The GEORES project is funded by the Italian Space Agency (ASI) and aims to develop a geospatial application meant to improve environmental sustainability and resilience to climate changes in urban areas, based on the synergistic use of the most advanced Earth Observation (EO) technologies, Artificial Intelligence (AI) and eXplainable AI (XAI). GEORES is organized into four main modules to support management of the main risks associated with land degradation: (1) Sediment Connectivity; (2) Land Displacement; (3) Urban Floods; (4) Urban Wildfires. For each module, EO data, calculation models and algorithms are integrated to identify "hot-spots" of urban and peri-urban territory at high risk from the point of view of land degradation caused by phenomena of hydrogeological instability, sediment flow or vegetation fires. The extracted information is expressed with specific indicators ("geo-analytics") calculated dynamically and automatically. The demonstration is undertaken in the Metropolitan City of Bari and Gargano Promontory, Apulia Region, southern Italy, and foresees the engagement of final users (i.e. Regional Civil Protection and Municipality of Bari).

Bayesian estimation of posterior probabilities for the presence of floodwaters, coupled with accurate time series regression methods, show good performance in the monitoring of inundations at high temporal and spatial resolution from long stacks of synthetic aperture radar (SAR) data. We report results on the integration of SAR intensity and cascaded InSAR coherence time series in different polarization channels within a Bayesian framework. The method is being tested over sites in both northern and southern Italy, with X- and C-band SAR data. The results indicate some advantage in using more than one independent channel in the Bayesian inference for some types of land cover, in terms of area under the curve (AUC) when compared to independent flood maps acquired over known events. Stacks of surface water confidence levels computed over both test sites show promising characteristics, both on agricultural and coastal areas.

Multi-temporal remotely sensed data are a precious source of information for high spatial and temporal resolution flood mapping. We present a methodology for flood mapping through processing of long time series of Sentinel-l SAR data, as well as ancillary information. A Bayesian framework is adopted to derive probabilistic maps of the presence of flood waters, through modeling of backscatter time series, based on the as-sumption that floods represent impulsive temporal anomalies. We illustrate some results on a time series of Sentinel-l data acquired from 2015 to 2021 over a test area on the Basento river watershed, Basilicata Region, in Southern Italy, recurrently subject to floods.

We observe relatively high InSAR mean coherence levels over badlands, i.e. clayey bare soil areas, on a test site in the Basilicata region, in southern Italy. Time series of InSAR coherences on cascaded short-baseline image pairs, obtained from stacks of Sentinel-1 SAR images, exhibit oscillating behaviour, with significant correlation with cumulated rainfall levels on bad-land areas, while on other areas with crops or spontaneous vegetation the correlation is lower, and a seasonal trend is instead statistically significant. These observations seem to point to the possibility of investigating erosion phenomena over badland areas through InSAR time series, which involves a significant step forward, in terms of spatial and temporal resolution, with respect to traditional measurements which require repeated topographic surveys at long intervals, or sparse in-field point measurements.

The Earth Observation is proven to be extremely useful in the management of the aftermaths of tsunami and earthquakes. ESA-Disaster Risk Reduction (DRR) is a project within the Earth Observation for Sustainable Development (EO4SD) program aiming to demonstrate on large scale in developing countries the usefulness of Earth Observation to reduce disaster risks and to support the management of the post-event rehabilitation and reconstructions. Among the activities of the project, the ESA-DRR team, following the dramatic event of the Sulawesi earthquake (28 September 2018) supported the Indonesian authorities, through the cooperation with the Asian Development Bank (ADB), the reconstruction and rehabilitation of Palu area. The provided EO services are presented.

Indonesia is periodically affected by severe volcanic eruptions and earthquakes, which are geologically coupled to the convergence of the Australian tectonic plate beneath the Sunda Plate. Multi-temporal SAR interferometry (MTI) can be used to support studying and modelling of terrain movements. This work is aimed at performing an analysis of ground displacements over Indonesian sites through MTI techniques. Two test sites in Sumatra and Java have been selected according to the availability of archived SAR data, GNSS networks, and geological data. Both COSMO-SkyMed (CSK) and Senitnel-1 data-sets have been processed through MTI algorithms. The derived displacement maps have been interpreted according to the available geological and geophysical information.

