This paper addresses the challenges of an early flood warning caused by complex convective systems (CSs), by using Low-Earth Orbit and Geostationary satellite data. We focus on a sequence of extreme events that took place in central Vietnam during October 2020, with a specific emphasis on the events leading up to the floods. In this critical phase, several hydrometeorological indicators could be identified thanks to Earth Observation satellites, which enable the characterization and monitoring of a CS, in terms of low-temperature clouds and heavy rainfall. Himawari-8 (H8) images, both individually and in time-series, allow identifying and tracking convective clouds. This is complemented by the observation of heavy/violent rainfall through GPM IMERG data, and the detection of strong winds using radiometers/scatterometers. Collectively, these datasets, along with the estimated intensity and duration of the event from each source, form a comprehensive dataset detailing the intricate behaviors of CSs. All of these factors are significant contributors to the magnitude of flooding and the short-term dynamics anticipated in the studied region.

Ship detection from satellite imagery using Deep Learning (DL) is an indispensable solution for maritime surveillance. However, applying DL models trained on one dataset to others having differences in spatial resolution and radiometric features requires many adjustments. To overcome this issue, this paper focused on the DL models trained on datasets that consist of different optical images and a combination of radar and optical data. When dealing with a limited number of training images, the performance of DL models via this approach was satisfactory. They could improve 5–20% of average precision, depending on the optical images tested. Likewise, DL models trained on the combined optical and radar dataset could be applied to both optical and radar images. Our experiments showed that the models trained on an optical dataset could be used for radar images, while those trained on a radar dataset offered very poor scores when applied to optical images.

Deep neural networks (DNNs) have demonstrated remarkable success across various domains, including Earth Observation applications. Despite their achievements, DNNs do not quantify the uncertainty of their predictions, which is particularly crucial for high-stakes applications such as flood mapping. We applied density-aware deep neural networks for uncertainty quantification in SAR-based flood mapping through a single forward pass. The aleatoric uncertainty is captured through softmax entropy, while epistemic uncertainty is quantified using density in the latent feature space. Our image segmentation results illustrate that the employed density-aware deep neural networks exhibit good performance in uncertainty quantification, surpassing Deep Ensembles for out-of-distribution (OOD) data detection.

We propose a novel deep learning-based method that adapts the domains of images acquired by different remote sensing sensors. It adapts a lower resolution image to the domain of an an higher resolution targeted sensor. This is effective in the case of change detection, where differences between sensors, such as spatial resolution and radiometry, can hinder the detection performance and where model hallucination artifacts are unwanted. The proposed technique divides the input image into patches and uses a diffusion-based model to generate translated patches in the style of the target sensor. The translated patches are stitched together to form the output image, which provides global generative consistency. Our approach can handle images with different resolutions and tonalities. We show its effectiveness on a Sentinel-II + Planet Dove data set and demonstrate its high generation quality and contribution to enhance change detection performance.

Climate change is increasing the frequency and severity of hydrological extremes in many parts of the world. In Europe as well as in Luxembourg, droughts have worsened in intensity and duration in recent years, causing significant loss to several sectors, such as agriculture and forestry. There is a pressing need for developing and setting up advanced drought monitoring and prediction systems. In this context, this research work aims to improve drought prediction accuracy by jointly assimilating, into a hydrological model, various EO-based datasets, namely soil moisture (SM) and total water storage (TWS) derived from S-1 and GRACE & GRACE-FO satellite missions respectively. The assimilation of satellite-observed water content enables an integrated assessment and modeling of water resources through the monitoring and modeling of SM and groundwater availability in Luxembourg and the Greater Region, between 2012 and 2022.

This paper proposes the observations of oil drift and the changes in oil shape and size based on the collocation of Sentinel-1 descending and ascending images with a time lag of 12 hours offshore Nigeria. The oil slicks are first detected from the descending and ascending images using the hierarchical split-based approach to identify oil objects and non-linear filters (mean and standard deviation) to determine oil contours. Then, the detected oil is collocated to estimate the distance and direction of its movements. Finally, surface wind and current data are used for analyzing the relationship between met-ocean conditions and the evolution of oil slicks.

Previous references indicated that surface wind gusts associated with deep convection can be observed and estimated from Sentinel-1 images. They also presented the relationship between surface wind patterns and deep convective clouds observed on Meteosat geostationary (GEO) images. To strengthen this relationship, this paper presents the comparison between surface wind speed retrieved from Sentinel-1 data, wind magnitude measured by the weather stations, and deep convective clouds observed on GOES-16 GEO images over the Gulf of Mexico. The results show that a mesoscale surface wind pattern (a squall line) observed on Sentinel-1 images corresponds to deep convective cloud locations. In particular, the peaks of wind intensity measured by the weather stations match the Sentinel-1 wind gusts and the deep convective clouds.

The accuracy of soil moisture estimated from Synthetic Aperture Radar backscatter data at high resolution is limited by speckle. Common practice to mitigate speckle is to multilook the data prior to retrieving soil moisture. While multilooking indeed reduces speckle, it also decreases the spatial resolution and removes possibly useful high resolution information from the data. We therefore hypothesised that using higher resolution backscatter data for soil moisture retrieval would lead to higher retrieval accuracies. A high-resolution field study combined with a synthetic experiment showed that calculating soil moisture prior to multilooking to the final target resolution (calculate-then-average, CtA) has substantial advantages over the average-then-calculate (AtC) approach. Currently, the AtC strategy is most often applied in soil moisture studies, mainly due to its computational advantage compared to the CtA approach. We show that by making use of a higher source resolution backscatter data than the target resolution, we could improve the soil moisture retrieval over an agricultural field.

Due to the complexity of urban environments, the synthetic aperture radar (SAR) based mapping of floodwater is impacted by different factors such as water depth, building orientation and the density of built-up areas. Several studies have proven that both SAR multitemporal intensity and interferometric SAR (InSAR) coherence data acquired in VV and VH polarizations support the urban flood mapping. We propose a deep learning (DL) based method using dual-polarization Sentinel-1 multitemporal intensity and coherence data combined with prior information to map floodwater in urbanized areas. The proposed method aims at mapping flooded areas in urbanized regions and bare soils/sparsely vegetated areas within the entire frame of a Sentinel-1 image. In this paper, our method is evaluated for the Houston (US) urban flood event in 2017 via a qualitative and quantitative comparison with two established DL models. The proposed method has the lowest number of false alarms in flooded urban areas, indicating that the prior information from the probabilistic urban mask is valuable.

The accuracy of high resolution soil moisture estimated from SAR backscatter data is limited due to speckle in the native resolution backscatter data. However, reducing this speckly by means of spatial aggregation also removes useful information from the data. We therefore hypothesised that using unfiltered backscatter data in a soil moisture inversion model can be valuable in high resolution soil moisture applications. A field study combined with a synthetic experiment showed that calculating soil moisture prior to spatial averaging to the final target resolution (CtA) has substantial advantages over the average-then-calculate (AtC) approach. Currently, the AtC strategy is most often applied in soil moisture studies, mainly due to its computational advantage compared to the CtA approach. However, especially at high resolutions, using a slightly higher source resolution backscatter data than the target soil moisture resolution, can already improve accuracy of the soil moisture estimates.

Previous studies have shown that the decrease of temporal interferometric synthetic aperture radar (InSAR) coherence could be exploited to detect the appearance of floodwater in urban areas. However, as of today, approaches based on this principle only make use of single co-polarization images for identifying the presence of floodwater in the double-bounce feature. In this study, we take advantage of both co- and cross-polarization images to detect significant decreases of the multitemporal InSAR coherence in order to enhance the mapping of floodwater in urban areas. We consider that not only double-bounce scattering, but also multiple-bounce may occur in urban areas depending on how the building facades are oriented with respect to the synthetic aperture radar (SAR) sensor’s line of sight. The Sentinel-1 (S-1) mission is particularly well suited for applying and testing this kind of approach due to the systematic availability of dual-polarization data. Using as a test case, the widespread flooding in the city of Houston, USA, caused by Hurricane Harvey in 2017, we demonstrate that the proposed methodology leads to an increase of the accuracy of the urban flood maps from 75.2% when only using the VV polarization, to 82.9% when using the dual polarization information.

