Classifying ground deformation processes, such as landslides, subsidence, deep-seated gravitational slope deformations (DSGSDs), and mining-induced deformations, is key for large-scale hazard assessment and national land use management. Earth observation provides heterogeneous data over the same geographic region, including Interferometric Synthetic Aperture Radar (InSAR) time series, multispectral imagery, and terrain products. However, effectively integrating such spatiotemporal information from multimodal datasets remains a major challenge. In order to fully utilize the rich information contained in the time series and to exploit the complementary strengths of spatial and temporal data, we propose a dual-branch deep learning approach that integrates InSAR ground deformation time series with geospatial information for classifying slow-moving ground deformation processes. To validate the approach, we construct a ground deformation dataset containing over 26,000 Active Deformation Areas (ADAs), labelled into four deformation types: Landslide, Subsidence, DSGSD, and Mining. Results demonstrate that our model achieves an overall classification accuracy exceeding 90% on both ascending and descending test dataset, though confusion remains between certain classes, such as landslides and DSGSD. Explainable AI (XAI) analysis indicates that spatial and morphological features contribute more significantly to classification performance than temporal deformation patterns, with clearer distinctions for subsidence and mining, but more overlap between landslides and DSGSDs. This work highlights the strength of multi modal data fusion method to classify ground deformation processes, while setting the stage for future research.

This work proposes a supervised classifier of ground motion phenomena using as main input the SAR Differential Interferometry (DInSAR) data contained in the European Ground Motion Service (EGMS) datasets. The classifier operates on the active deformation areas (ADAs) extracted from the EGMS data, which are categorized into three deformation classes, i.e. deep-seated gravitational slope deformation (DSGSD), landslide and subsidence. Digital Elevation Models (DEM) and Land Cover maps are used as ancillary input data, whereas landslide and subsidence inventories are employed as ground truth to form the training dataset. The implemented machine learning (ML) classifier employs the Extreme Gradient Boosting technique on a set of features extracted from the input data. Feature importance is analysed to provide an insight on the physical meaning of the implemented classification, moving towards the explainable artificial intelligence paradigm. The results show good classification performance on the test dataset. The classification reliability is evaluated on the unseen data through class probabilities.

This work addresses the automatic extraction of active deformation areas (ADAs) using a pan European ground deformation dataset, provided by the European Ground Motion Service. The ADA extraction routine is based on the ADA Finder tool, which selects the persistent scatterers that are likely to belong to a surface affected by a ground deformation phenomenon. The result consists of two European ADA maps, associated with the Sentinel-1 ascending and descending trajectories. A preliminary validation of the European ADA maps is addressed by analysing the detected ADAs in the territory of Valle d'Aosta region (Italy), that is affected by the presence of landslides and deep-seated gravitational slope deformations, showing a limited number of false positives.

Measuring glacier elevation change is crucial information for estimating glacier mass balance, calibrating mass balance and climate models, and assessing the impact of global warming. We examined the potentiality of clinometry to quantify glacier elevation changes. This technique allows calculating the elevation of the points that lie on the margins of the shadows cast by the local topography on monoscopic optical images. Mapping the shadow position across different images permits quantifying surface elevation changes. We applied clinometry to Sentinel-2 images of the Aletsch Glacier (Switzerland) and we measured a glacier thinning of −1.9 ± 1.7 ma−1 between 2017 and 2021, in agreement with previous observations.

This letter aims at investigating the potential of high-resolution (up to$0.7\times0.5\,\,\text{m}^{2}$) synthetic aperture radar (SAR) images in generating digital elevation models (DEMs) using the radargrammetry technique. In this work, we process two SAR images recorded by the Capella Space X-band satellite-borne radar sensor on two consecutive days, October 2 and 3, 2021, over La Palma (Canary Islands, Spain) during the Cumbre Vieja volcanic eruption. We adopt an iterative point-aggregation algorithm to identify matching pixels between the two images; then, the height estimation is performed using a distance minimization routine over the previously identified pairs of matching points. The resultant radargrammetric DEM is validated against a lidar-based DEM for various land cover (LC) classes, showing a good agreement in the areas less affected by lava flow. An estimation of the lava thickness is performed, yielding profiles of the cone area, which are compared to the photogrammetry estimates obtained from the Pléiades mission data.