The Rheticus® cloud-based platform provides continuous monitoring services of the Earth's surface. One of the services provided by Rheticus® is the Displacement Geoinformation Service, which offers monthly monitoring of millimetric displacements of the ground surface, landslide areas, the stability of infrastructures, and subsidence due to groundwater withdrawal/entry or from the excavation of mines and tunnels. To provide this information, the Rheticus® platform processes a large amount of Geospatial Big Data. In particular, Rheticus® is capable to process Synthetic Aperture Radar images acquired by the X-band COSMO-SkyMed constellation, as well as satellite Open Data provided by Copernicus Sentinels, and it is capable to integrate local INSPIRE data sources. In this paper, we summarize the main features of the Rheticus® services and we provide examples of the detection and monitoring of geohazard and infrastructure instabilities through Multitemporal InSAR techniques. Furthermore, we outline the porting activity and the efficient implementation of the most time-consuming algorithmic kernels in the GPGPU environment.

Satellite remote sensing radar technologies provide powerful tools for geohazard monitoring and risk management at synoptic scale. In particular, advanced Multi-Temporal SAR Interferometric algorithms are capable to detect ground deformations and structural instabilities with millimetric precision, but impose strong requirements in terms of hardware re-sources. Recent advances in GPU computing and programming hold promise for time efficient implementation of imaging algorithms, thus enhancing the development of advanced Emergency Management Services based on Earth Observation technologies. In this study, a preliminary assessment of the potentials of GPU processing is carried out, by comparing CPU (single- and multi-thread) and GPU implementations of InSAR time-consuming algorithm kernels. In particular, it is focused on the fine coregistration of SAR interferometric pairs, a crucial step in the interferogram generation process. Experimental results are discussed.

Thanks to the technological maturity as well as to the wide availability of SAR data, Multi-temporal SAR Interferometry (MTInSAR) can be used to support systems devoted to environmental monitoring and risk management. In particular, high resolution X-band MTInSAR applications are also suitable for monitoring single man-made structures (buildings, bridges, railways and highways). The paper presents examples concerning the application of MTInSAR techniques and COSMO-SkyMed constellation for instability monitoring of infrastructures and, in particular, harbor docks and railways.

By mapping the concentration of chlorophyll-a (CHL) and the temperature of the sea surface (SST), satellite images reveal the complex dynamics of marine waters and prove to be a very powerful tool when used to detect potential fishing areas, significantly reducing the time of the search, the fuel consumption and the human effort, and simultaneously increasing the CPUE (catch per unit effort). In the present work, various techniques of multi-sensor, multi-resolution and multi-temporal data fusion are applied to multi-spectral satellite image data of MODIS-AQUA, MODIS-TERRA and VIIRS sensors, in order to detect "fronts" of chlorophyll concentration and temperature on the sea surface. According to the physical model of the phenomena, these fronts are generated by the upwelling of cold waters rich of nutrients (phytoplankton) which correspond to areas with a high concentration of pelagic fish and are characterized by high values of local gradients of SST and CHL with anti-parallel orientation. An automatic procedure has been developed to calibrate and validate the production in near-real time of daily maps of expected good fishing grounds to be provided to the FEDERPESCA fleet. The same procedure could be optimized also for other seas.

We present a research project, funded by the Italian Space Agency (ASI), aimed at performing 2D and 3D Focusing of COSMO/SkyMed (CSK) SAR Data. We describe the main objectives of the project, briefly illustrate employed techniques, and finally present the obtained results. The latter show that sub-meter resolution can be achieved in the enhanced spotlight CSK acquisition mode, and that by using 3D focusing it is possible to resolve scatterers at different slant heights within the same range-azimuth resolution cell, even in areas characterized by severe height discontinuities and large thermal dilations effects.

The present study is aimed at investigating the potentialities of the COSMO/SkyMed (CSK) constellation for ground elevation measurement with particular attention devoted to the impact of the improved spatial resolution wrt the previous SAR sensors. Assuming no movement and successful orbital error removal, the main problem in height computation through InSAR techniques derives from the interferometric phase artifacts related to the interaction between microwave and the lower layers of the atmosphere (APS, Atmospheric Phase Screen). Different strategies can be adopted to filter out this signal, ranging from the exploitation of the well-known spatial and temporal statistics of the APS to the estimation of independent APS measurements through Numerical Weather Prediction (NWP) models. Their feasibility and the achievable accuracies are discussed here.

COSMO/SKYMED is currently the unique constellation of synthetic aperture radar (SAR) sensors operative, which is also for civilian use. On April 6, 2009, an Mw 6.3 earthquake struck the city of l'Aquila in Central Italy. The constellation acquired data stacks over the hit area at an unprecedented temporal rate. In this letter, the results obtained by processing several data set via two independent multitemporal differential interferometric SAR techniques are presented to demonstrate the capability of this constellation in postseismic deformations monitoring.