Due to the complexity of backscattering mechanisms in built-up areas, the synthetic aperture radar (SAR)-based mapping of floodwater in urban areas remains challenging. Open areas affected by flooding have low backscatter due to the specular reflection of calm water surfaces. Floodwater within built-up areas leads to double-bounce effects, the complexity of which depends on the configuration of floodwater concerning the facades of the surrounding buildings. Hence, it has been shown that the analysis of interferometric SAR coherence reduces the underdetection of floods in urbanized areas. Moreover, the high potential of deep convolutional neural networks for advancing SAR-based flood mapping is widely acknowledged. Therefore, we introduce an urban-aware U-Net model using dual-polarization Sentinel-1 multitemporal intensity and coherence data to map the extent of flooding in urban environments. It usesa prioriinformation (i.e., an SAR-derived probabilistic urban mask) in the proposed urban-aware module, consisting of channel-wise attention and urban-aware normalization submodules to calibrate features and improve the final predictions. In this study, Sentinel-1 single-look complex data acquired over four study sites from three continents have been considered. The qualitative evaluation and quantitative analysis have been carried out using six urban flood cases. A comparison with previous methods reveals a significant enhancement in the accuracy of urban flood mapping: the F1 score of flooded urban increased from 0.3 to 0.6 with few false alarms in urban area using our method. Experimental results indicate that the proposed model trained with limited datasets has strong potential for near-real-time urban flood mapping.

Operational services in the fields of flood monitoring and prevention are benefitting from the large scale and systematic availability of synthetic aperture radar (SAR) data. The main advantages of SAR data are that they provide synoptic views over wide areas, day and night and all-weather condition acquisitions and a reliable data acquisition schedule. Satellite SAR data availability has increased over the past few years due to renewed efforts of several space agencies to put in place new satellite constellations. The latter enable the reduction of the satellite time access to areas of interest and provide enriched information with increased spatial resolution as well as variable polarizations and frequencies. The current situation tells us that there are regions in the world and land cover classes where SAR-derived flood maps are very reliable and accurate, but others where uncertainty is still very high, or where SAR is even unable to provide flood extent information. Therefore, the aim of this paper is to provide an overall picture of SAR-based floodwater mapping algorithms and their suitability for operational applications.

Space-borne SAR instruments can provide backscatter on a high spatial resolution, and with the introduction of the Sentinel-1 satellites, these can co-exist with relatively high temporal resolutions. Here, we use a combination of active microwave Sentinel-1 and optical Sentinel-2 data in the MULESME algorithm to estimate soil moisture on a field in Southeastern Luxembourg. Satellite data were compared to data gathered in the field and semi-continuous measurements from a nearby permanent station. Our results indicate that the accuracy of MULESME soil moisture estimates increases with a decrease in spatial resolution, but that this increase stagnates rather soon after the first few spatial aggregations, thus confirming the value of high resolution data. Future endeavours will focus on the analysis of soil moisture variation in time, compared to soil moisture measurements from a nearby permanent station.

This study addresses the effects of SAR signatures of moving vessels extracted from Sentinel-1 imagery, that suffer from a loss of focus due to the azimuthal velocity, e.g. target defocusing. The effects generated by SAR moving targets in the azimuthal direction result in residual Doppler chirps that can be estimated and characterized by processing Sentinel-1 Single Look Complex (SLC) images. We propose to employ the fractional Fourier transform (FrFT) in order to compensate the moving target defocusing effects in the SLC domain. In addition, different Sentinel-1 polarimetric representations of a target within the FrFT domain are also addressed and fused. The experimental results are based on Sentinel-1 Stripmap images and are cross compared with AIS data.

The new, systematic Global Flood Monitoring (GFM) product of the Copernicus Emergency Management Service will provide a continuous monitoring of floods worldwide by immediately processing and analysing all incoming S-1 Interferometric Wide Swath data and making use of the data cube approach enabling a high product timeliness and the implementation of flood mapping algorithms that require data-driven model training. It integrates three independently developed flood mapping algorithms to improve the robustness and accuracy of the flood and water extent maps and to build a high degree of redundancy into the service.

With the ongoing proliferation of satellite data, in particular open-access satellite imagery, from both optical and synthetic aperture radar (SAR) sensors, the number of downstream applications is rapidly growing. Developers of Earth Observation (EO)-based products and services, as well as expert and non-expert users of such tools, thus need access to a cloud computing infrastructure offering interoperable analysis functionality. Here, we present the versatility of such a cloud-based infrastructure called WASDI. WASDI, a web-advanced space development interface, is an online EO analytics platform where EO experts can develop and deploy applications (apps) and users can use them to processes satellite images on demand to generate value-added content.

Due to the similarity of the radar backscatter in flooded and unflooded conditions over particular areas, it is not possible to carry out a comprehensive SAR-based flood mapping at large scale. In this paper, an additional information layer derived from Sentinel-l time series data, called Exclusion map (EX-map), is introduced. Its aim is to enhance and complement the results of automatic change detection-based flood mapping methods. The EX-map aims at delineating areas where observed variations of SAR backscatter do not allow detecting the appearance of floodwater. The EX-map is mainly composed of the following land cover classes: topographic shadow/layover, double bounce and smooth tarmac in urban areas, arid areas, dense vegetation and permanent water bodies. The method is evaluated over six study sites across the globe and tested for different flood events. The EX-map not only increases the classification accuracy of change detection-based flood maps derived from Sentinel-l data from 95.92% to 97.02%, but also enables a better interpretation of any SAR-based floodwater map.

This letter addresses the use of the Sentinel-1 time series with the aim of proposing an automatic and unsupervised coastline detection method that averages the dynamical variations of coastal areas over a limited period of time, e.g., one year. First, we propose applying a temporal averaging filter that allows the temporal variations in coastal areas, e.g., due to tides or vegetation, to be encapsulated, and, at the same time, the speckle to be reduced, without decreasing the spatial resolution of the synthetic aperture radar (SAR) time series. Then, based on the distinctive backscattering values of the sea and land pixels, we will employ an iterative hierarchical tiling method in order to accurately characterize the two classes using bimodal distribution. The distribution is then segmented by a thresholding and region-growing procedure to separate the sea and land classes. A large-scale quantitative comparison between the SAR-derived and open street map (OSM) coastlines allows for a numerical evaluation of the results, i.e., an overall agreement ranging from 80% to 90%. In addition, Sentinel-2 images are used to evaluate the estimated SAR coastline qualitatively. Furthermore, the benefits of having an accurate SAR coastline are shown in the case of two well-known Earth observation-monitoring applications, ship detection, and floodwater mapping.

We introduce a new SAR-based flood extent mapping algorithm enabling systematic and automatic monitoring of water bodies at large scale in near real time. The algorithm is both efficient and robust, especially in areas where flood events are not sporadic but long lasting (e.g. monsoon-related floods). It is based on the regular processing of subsequently acquired pairs of Sentinel-1 images. The algorithm has been developed in the framework of the ESA-funded e-DRIFT project, with the aim to respond to the needs of the disaster risk financing sector in Southeast Asia. The approach has been validated after intensive testing over different areas of interest in South East Asia i.e. Myanmar and Laos, where risks associated with flooding are currently not well-known. Moreover, the algorithm is implemented on a virtual platform that efficiently handles large collections of Sentinel-1 data from all the orbits and dates available over areas of interest affected by floods. The output of this near real-time system are reliable and useful input data for the parametric modelling carried out by the (re-)insurance companies, allowing them to better anticipate risk of natural disasters.

In this paper, we present a fully automatic algorithm capable of mapping floodwater in urban areas using 20 m Sentinel-1 SAR data. It is composed of a two-steps approach that first uses the SAR data to identify buildings and then takes advantage of the Interferometric SAR coherence feature from both co- and cross-polarizations to detect the presence of floodwater in urbanized areas. The preliminary detection of buildings is a pre-requisite for classifying them as flooded based on the InSAR coherence temporal decrease when water is present in urban areas, given that in general buildings show a strong temporal coherence. In addition, the short temporal and perpendicular baselines of the intereferomeric Sentinel-1 image acquisitions is an advantage for this kind of approach. The algorithm is applied to Sentinel-1 images acquired during the major flood event that hit Jakarta (Indonesia) in January 2020.

This research addresses the use of Sentinel-1 time series with the aim of detecting spatio-temporal changes in the coastal environment. To this end an automatic and unsupervised coastline detection method is proposed. First, we apply a temporal averaging filter that allows encapsulating the temporal variations in coastal areas, e.g. due to tides or vegetation, and at the same time it allows reducing the speckle, without decreasing the spatial resolution of the Synthetic Aperture Radar (SAR) images. Then, based on the distinctive backscattering values of the sea and land classes we employ an iterative hierarchical tiling method in order to accurately characterize the two classes by a bimodal distribution. The latter is then segmented by a thresholding and region-growing procedure to separate the sea and land classes. The proposed method is applied to two different SAR time-series, each one acquired throughout one year. The extracted yearly coastlines are then analyzed in order to identify spatio-temporal changes. Experimental results showcase coastal area changes between occuring 2018 and 2019 and that were caused by the hurricane Michael hitting Northwest Florida in October 2018.

The availability of many platforms carrying on board SAR payloads working at different frequency bands is paving the way to the development of new or improved products for different applications. Constellations of satellites also enable acquisitions close in time, thus not only improving the temporal resolution but also enabling almost coincident multifrequency observations of the same target. This work resumes previous investigations demonstrating the role of multifrequency data for the generation of thematic maps and the observation of flooded vegetated areas. The different signatures of the targets made it possible to improve the classification accuracy when SAR data, at different frequencies, were added to optical data. Although optical data still keep the best discrimination capability, multifrequency SAR data play a role to improve thematic accuracy. The higher penetration through the vegetation of L-band signal, with respect to C- and X-band, and the polarimetric mode made possible to clearly detect the enhancement of the double bounce scattering due to the presence of standing water under rice fields, although some evidence of that mechanism can be also detected even at X-band by change detection approaches.

This research addresses the use of machine learning for detecting building footprints from dual-polarization multitemporal SAR data. We assume that in SAR images, built-up areas exhibit very high backscattering values, making them brighter than other land cover classes in both the co- and cross-polarization channels. In addition the building class is considered to be stable in time. In this context, we propose to employ a convolutional neural network that integrates a set of SAR images of different dates and polarizations for delineating the building footprint. The algorithm was tested on Sentinel-1 datasets from eight different test sites located in Iraq. Four of the test sites along with their corresponding annotations (i.e. ground truth) constitute the training dataset, while the other four test sites are used for evaluation. All the datasets were provided in the framework of a humaritan AI4EO UNOSAT challenge launched in 2019 by Phi-Unet. The results evaluated by the F1 score with values of about 0.75 indicate that the proposed method is able to accurately detect the building footprint.

In this work we present an automatic algorithm for providing probabilistic flood maps, not only on bare soils, but also within urban areas. The probabilistic flood mapping procedure is based on synthetic aperture radar (SAR) data and the Bayesian inference. Both intensity data and Interferometric SAR (InSAR) coherence feature are used. The approach improves the information content of a binary SAR-based floodwater map, which does not give any indication on the uncertainty in the pixel state.The proposed methodology is tested for the flood event that heavily affected the city of Houston (Texas) during the 2017 hurricane season. Data provided by the Sentinel-1 mission are used, with a geometric resolution of 20m. The algorithm takes fully advantage of the Sentinel-1 mission's repeat cycle of six days and narrow orbital tube to fully exploit the potentialities of InSAR coherence feature to detect floodwater in complex environments. The application of the proposed method to the Houston case study showed promising results.

This study addresses the use of Sentinel-1 data for innovative improvements of automatic classic ship detection detection chains. Firstly, we propose to extract the complex coherence from the two polarization channels and to perform the vessel detection the vessels in this domain. A comparative assessment between the use of the complex coherence and the intensity images together with AIS validation demonstrates that the fusion of the different results allows to reduce the number of false alarms while maintaining an optimal detection rate. Secondly, we propose to make use of Sentinel-1 time series in order to delineate the coastline which is an essential parameter for ship detection chains. Experimental results are conducted on Sentinel-1 images acquired in different areas of interest for maritime surveillance such as the Gulf of Califoria (Mexico) or the English Channel.

The utility of synthetic aperture radar (SAR) data to produce flood delineation maps is well established. However, flood mapping still represents a challenge in urban settlements, because the radar signatures of flooded urban pixels are generally ambiguous. As a matter of fact, flood mapping algorithms generally do not consider urban areas, thus producing a lot of missed detection errors. Recent studies demonstrated that SAR Interferometry (InSAR) represents a suitable tool to at least mitigate this problem. Following these studies, here we analyze time series of complex coherence data in stable scatterers, i.e., pixels exhibiting high backscatter combined with high temporal stability. Our idea is based on the fact the water surfaces show no coherence in a repeat-pass interferogram, so that a decrease of coherence may occur even for stable scatterers if floodwater is present in a resolution cell. The analysis was performed considering the floods that hit the city of Alicante (Spain) in March 2017. This event was observed by Sentinel-1 in Interferometric Wide Swath mode.

The Zambezi river basin, one of the world's largest flood-plains, located in south-eastern Africa, is recurrently subject to floods [1] . It has been the subject of several studies exploiting multi-temporal SAR data to monitor its hydrological cycle and periodic inundations, e.g. [2] .

The main objective of this study is to introduce and evaluate a SAR-based flood mapping algorithm enabling the automatic generation of a large-scale flood record from the ENVISAT ASAR data archive. The flood mapping algorithm is based on a change detection approach and requires an automatic selection of optimal reference images. The flood mapping algorithm is applied to selected pairs of images to sequentially generate a record of flood extent maps. False alarms caused by water-like areas are reduced using auxiliary data sources such as the Height Above Nearest Drainage (HAND) index derived from topography data. The proposed method is applied to several ENVISAT WS ASAR datasets acquired over the UK and results are validated with a flood extent map derived from aerial photography. Results presented in this paper demonstrate the effectiveness of the methodology.

This study aims to map built-up areas using SAR data provided by the Sentinel-1 mission. The proposed algorithm exploits several features offered by the satellite mission such as: high revisit time, dual-polarization data and Interferometric SAR coherence. The algorithm is based on an adaptive parametric thresholding methodology that identifies pixels with high backscattering values in both VV and VH channels corresponding to built-up areas. The Interferometric SAR coherence allows discriminating false alarms caused by other land cover classes characterized by high backscattering values which are not coherent in time (e.g. certain types of vegetated areas). Both the intensity in VV and VH, as well as coherence features are obtained by averaging multi-temporal SAR series. This allows reducing the speckle without any loss in spatial resolution. The algorithm has been tested on Sentinel-1 Interferometric Wide Swath data from five different test sites located in semiarid and arid regions in the Mediterranean region and Northern Africa.

As of today, SAR imagery represents the most commonly used data source for remote sensing-based flood mapping. The data are characterized by a good sensitivity to water and are available day and night, regardless of cloud cover. Many studies have demonstrated that SAR systems are suitable tools for flood mapping on bare soils and scarcely vegetated areas. In spite of the progress in the development of Near Real Time SAR based flood mapping algorithms, the detection of inundation in urban areas still represents a critical issue. Here we propose a methodology for identifying floods that heavily affected the city of Houston (Texas) during the 2017 hurricane season. Our approach takes advantage of the Interferometric SAR coherence feature to detect the presence of floodwater in urbanized areas. In particular, data provided by the Sentinel-1 mission in both, Strip Map and Interferometric Wide Swath modes, have been used, with a geometric resolution of 5m and 20m, respectively. The algorithm takes fully advantage of the Sentinel-1 mission's repeat cycle of six days, thereby providing an unprecedented possibility to develop an automatic, high frequency flood mapping application that is suitable for complex environments. The test of the algorithm for the Houston case study showed promising results for mapping flood in urban areas.

This study addresses the use of dual-polarimetric descriptors for ship detection and characterization from Synthetic Aperture Radar (SAR) data. Ship detection is usually performed independently on each polarization channel and the results are merged subsequently. We propose to extract polarimetric descriptors from the two polarization channels and to perform the vessel detection the vessels in this domain. Several polarimetric descriptors, such as those derived from the the Eigenvector-Eigenvalue decomposition, are employed for this purpose. A comparative assessment between the use of intensity images and polarimetric descriptors for the detection and identification of ships is then realized. The proposed methodology is tested on Sentinel-1 data acquired over the English channel. Automatic Identification System (AIS) data flows are considered as ground truth.