Image alignment is a crucial step in synthetic aperture radar (SAR) interferometry. Interferogram formation requires images to be coregistered with an accuracy of better than a few tenths of a resolution cell to avoid significant loss of phase coherence. In conventional interferometric precise coregistration methods for full-resolution SAR data, a 2-D polynomial of low degree is usually chosen as warp function, and the polynomial parameters are estimated through least squares fit from the shifts measured on image windows. In case of rough topography or long baselines, the polynomial approximation may become inaccurate, leading to local misregistrations. These effects increase with spatial resolution of the sensor. An improved elevation-assisted image-coregistration procedure can be adopted to provide better prediction of the offset vectors. This approach computes pixel by pixel the correspondence between master and slave acquisitions by using the orbital data and a reference digital elevation model (DEM). This paper aims to assess the performance of this procedure w.r.t. the “standard” one based on polynomial approximation. Analytical relationships and simulations are used to evaluate the improvement of the DEM-assisted procedure w.r.t. the polynomial approximation as well as the impact of the finite vertical accuracy of the DEM on the final coregistration precision for different resolutions and baselines. The two approaches are then evaluated experimentally by processing high-resolution SAR data provided by the COnstellation of small Satellites for the Mediterranean basin Observation (COSMO/SkyMed) and TerraSAR-X missions, acquired over mountainous areas in Italy and Tanzania, respectively. Residual-range pixel offsets and interferometric coherence are used as quality figure.

On April 6th2009 a Mw=6.3 earthquake struck the area around the city of L'Aquila in Italy. SAR systems have been proven to be valuable sensors for analyzing the effect of earthquakes and monitoring post-seismic displacements. Due to the low deformation rate, the study of post-seismic events requires the use of a multi-temporal InSAR approach. COSMO/SKYMED is a constellation of SAR sensors of 4 X-band sensors operative also for the civilian use. Thanks to the availability of a stack of ascending acquisitions, ad hoc programmed by ASI on the area stricken by the earthquake, it was possible to provide post-seismic deformation maps by using two different multi-temporal interferometric approaches: the SPINUA and SBAS techniques. The work is carried out in the framework of the MORFEO project dedicated to the monitoring of the landslides risk by means of Earth Observation data. The displacement maps related to the post-seismic activity are presented and commented. The results clearly show the potentiality of the COSMO/SKYMED constellation use for emergency monitoring.

This work is focused on the phase validation of interferograms obtained by combining COSMO-SkyMed SAR images acquired by a single satellite (temporal baseline coincident with the orbital repeat cycle) or even by two satellites of the SAR constellation in equi-phased configuration on the orbital plane (temporal baseline: 8 days), thus minimizing the temporal decorrelation. Both qualitative and quantitative analyses have been therefore carried out for HI (HIMAGE: stripmap, single polarization) and S2 (enhanced spotlight) imaging modes, in order to proof whether or not COSMO-SkyMed constellation is well suited for SAR interferometry.

The detectability of the deformation pattern produced by the June 15, 1995 Aigion earthquake with DInSAR techniques is ensured by its magnitude (Mw=6.3), shallow depth and dip-slip mechanism. In this paper, stacking procedures are applied to a series of ERS interferograms in order to filter out from the differential phase field the atmospheric signal, and an a posteriori test is used to check the statistical properties of the atmospheric signal both in time and space. Based on the DlnSAR-derived deformation pattern, a new fault model is proposed that takes into account the crustal layering of the western part of the Gulf of Corinth.

This paper proposes a unified framework for predicting optimized pairing strategies for interferometric processing of multipass synthetic aperture radar data. The approach consists in a minimum spanning tree (MST) structure based on a distance function encoding an a priori model for the interferometric quality of each image pair. Using a distance function modeled after the interferometric coherence allows reproducing many "small baseline" strategies presented in the recent literature. A novel application of the method to the processing steps of image coregistration and equalization is illustrated, using a test European Remote Sensing Satellite dataset. Widespread methods used for these two operations rely on the computation of the amplitude cross correlation over a large number of corresponding tie patches distributed over the scene. Geometric shift and radiometric equalization parameters are estimated over the patches and used, respectively, within a polynomial warp model and a radiometric correction scheme. The number of reliable patches available behaves similarly to the interferometric synthetic aperture radar (InSAR) coherence with respect to the baselines, and can be assimilated to a quality figure for the derivation of the MST. Results show an improvement in the quality of the stepwise (SW)-processed image stack with respect to the classical single-master procedure, confirming that the SW approach is able to provide better conditions for the estimation of correlation-related InSAR parameters