Monitoring of wetland inundation dynamics is important for flood management and the characterisation of hydrological connectivity. SAR-based inundation extent monitoring in wetlands is often challenging due to different factors, such as waves, vegetation cover and wet snow. The presented study targets the mapping of inundation dynamics in the Prairie Pothole Region (PPR) of North Dakota, USA. A 3-year water extent time series was derived from Sentinel-1 SAR data by first delineating permanent water bodies using a clustering approach. In a second step, water body dynamics were mapped using region growing and automatic thresholding. Results suggest that there is considerable potential for mapping surface water dynamics in late spring, summer and autumn, whereas confusion with wet snow may take place in early spring.

The assessment of satellite image classifications is usually carried out using a test sample assumed as the ground truth, from which a confusion matrix is derived. There are cases where the reference data, even those coming from a ground survey, are affected by errors and do not represent a reliable truth. In the field of geophysical parameter retrieval, the triple collocation (TC) technique is applied for validating remotely sensed products when the source of test data (e.g., ground data) does not represent a reliable reference. TC is able to retrieve the error variances of three systems observing the same target parameter, assuming that their errors are independent. In this paper, we exploit the same idea to test the classification accuracy in cases where the ground truth is not available. We extend the TC approach to the classification problem for a general number of classes, but we solve it numerically for a two-class problem (i.e., collapsed and noncollapsed buildings). The specific case refers to the detection of L'Aquila 2009 earthquake damage from very high-resolution optical data. The image classification, performed by exploiting an object-based analysis, is compared with those from two different ground surveys carried out after the earthquake by different teams and with different purposes. This paper demonstrates the power of the TC approach for assessing the classification accuracy with no reliable ground truth available, and provides an insight into the problem of assessing damage, from satellite and on ground, in a very critical and unsafe situation, like the one occurring after an earthquake. Moreover, it was found that the remotely sensed product can have an order of accuracy comparable to that of at least one of the ground surveys.

The objective of this research was to develop a method for water level retrieval in natural and artificial lakes. It was thought to be applied for monitoring purposes and flood control applications, especially in data-scarce environments. The method is based on a combined GIS, remote sensing and statistical modeling approach. It was tested on both optical (Landsat 8) and SAR (Cosmo-SkyMed®) data. The topographic information, required by the method, were obtained from freely available digital elevation models (SRTM and ASTER) to compare their performances. The Place Moulin Lake, an Alpine reservoir, was selected as study area since it represents a very challenging case study for developing the proposed methodology. The results showed that: i) the method provided reasonably accurate results when the degree of filling of the reservoir was high. ii) The accuracy of the results strongly relied on the accuracy of the topographic information. iii) The combination of Cosmo-SkyMed® and SRTM data provided more reliable results. Further analyses are required to evaluate the method in different environmental conditions.

As part of the Copernicus Programme, Sentinel 1 (S1) synthetic aperture radar (SAR) mission represents a unique monitoring tool whose potentialities for hydrological risk mitigation need to be evaluated. To this aim, S1-A derived soil moisture maps with high spatial resolution (100 m) and moderate temporal resolution (12 days) were assimilated within a time-continuous, spatially-distributed, physically-based hydrological model (Continuum) with the specific objective to evaluate the impact on discharge predictions and (flash) flood modelling. A Nudging assimilation scheme was chosen for the DA experiment due to its computational efficiency, particularly useful for operational applications. Results were evaluated in the Orba River catchment (Italy) in the time period October 2014 — November 2016, corresponding to the first two years of activity of the S1-A mission.

The main objective of this study is to generate inundation maps of past flood events based on an archive of Synthetic Aperture Radar (SAR) data. Within a hierarchical image splitting framework, the flood mapping algorithm uses a histogram thresholding operation and a region growing process to delineate the flood extent. This algorithm is applied to an archive of SAR images in order to generate a flood frequency map. We define the flood frequency of a specific area as the ratio between the number of images where the area was detected as flooded and the total number of images within the employed data collection. SAR water-like ambiguities (e.g. urban areas, crops or shadow regions) are filtered out using auxiliary data sources such as the Height Above Nearest Drainage (HAND) index or land cover maps. The proposed methodology is applied to an ENVISAT ASAR image archive over the UK area. Results presented in this article demonstrate the effectiveness of this methodology.

In this paper we investigate the double bounce enhancement due to standing water in flooded agricultural fields to assess the capability of an X-band radar to recognize the presence of floodwater under vegetation. The investigation was carried out by analyzing a polarimetric and multifrequency SAR dataset (COSMO-SkyMed, Alos-2, Radarsat-2) collected over the Vercelli district in North Italy, characterized by a widespread and intense cultivation of rice crop, were the fields were routinely artificially flooded and dried according to the agricultural practice. The investigation demonstrated that in July, when rice is well developed, high backscatter in X-band was observed in fields were the L-band polarimetric data recognized the double bounce return. The presence of a double bounce scattering enhancement at X-band was then established. At C-band the dihedral type of return was not clearly recognized because of the smaller incidence angle of Radarsat-2 acquisitions.

The utility of synthetic aperture radar (SAR) data to produce flood delineation maps is well established. However, for what concerns urban settlements, flood mapping still represents a challenge, because the radar signatures of flooded urban pixels are generally ambiguous. As a matter of fact, flood mapping algorithms generally do not consider urban areas, thus producing a lot of missed detection errors. To cope with this problem, this study proposes a new method that basically analyzes the complex coherence of urban pixels characterized by high backscatter combined with high temporal stability (stable scatterers). Contextual information is also used to reduce the noise of the maps. The rationale is that, since water surfaces show no coherence in a repeat-pass interferogram, a decrease of coherence may occur even for (some) stable scatterers if floodwater is present in a resolution cell. To develop the algorithm, the inundation that hit Houston (Texas, USA) in April 2016 was considered. This event was observed by Sentinel-1 in Interferometric Wide Swath mode. Results showed that looking at the coherence of stable scatterers could represent a reliable road to at least mitigate the problem of missed detection of flooded urban settlements.

Parametric thresholding algorithms applied to synthetic aperture radar (SAR) imagery typically require the estimation of two distribution functions, i.e., one representing the target class and one its background. They are eventually used for selecting the threshold that allows binarizing the image in an optimal way. In this context, one of the main difficulties in parameterizing these functions originates from the fact that the target class often represents only a small fraction of the image. Under such circumstances, the histogram of the image values is often not obviously bimodal and it becomes difficult, if not impossible, to accurately parameterize distribution functions. Here we introduce a hierarchical split-based approach that searches for tiles of variable size allowing the parameterization of the distributions of two classes. The method is integrated into a flood-mapping algorithm in order to evaluate its capacity for parameterizing distribution functions attributed to floodwater and changes caused by floods. We analyzed a data set acquired during a flood event along the Severn River (U.K.) in 2007. It is composed of moderate (ENVISAT-WS) and high (TerraSAR-X)-resolution SAR images. The obtained classification accuracies as well as the similarity of performance levels to a benchmark obtained with an established method based on the manual selection of tiles indicate the validity of the new method.

First results of the assimilation of high-resolution Sentinel-1A based soil moisture products in a distributed, physically based, hydrological model are presented. A comprehensive evaluation of the assimilation's impact on discharge predictions is provided. Results are further compared to those obtained when assimilating the lower-resolution ASCAT-based soil moisture product. The exercise was carried out within the MIDA project framework (funded by the Italian Space Agency) aiming at producing root zone soil moisture maps useful for flood risk management applications. The experimental site is the Orba River Catchment (Italy). The period of investigation is October 2014-February 2015. Using a relatively simple data assimilation technique (Nudging) the results of our case study show that overall the assimilation of currently available Sentinel-1 data only marginally improves discharge simulations. However, the impact becomes more significant when specifically considering predictions of high flow. Further improvements are expected when both Sentinel-1A and B data will be available.

Classification algorithms for Synthetic Aperture Radar (SAR) imagery that are based on statistical modelling typically require the parameterization of distribution functions of backscatter values of different land cover classes to classify a scene. To parameterize accurately the distribution functions of the individual classes a sufficient number of pixels is needed and this criterion is not always satisfied, especially for classes occupying only a small fraction of the scene Here we propose an automatic algorithm that aims to map buildings based on the SAR intensity backscattering feature. It makes use of a hierarchical split-based approach that does not fix the size of the tiles a priori but, rather, searches for tiles of variable size where the distribution functions attributed to classes of interest can be parameterized in a robust way. The algorithm has been developed in the framework of the Urban Round-Robin exercise, supported by the European Space Agency (ESA) through the ESA Land Cover Climate Change Initiative (CCI), and tested on Sentinel-1 data from five different test sites located in semiarid and arid regions in the Mediterranean region and Northern Africa.

By providing high quality flood maps a spaceborne SAR can be an effective source of information. These maps support civil protection authorities for disaster risk reduction. Here we propose a methodology for identifying floods that occur on different types of land cover, such as urban areas, bare and poorly vegetated soil and vegetated areas. Our approach takes advantage of polarimetric SAR data and InSAR coherence to better characterize the landscape. Indeed, the Sentinel-1 repeat cycle of six days, and its systematic acquisition of dual-pol SAR data, provides an unprecedented chance to develop automatic, high frequency flood mapping algorithms for complex environments. The algorithm has been tested on two different Sentinel-1 datasets, acquired, respectively over Greece and Italy, showing promising results.

Observations of the temporal and spatial variations of water depth in rivers and floodplains are very important in operational hydrology. However, our capacity to monitor water depth at large scale is still very limited. As a result, the need to measure water storage changes in all wetlands, lakes, and reservoirs has motivated the radar interferometry-based Surface Water Ocean Topography Mission (SWOT) scheduled for launch in 2020.

In this paper, we investigate the role of polarimetric features to improve flood mapping in agricultural areas. Considering that the double bounce enhancement due to standing water can increase the backscatter from flooded agricultural fields, polarimetry can potentially detect this mechanism and mitigate the misdetection of algorithms based on the identification of dark areas in the image. The investigation was carried out by analysing a polarimetric and multifrequency SAR dataset (COSMO-SkyMed, Alos-2, Radarsat-2) collected over the Vercelli district in North Italy, characterized by a widespread and intense cultivation of rice crop, were the fields were routinely artificially flooded and dried according to the agricultural practice. The investigation demonstrated that the polarimetric data are able to recognize the double bounce return in areas with high backscattering. They overcome the need of a pre-flood image, otherwise required to identify a sudden increase of backscatter to be ascribed to the standing water.

Probabilistic flood mapping offers flood managers, decision makers, insurance agencies, and humanitarian relief organizations a useful characterization of uncertainty in flood mapping delineation. Probabilistic flood maps are also of high interest for data assimilation into numerical models. The direct assimilation of probabilistic flood maps into hydrodynamic models would be beneficial because it would eliminate the intermediate step of having to extract water levels first. This paper introduces a probabilistic flood mapping procedure based on synthetic aperture radar (SAR) data. Given a SAR image of backscatter values, we construct a total histogram of backscatter values and decompose this histogram into probability distribution functions of backscatter values associated with flooded (open water) and non-flooded pixels, respectively. These distributions are then used to estimate, for each pixel, its probability of being flooded. The new approach improves on binary SAR-based flood mapping procedures, which do not inform on the uncertainty in the pixel state. The proposed approach is tested using four SAR images from two floodplains, i.e., the Severn River (U.K.) and the Red River (U.S.). In all four test cases, reliability diagrams, with error values ranging from 0.04 to 0.23, indicate a good agreement between the SAR-derived probabilistic flood map and an independently available validation map, which is obtained from aerial photography.

The use of synthetic aperture radar (SAR) data is presently well established in operational services for flood management. Nevertheless, detecting inundated vegetation and urban areas still represents a critical issue, because the radar signatures of these targets are often ambiguous. This paper analyzes the role of the interferometric coherence in complementing intensity SAR data for mapping floods in agricultural and urban environments. The advantages of the joint use of intensity and coherence are first discussed in a theoretical way and then verified on a case study, namely, the flood that hit the Emilia-Romagna region (Northern Italy) in January 2014. The short revisit time of the COSMO-SkyMed images, as well as a dedicated acquisition plan tailored to the requirements of the Italian Civil Protection Department, has allowed us to build a data set of radar interferometric observations of the event. Results show that the analysis of the multitemporal trend of the coherence is useful for the interpretation of SAR data since it enables a considerable reduction of classification errors that could be committed considering intensity data only. Interferometric data have permitted us to distinguish zones where water receded from areas where it persisted for a longer time and, in one case, to measure changes of water level.

Nowadays very high resolution (VHR) Synthetic Aperture Radar (SAR) systems can provide near real time earthquake damage maps with an high degree of details to stakeholders in charge of managing the emergency phase. However, the increased resolution introduces new challenges to interpret and detect changes in urban areas caused by seismic events. In metric resolution SAR sensors a building appears as a complex of image structures associated to different scattering mechanisms, preventing the use of pixel-based algorithms. In this paper we propose an object oriented approach, focusing the attention on the double-bounce return from buildings, trying to detect damages looking at changes of these particular image patterns. The identification of double-bounce regions is performed using open and close morphological filters and assuming linear structuring elements with different orientation and length. The change detection analysis based on a pre- and a post-event image is carried out using four change detection indicators, such as: intensity ratio, interferometric coherence, intensity correlation and Kullback-Leibler divergence. All change features are extracted using all pixels within each identified object, i.e., double-bounce regions. The test case is the earthquake that hit L'Aquila city (Italy) on April 6, 2009, while the dataset is composed of two X-band COSMO-SkyMed SAR images acquired before and after the event. A macro-seismic survey map was available to evaluate the obtained results.

Volcanic eruptions affect at different levels the population and economy of interested areas. Moreover, volcanic ash detection represents a key issue for aviation safety due to the harming effects on aircraft. For these reasons, an accurate and fast analysis of the data is needed to monitor the phenomena's evolution and to manage the risk mitigation phase. In this scenario, the introduction of an inversion approach based on Neural Networks (NNs) has significant interest to reduce the need of human interpretation of the ash detection maps as those generated by the application of brightness temperature difference approach. In this work we show that NNs algorithms are suitable for an accurate mapping of ash cloud on Moderate Resolution Imaging Spectroradiometer (MODIS) images in a very cloudy scenario as the ones of 2010 Eyjafjallajökull and 2011 Grimsvötn eruptions.

The latest generation of synthetic aperture radar (SAR) systems allows providing emergency managers with near real time flood maps characterized by a very high spatial resolution. However, mapping inundations in vegetated and urban areas still represents a critical issue, because the radar signatures of these targets are often ambiguous. This paper proposes a possible strategy to cope with flood mapping using SAR data in vegetated and urban areas. In particular, the use of the interferometric coherence is proposed to complement the information brought by intensity. The SAR images used for verifying the potentiality of the joint use of intensity and coherence data are the COSMO-SkyMed observations of the flood that hit the Emilia region (Northern Italy) in January 2014.

This work presents the main outcomes of a land use change detection analysis by means of a classification algorithm based on very high resolution (VHR) optical images. The satellite data we used were captured by the sensor on board of IKONOS and GeoEye-1, on January 2002 and June 2012, respectively. Land use map at each day has been obtained merging the results of a supervised multispectral per-pixel classification and an unsupervised hierarchical segmentation aiming at classifying the objects in the VHR images selecting the spatial scales which maximize the final classification accuracy. The change detection land use analysis has been performed in post classification phase, comparing the resulting land use maps. The implemented classification architecture, called Scale Object Selection (SOS), allowed to obtain an overall classification accuracy higher than 97%, and a K-coefficient of about 0.95.

A new method for flood hazard mapping that integrates global flood inundation modeling and microwave remote sensing is presented. It combines the time and space continuity of a global inundation model with the limited revisit time but high spatial resolution of satellite observations. The availability of model simulations over a long time period allows a robust estimate of non-exceedance probabilities that can be attributed to the corresponding satellite observations. The resulting flood hazard map will have a spatial resolution equal to that of the used satellite images, generally higher than that of the global inundation model. This can theoretically be done for any point in the world, allowing the estimation of flood hazard at a global scale, provided that a sufficient number of remote sensing images are available. The method is tested on the Severn River (UK), with a high number of flood events observed by ENVISAT ASAR. The global ECMWF flood inundation model is considered for this study.