We illustrate, through a sample application to a difficult landslide test site, the use of a novel method to detect potentially stable objects in Persistent Scatterers SAR Interferometry (PSI). Conventional PSI processing involves selecting first-guess potential stable objects, called PS Candidates (PSC), through thresholding of the amplitude dispersion index. This method can lead, in applications to scenes characterized by scarce urbanization, to very low PSC numbers, insufficient for a successful subsequent phase analysis if their spatial distribution is very sparse. Our classification-based approach relies on the proven fact that urban areas are more likely to contain PS pixels than any other land-cover class. Therefore, using pixels belonging to the urban land-cover class as PSC is a convenient way of increasing the number of initial fiducial points while keeping false alarm probabilities to reasonable levels. Results show that PSC belonging to the urban class, selected through simple external classification algorithms, lead to more consistent results for the final PS, both in terms of spatial density, and of reliability of displacement series.

We present an assessment of stepwise co-registration procedures applied to multitemporal SAR datasets. Images are connected in pairs through a minimum spanning tree structure, obtained by adopting a distance measure which is a function of the expected co-registration quality. Experiments have been performed on a test dataset by a) directly estimating the (a posteriori) co-registration quality over all possible image combinations, b) using an a priori model inspired by similar models for the multitemporal InSAR coherence, with parameters obtained experimentally, c) using the same a priori model with first-guess parameters. Performances were evaluated by analyzing the amplitude inverse coefficient of variation distribution over the co-registered image stacks obtained by the three procedures above. Results show that, although the best coupling strategy depends on the particular dataset and is thus difficult to model via general rules, a nonnegligible improvement in the performance of persistent scatterers interferometry techniques can be obtained by adopting stepwise approaches based on a priori models for the expected co-registration quality, rather than using a single acquisition as master

The analysis of multi-temporal SAR data-sets encountered large interest in the remote sensing community during the past few years.The main effort goes toward the extraction of ground displacements signals by means of differential interferometric techniques.In this opera- tional framework an important processing step concerns the estimation and subtraction of signals due to atmospheric artifacts and processing errors.In the present work we apply the technique of Blind Source Separation (BSS) by using the algorithm of Independent Component Analysis (ICA) to Permanent Scatterer processing in order to perform the separation of different signal components.Preliminary investigations are carried out both on simulated and real ERS-1/2 data and results are reported and commented. I. INTRODUCTION The permanent scatterers (PS) technique (1) allows the analysis of the phase information over single objects and thus typically on man-made structures, characterized by a high temporal phase stability. Where such object are present, even when surrounded by low-coherence condi- tions, a non-conventional spatio-temporal phase analysis can be performed, together with an accurate atmospheric phase screen estimation and removal. This new technique, although very promising as an operational tool for the analysis of long series of SAR images, represents still a challenge in termsof computational burden and exhibits modelling aspects not yet fully understood. The main problemsencountered when dealing with data on moun- tainous, un-urbanised areas involve the steep topography causing considerable DEM errors (∆h(x, y)), increased atmospheric effects compared to flat areas, and increased geometric distortions; also, the scarcity of man-made structures lowers considerably the number of detectable stable points, further complicating the processing (5). Many as pectsof the complicationsintroduced in the

Quantitative analysis of multi-temporal SAR datasets requires accurate radiometric calibration. This can usually be achieved by considering a number of multiplicative factors, to be applied to statistically-homogeneous areas on each image to be calibrated. We propose an automated procedure to easily obtain a relative radiometric calibration of a stack of an arbitrary number of ERS SAR images, all coregistered to a unique master. The procedure relies upon the theoretical invariance of both system and scene parameters for targets whose amplitude values are statistically correlated with each other. In practice, with the above-mentioned assumptions, the only varying quantity to take into account is the temporal variation of the calibration factor. We show, based on experiments on two multi-temporal datasets, that the ERS-1 calibration constant exhibits practically no variations except for the expected statistical fluctuations of about /spl plusmn/1 dB. This observation is consistent with the reports on the operations of the ERS-2 satellite, published by ESA. The proposed procedure then consists in applying the full absolute calibration procedure only to the master image, and then relatively correct all other images in the stack by the method described.

The first basic step in the permanent scatterers analysis technique is the location of the stable reflectors. For this purpose, an inter-image amplitude analysis, on a pixel-by-pixel basis, is performed, and the scatterers showing stable amplitude response are named Permanent Scatterer Candidates (PSC). The effects of the interpolation of the SLC (Single Look Complex) images are considered, with particular attention to the oversampling factor and to the interpolation kernels. Finally, experimental results are shown to confirm the PSC theory and the interpolation analysis.