Predictions of hydrological models are highly uncertain due to both the nature of the modelled system and the meteorological forcings. Soil moisture information derived from satellite data can help to reduce this uncertainty. Indeed, data assimilation techniques offer the possibility to dynamically correct the model evolution in order to improve the model output. However, several questions concerning the use of these techniques are still without answer. The aim of this work is trying to better understand what conditions allow a successful assimilation of satellite soil moisture products. To this end, three different products, along with several options in terms of filter design, were tested and their impact on data assimilation performances was evaluated.

The latest generation of synthetic aperture radar (SAR) systems allows providing emergency managers with near real time flood maps characterized by a very high spatial resolution. Near real time flood detection algorithms generally search for regions of low backscatter, thus assuming that floodwater appears dark in a SAR image. It is well known that this assumption is not always valid. For instance, in urban areas, the double bounce backscattering involving ground and vertical walls produce high radar return that can be further increased by the presence of the highly reflective floodwater. In addition, even mapping bare or scarcely vegetated inundated terrains, or crops totally submerged by water can turn out to be a difficult task. In fact, in the presence of significant wind that roughens the water surface, floodwater can appear bright in SAR images. This paper proposes possible strategies to cope with flood mapping using SAR data in urban areas and in the presence of significant wind. In particular, the use of the interferometric coherence for floodwater detection in urban areas and the use of an electromagnetic model able to simulate the radar return from shallow water as function of the wind field are proposed.

An automatic method to distinguish water surfaces (either flooded or permanent water bodies) from artifacts caused by heavy precipitation and wet snow is designed to improve flood detection accuracy in X-band synthetic aperture radar (SAR) images. The algorithm implementing the proposed method, mainly based on image segmentation techniques and on the fuzzy logic, consists of two principal steps: 1) detection of regions (or segments) of low-radar backscatter that appear dark in a SAR image, and 2) classification of each detected segment. Ancillary data, such as a local incidence angle map, a land cover map, and an optical image (helpful to detect wet snow), are also used. Through the fuzzy logic, the algorithm integrates different rules for the detection of dark areas, as well as for their classification based on radiometric, geometrical and shape features extracted from the segmented SAR image and on the ancillary data. The algorithm is tested on the COSMO-SkyMed imagery of the severe weather event that hit Northwest Italy on November 2011. A comparison with measured data, provided by the weather radars belonging to the Italian radar national network, and with the ground precipitation, forecasted by a numerical weather prediction model routinely used within the framework of the EUMETSAT Hydrology Satellite Application Facility project, indicates that the algorithm produces reliable classification maps, being able to distinguish the rainfall signature on X-band SAR images from that of flooded areas.

We have investigated the recent deformation affecting the Messina Strait area in South Italy, across the Calabria and the Sicilia Regions. Messina Strait is an extensional tectonic feature that separates peninsular Italy from Sicily, and it is also one of the most seismically active and deforming area of the Mediterranean. It is located between the Neogene Tyrrhenian backarc basin, developed in the hanging wall of the Apennines, and the Mesozoic-Paleogene Ionian basin, which is the foreland and the area of propagation of the Neogene to present Apennines accretionary prism. We have exploited Synthetic Aperture Radar (SAR) data along the time period between 2002 and 2010 by applying the advanced Differential Interferometric Synthetic Aperture Radar (DInSAR) technique referred to as Small BAseline Subset (SBAS) algorithm. GPS data relative to permanent stations within the study area have been also used.

We studied the disastrous effects of the tsunami triggered by the Mw 9.0 earthquake that occurred on March 11, 2011, offshore the Honshu island (Japan). The tsunami caused a huge amount of casualties and severe damage along most of the eastern coastline of the island. The data set used is composed of images from ASTER, visible and thermal, and ENVISAT SAR sensors. The processing and the analysis of data from different sources were performed in order to obtain the tsunami inundation map of the Sendai coastal area, to analyze inland factors driving the tsunami inundation, and to detect the liquefaction effects in the Chiba bay area as well. The obtained inundation line, with a maximum value of about 6 km, has been jointly analyzed with digital elevation model providing the run-up values, which are generally below 21 m in the ca. 60-km-long study area of Sendai. Moreover, from SAR coherence and intensity correlation, a wide area of subsidence is mapped at Chiba bay, which is reasonably related to strong ground shaking and pervasive liquefaction.

On March 11, 2011 a Mw 9.0 earthquake hit Honsu island in Japan. The so called “Tohoku-Oki” giant earthquake occurred near the northeast coast resulting from thrust faulting on or near the subduction zone plate boundary between the Pacific and North America plates. The rupture zone is roughly centered on the earthquake epicenter alongstrike, while peak slips were up-dip of the hypocenter, towards the Japan Trench axis. The rupture was also responsible of a big tsunami that struck a large portion of the coastal area of Tohoku-Oki region. Soon after the earthquake numerous space remote sensing sensors were tasked to acquire as much as possible information about the situation on ground. In particular, a large numbers of SAR images were collected from the ENVISAT and ERS-2 satellite. For this event, a group of researchers from the Istituto Nazionale di Geofisica e Vulcanologia (INGV), named Tohoku-Oki INGV Team, decided to apply a multidisciplinary approach to carry on some analysis addressed to achieve added value outcomes. In this work we present part of the work done by the Team concerning DInSAR data analysis and the methodology to retrieve the seismic source of this extraordinary and disastrous event.

The volume estimation of products is critical for volcanic hazard mitigation, especially for lahars (mudflow) occurrences during rainy season at Merapi, and at active volcanoes subject to lahars elsewhere. Lahars can affect inhabited areas around a volcano, even several years after an eruption. In this work, we present an innovative approach to detect and estimate the volume of pyroclastic flow deposits. We exploited data collected from the very high resolution SAR sensor on board of the COSMO-SkyMed satellite constellation. By comparing a pre-eruption airborne Digital Surface Model (DSM) with a new one obtained applying SAR Interferometry technique to COSMO-SkyMed data, we estimate the volume of the pyroclastic material emitted during the 2010 Merapi eruption. Results show pyroclastic flow deposit thicknesses of up to 75 m that fill canyons on flank of volcano, and are observed up to 16 km far from the mountain summit. The total volume of the deposits is around 117×106m3.

This paper presents an investigation on the rainfall signature for two COSMO-SkyMed (CSK) satellite case studies. Both of them are relative to a severe precipitation weather event, occurred in northwestern Italy (close to Liguria region) on November 3-8, 2011. This event was monitored by using a number of CSK images provided by the Italian Space Agency (ASI). In this case CSK X-SAR data have been compared with the weather radar (WR) Italian Radar National Mosaic. A third case study is relative to Hurricane “Irene” event, occurred in Eastern United States (close to Delaware) on late August 2011. CSK X-SAR images are compared with respect to concurrent ground-based S-band NEXRAD weather radar reflectivities. The correlation of the precipitating cloud fields between CSK X-SAR and WR images is significant in all case studies. An application of a refined XSAR-based precipitation retrieval method is presented. The X-SAR surface response is estimated using ancillary data, such as land cover maps and a digital elevation model (DEM).

We have analysed the seismic source of the active fault generated Van Mw=7.1 earthquake occurred in Eastern Turkey the 23rdOctober 2011. To this aim the surface displacement field has been measured applying SAR Interferometry (InSAR) technique to the available dataset of coseismic COSMO-SkyMed image pairs. The seismic source model has been obtained by the use of a data inversion procedure based on the concurrent application of InSAR techniques and Neural Networks. The proposed approach elaborates the information on the coseismic deformation pattern stemming from available differential interferograms. The interferogram is the expression of the active fault at depth, thus its shape, size and its features somehow refer to the geometry and slip of the fault generating the seism. A Neural Network has been trained to recognize some fault parameters (Length, Width, Strike, Dip, Depth) from the unwrapped interferogram. The retrieval exercise consists in estimating these parameters from the coseismic interferogram exploiting Neural Networks.

The possibility offered by space-based sensors represents an irreplaceable resource for monitoring in near real time the eruption activities. The high revisit time of sensor like MODIS, seems to be the most effective way to mitigate the aviation hazard imaging the phenomenon evolution. In this work we propose a neural networks based approach to the volcanic ash mass retrieval. In comparison with the techniques based on radiative transfer models, the proposed algorithm has shown similar accuracy and faster computation. This issue can be of real interest to address the problems inherent the volcanic activity in short time. A set of MODIS images collected during the Eyjafjallajokull eruption, occurred from the 14thof April to the 23rdof May 2010, has been used to analyze the performance variations due to different selection of the algorithm inputs, i.e. the MODIS channels from visible to thermal infrared electromagnetic spectrum. The best wavelength sets for the retrieval of the ash mass, optical thickness and effective radius have been identified by means of neural network pruning algorithm.

In this work, an automatic method to distinguish, in X-band SAR images such as those supplied by Cosmo-SkyMed, water surfaces (either flooded, or permanent water bodies) from artifacts due to heavy precipitation, is designed to improve flood detection accuracy. The method, mainly based on the fuzzy logic, consists of two main steps, i.e., the detection of low backscatter areas and the classification of each dark object present in the considered SAR image. The algorithm uses ancillary data, such as a local incidence angle map and a Land Cover map. Through the fuzzy logic, it integrates different rules for the detection of low backscatter areas (based on the standard deviation of the backscattering coefficient and on a well-established radar backscattering model), as well as different rules for the classification of the low backscatter (dark) areas (i.e., to distinguish water surfaces from artifacts) based on their geometrical and shape features and on both land cover and local incidence angle.

The major outcomes of the analysis of the COSMO-SkyMed (CSK) synthetic aperture radar (SAR) observations of the area hit by the 2011 Japan tsunami are presented. The height of the tsunami waves was such as to cause a widespread inundation of the coastal area. The SAR acquisitions have been performed on March 12 (i.e., one day after the tsunami occurred) and March 13, 2011 in interferometric mode, so that not only the information on the intensity of the radar signals, but also the complex coherence has been used. The interpretation of the available data has allowed us to detect the flooded areas, as well as the receding of the floodwater from March 12 to March 13, 2011 and the presence of the debris floating above the water surface. Moreover, thanks to the high spatial resolution of the CSK images, the presence of floodwater in some urban areas in the Sendai harbor has been revealed by exploiting the information on the coherence. Our interpretations have been confirmed by a couple of optical images used as benchmarks.

The great eruption of the Icelandic Eyjafjallajokull volcano that occurred from the 14thof April to the 23rdof May 2010 injected large and dense ash clouds into the atmosphere, causing major international air traffic disruption worldwide.

This work aims to assess the potential of Synthetic Aperture Radar (SAR) data combined with optical data to support local administrations in the knowledge of the land use and land cover at regional scale. In particular, the contribution of data available in the future through the SIASGE project, combining L-band and X-band radar imagery, is assessed in order to produce thematic maps. Moreover, the further contribution brought by C-band has been evaluated. The classification, focused on two regions in the north side of Italy, is driven by the legend of already existing maps tackling the real needs of the land managing authorities. As the combination of data from optical imagery is fundamental to achieve good thematic accuracy, the work has exploited the Support Vector Machine learning technique, which is more suitable than standard statistical parametric approaches in this respect. Concerning the classification step, some algorithmic issues has been faced to improve the results, such as training set selection strategy and data fusion techniques. The work has proved as the multi source data set (SAR and optical) is fairly suitable to produce thematic maps comparable to what already in use at local administrative level, allowing to obtain reliable maps with a classification accuracy in the order of 90%.

The Cosmo-SkyMed mission offers a unique opportunity to obtain radar images useful for flood mapping, being characterized by high revisit time, thanks to the four satellites that form its constellation. In the context of a study aiming at evaluating the usefulness of Earth Observation data for managing flood events, particularly focused on Cosmo-SkyMed, an algorithm to map flooded areas from synthetic aperture radar imagery has been developed. It is based on methods developed in previous studies and aims at combining an image segmentation technique based on mathematical morphology and the fuzzy logic that allows us to label the identified objects as flooded or non-flooded. The default parameters of the fuzzy classifier are derived from the outputs of well-established electromagnetic scattering models.

This letter compares the coseismic deformation maps obtained from different synthetic aperture radar (SAR) sensors using the well-known differential SAR interferometry technique. In particular, four deformation maps have been obtained from X-, C-, and L-band SAR sensors onboard COSMO-SkyMed, Envisat, and ALOS satellite missions correspondingly. The test case is the April 6,2009, earthquake (Mw= 6.3). This seismic event struck a densely populated region of the Apennines and was felt all over Central Italy. The SAR data set is rather inhomogeneous, since it includes interferograms with three different wavelengths, four acquisition geometries, different spatial resolutions, variable temporal and spatial baselines, and differently emphasized signal noise. However, we find that the detected displacements are highly comparable. The outcome of this work is that, even though such differences have an impact on the properties of the interferograms, the displacements can be measured with an overall discrepancy of about half the value of the shortest wavelength (COSMO-SkyMed) data set.

In this work is presented a new robust unwrapping-free phase correlation method, for retrieving the coseismic displacement field and the surface rupture fault-trace mapping using optical data. Phase-corr method does not need phase unwrapping and has been proved to be robust under a wide variety of circumstances. The method has been applied at two different test cases, Izmit (Turkey) and Kashmir (Pakistan) earthquakes, occurred on August 17, 1999, and October 8, 2005, respectively. We measured the near-field deformations exploiting two geometrically corrected IRS images with similar look angles in the case of Izmit earthquake, while the Kashmir earthquake coseismic displacement has been retrieved by ASTER data. The results show that Phase-corr method can be used for deriving the coseismic slip offsets due to a large earthquake (and to map its fault trace) using optical data from different sensors.

In this paper, we investigate the performance of pulse-coupled neural networks (PCNNs) to detect the damage caused by an earthquake. PCNN is an unsupervised model in the sense that it does not need to be trained, which makes it an operational tool during crisis events when it is crucial to produce damage maps as soon as the post-event images are available. The damage map resulting from PCNN was validated at a block scale of 120×120m using ground truth obtained by a combination of ground survey and visual inspection of the before- and after-event images. The comparison showed agreement between the change measured by PCNN on block scale and the damage occurred.

The Cosmo-SkyMed mission offers a unique opportunity to obtain radar images useful for flood mapping, being characterized by high revisit time, thanks to the four satellites that form its constellation. In the context of a study aiming at evaluating the usefulness of Earth Observation data for managing flood events, particularly focused on Cosmo-SkyMed, an algorithm to map flooded areas from synthetic aperture radar imagery has been developed. It employs also ancillary data as a land cover map and a digital elevation model. The approach is based on the fuzzy logic because such a theory allows us to exploit the theoretical knowledge about the radar return from inundated areas and to account for simple hydraulic considerations and contextual information.

A destructive (Mw 7.9) earthquake affected the Sichuan province (China) on May 12, 2008. The seismic event ruptured approximately 270 km of the Yingxiu-Beichuan fault and about 70 km of the Guanxian-Anxian fault. Surface effects were suffered over a wide epicentral area (about 300 km E-W and 250 km N-S). We apply the differential synthetic aperture radar interferometry (DInSAR) technique to detect and measure the surface displacement field, using a set of ALOS-PALSAR L-band SAR images. We combine an unprecedented high number of data (25 frames from six adjacent tracks) to encompass the entire area which has coseismically displaced. The resulting mosaic of differential interferograms covers an overall area of about 340 km E-W and 240 km N-S. We investigate the source of the Sichuan earthquake by modeling the DInSAR data. The geometry and position of the fault parameters are inferred by a nonlinear inversion, followed by a linear inversion to retrieve the relative slip distribution. Our results show two different source mechanisms for the 145-long Yingxiu-Beichuan fault and for the 105-long Beichuan-Qingchuan fault. Both faults are characterized by slip concentrations of up to 8 m.

On August 17, 1999, a strong earthquake (Mw ¿ 7.4) occurred along the western sector of the North Anatolian Fault system in Turkey. The epicenter was located near the city of Izmit, 50 km east of Istanbul. Previous works determined the coseismic surface displacements by satellite synthetic aperture radar (SAR) interferometry (InSAR) and satellite optical-image correlation. In 1999, the highest spatial resolution orbiting camera was the panchromatic sensor (PAN), a 5.8-m pixel sensor (SPOT 2 was a 10-m pixel sensor) onboard the Indian Remote Sensing (IRS) satellite. We propose to apply a new phase-correlation method to PAN images to study the coseismic rupture due to the Izmit earthquake. The phase-correlation method does not need phase unwrapping and was proved to be robust under a wide variety of circumstances. Image correlometry deals with the quantification of the subpixel offsets over the whole image, allowing displacement measurement with an accuracy that is proportional to the pixel size. We measured the near-field deformations exploiting two geometrically corrected IRS images with similar look angles. A quality check of the derived offset map was performed by comparison with GPS benchmarks and SPOT offsets. The results show that IRS PAN images can be correlated to derive coseismic slip offsets due to a large earthquake (and to map its fault trace).

This paper investigates the sensitivity of backscattering coefficient to variations of soil moisture and flooding for two kinds of crops, such as wheat and maize, and for deciduous forests. Investigations are based on model simulations at L and C band, VV and HH polarization. At L band, a significant sensitivity to flooding effects is observed for all vegetation covers. At C band, the sensitivity is still acceptable for wheat, while for maize it is present only in case of non uniform cover. For forests, the performance of C band is poor.

This study provides an assessment of the potential for using contextual information with TerraSAR-X backscattering images in classifying urban land-use. Due to the lack of multi-frequency data, a contextual analysis was carried out to extract geometrical information of objects/classes within the images. Anisotropic morphological filters were applied to the backscattering image using a multi-scale approach. A range of different spatial domains were investigated by neural network pruning. The final map of land-use composed of seven different classes of interest was obtained using a Multi-Layer Perceptron neural network with an accuracy of 0.91 in terms of K-coefficient.

The COSMO-SkyMed mission is expected to give a fundamental contribution for flood mapping, because of the high revisit time and throughput achieved by the four satellites that form the constellation. To study the potentiality of COSMO-SkyMed radar data for this purpose, two inundation events are analyzed in this paper, namely the flood occurred in Myanmar in May 2008 and the event that took place in the city of Alessandria (Italy) in April 2009. For the first event, two radar images were considered, one temporally close to the peak of the event, and the other one acquired one week later. As for the Alessandria flood, a time series of images was available, so that an attempt to monitor the temporal evolution of the inundation was accomplished.

Using satellite sensors to detect urban damage and other surface changes due to earthquakes is gaining increasing interest. Optical images at different resolutions and radar images represent useful tools for this application, particularly when more frequent revisit times will be available with the implementation of new missions and future possible constellations of satellites. Very high resolution (VHR) images (on the order of 1 m or less) may provide information at the scale of a single building, whereas images at resolutions on the order of tens of meters may give indications of damage levels at a district scale. Both types of information may be extremely important if provided with sufficient timeliness to rescue teams. The earthquake that hit the city of Bam, Iran, has been taken as a test case, where QuickBird VHR optical images and advanced synthetic aperture radar data were available both before and after the event. Methods to process these data in order to detect damage and to extract features used to estimate damage levels are investigated in this paper, pointing out the significant potential of these satellite data and their possible synergy.

SIGRIS (SIstema di osservazione spaziale per la Gestione del RIschio Sismico) is a pilot project aiming to the realization of a system, based on satellite remote sensing data, for the seismic risk management. The project is funded by the Italian Space Agency (ASI). ASI is deeply interested on the development of new applications, using satellite data, dedicated to the monitoring and management of the natural hazards. SIGRIS is focused on providing the information services for mapping, monitoring, forecasting and awareness of seismic risk. The Earth Observation products are generated by using GPS data, optical and SAR (Synthetic Aperture Radar) images. This project deals with the data exploitation of the new Italian Earth Observation mission: COSMO-SkyMed, a constellation of four satellites equipped with an X-band high resolution SAR.

Remote sensing sensors for detecting urban damage and other surface changes due to earthquakes is gaining increasing interest. To this aim optical images can represent useful tools for this application thanks to their very high ground geometric resolution, especially when more frequent revisit times will be feasible to the implementation of new missions and future possible constellations of satellites. Sub-meter resolution images at visible frequencies are able to provide information at the single building scale. This kind of information is extremely important if provided with sufficient timeliness to rescue teams. In this work, the December 26th, 2003, earthquake that hit the ancient city of Bam (Iran) has been investigated. The urban area was very close to the epicenter of the seism thus causing strong damage to the urban structures. Pre- and post-earthquake QuickBird panchromatic images have been used to show the capability of this data to map damage at building scale by means of segmentation approach based on the application of morphological operators. A validation process has been performed by comparing the map of damage levels at single building scale with a detailed ground-based damage map provided byinsitusurvey.

Urban areas are composed of numerous materials arranged by humans in complex ways. A simple building may appear as a complex structure with many architectural details surrounded by gardens, trees, buildings, roads, social and technical infrastructure and many temporary objects, such as cars, buses or daily markets. In this paper, we analyze the effectiveness of 8 textural features (resulting from the Grey Level Co-occurrence Matrix) derived from a 50 cm WorldVieW-1 image of Washington D.C. (U.S.A.). The information extracted from the panchromatic and textural features are fused and processed by a Multi-Layer Perceptron (MPL) neural network producing a land-use map with accuracy above 0.90 in term of K-coefficient.

Parametric and nonparametric approaches to evaluate land-cover change detection using very high resolution (VHR) satellite imagery are applied to the analysis of the demolition of the Rocky Flats nuclear weapons facility located near Denver, CO. Both maximum-likelihood and neural network classifiers are used to validate a new parallel architecture which improves the accuracy when applied to VHR satellite imagery for the study of land-cover change between sequential satellite acquisitions. An enhancement of about 14% was found between the single-step classification and the new parallel architecture, confirming the advantage and the robust improvement obtained with this architecture regardless of the classification algorithm used. In this paper, we demonstrate and document the demolition and removal of hundreds of buildings taken down to bare soil between 2003 and 2005 at the Rocky Flats site.

We present two different change detection techniques to monitor surface changes that occurred at the Rocky Flats nuclear weapons facility located immediately to the North West of the city of Denver, Colorado, USA. The site started being cleaned up and dismantled in 1998 and was completed in 2005. The first Change Detection method is based on a Maximum Likelihood classifier, while the other is an approach based on a Neural Network architecture called NAHIRI (Neural Architecture for High-Resolution Imagery) to produce change detection maps from very high-resolution satellite imagery. NAHIRI simultaneously exploits spectral and temporal information by adding a filter, directly stemming from the multi- temporal information, to the classification changes derived from the multi-spectral data. In fact, the distinctive feature of this method is that the NNs exploit both the multi-spectral and the multi-temporal information in parallel that are associated with the changed values of the pixel spectral reflectances. The quantitative results are analyzed in order to single out advantages and shortcomings of the two different approaches.

An inundation event occurred in Italy on November 1994 has been analysed in order to asses the capability to map the flooded areas by radar (SAR) images in support to civil protection interventions. ERS images collected before and after the inundation have shown the presence of different scattering mechanisms occurring in the flooded areas and originating either an increase or a decrease of the backscattering coefficient, depending on the land covers type. This analysis has allowed us to develop a flood discrimination procedure based on a fuzzy approach able to integrate all the possible sources of available data (SAR images, land cover and a DEM) and prior information (cover dependent backscattering behaviour). A comparison with a ground survey providing the maximum flood extension has given encouraging results.

The prompt detection, mapping and assessment of urban damages due to earthquakes is a key point, particularly in remote areas or where the infrastructures are not well developed to ensure the necessary communication exchanges or where their operability has strongly decreased as a consequence of the event. The combination of Synthetic Aperture Radar (SAR) data and optical images is a promising and suitable approach. We propose two test cases, the 1999 Izmit (Turkey) and the 2003 Bam (Iran) earthquakes where we investigate the capability to detect urban changes and classify them. Moreover, a comparison with ground based data is also shown.