Generation of Digital Elevation Models (DEMs) using Synthetic Aperture Radar (SAR) data is a valid flexible alternative to other surveying techniques. In this work we adopt radargrammetry on two SAR images recorded by the Capella Space X-band spaceborne sensor on two consecutive days over La Palma Island (Canary Islands, Spain) during the Cumbre Vieja volcanic eruption, with the aim of investigating the potential of these high-resolution (up to 0.7 by 0.5 m2) SAR images in generating precise DEMs. The adopted method employs a multilook-based spatial filter followed by an iterative region-growing algorithm to identify matching pixels between the two images. The resultant radargrammetric DEM is compared with a Lidar-based DEM, showing a good agreement in the areas less affected by lava flow.

In this letter, we propose a Deep Learning (DL) based approach which exploits multispectral Sentinel-2 open-source data and a small-size inventory to map artisanal and small-scale mines (ASM). The study area is in central northern Burkina Faso (Africa) and is characterized by a semi-desert environment that makes mapping challenging. In sub-Saharan Africa, artisanal and small-scale mining represents a source of subsistence for a significant number of individuals. However, because ASM are often illegal and uncontrolled, the materials employed in the excavation process are highly dangerous for the environment as well as for the lives of the people involved in the mining activities. One of the most important aspects regarding ASM is the record of their spatial location which, at the moment, is missing in most of the African regions. Performance evaluation of two state-of-art DL architectures (U-Net, and Attention Deep Supervised Multi-Scale U-Net - ADSMS U-Net) is provided, along with an in-depth analysis of the predictions when dealing with both dry and rainy seasons. The ADSMS U-Net architecture yields generally more accurate predictions than the basic U-Net allowing us to better discriminate ASM in such an environment. The findings show that the proposed approach can detect ASM in semi-desertic areas starting with a few samples at a low cost in terms of both human and financial resources.

In this letter, we use deep learning convolutional neural networks (CNNs) to compare the landslide mapping and classification performances of optical images (from Sentinel-2) and synthetic aperture radar (SAR) images (from Sentinel-1). The training, validation, and test zones used to independently evaluate the performance of the CNN on different datasets are located in the eastern Iburi subprefecture in Hokkaido, where, at 03.08 local time (JST) on September 6, 2018, an Mw 6.6 earthquake triggered about 8000 coseismic landslides. We analyzed the conditions before and after the earthquake exploiting multipolarization SAR as well as optical data by means of a CNN implemented in TensorFlow that points out the locations where the landslide class is predicted as more likely. As expected, the CNN runs on optical images proved itself excellent for the landslide detection task, achieving an overall accuracy of 98.96%, while CNNs based on the combination of ground range detected (GRD) SAR data reached overall accuracies beyond 95%. Our findings show that the integrated use of SAR data may also allow for rapid detection even during storms and under dense cloud cover and provides comparable accuracy to classical optical change detection in landslide recognition and detection.

This work addresses the problem of estimating and filtering the Atmospheric Phase Screen from a stack of image phases, extracted from a sequence of SAR images using the PSInSAR approach. Leveraging the assumption that the atmospheric disturbance is a signal with a high spatial correlation and low temporal correlation, a Nonequispaced FFT is used to estimate the spatial spectrum of the input data, which are distributed spatially on a non-uniform grid, then perform a lowpass filter, followed by a highpass temporal filter. The obtained APS estimate is then removed from the input data. The performance of the proposed method is evaluated over Sentinel-1 SAR data to study mining-induced terrain deformations in the Polkovice area, Poland.

In this paper, an approach to Persistent Scatterer Interferometry (PSI) is used to derive deformation measurements over the Catalonia region (Northern Spain). A set of tools to control the quality of the 2+1d phase unwrapping, one of the key steps of the proposed procedure, are described and applied over a set of Sentinel-1A (S-1A) images. The results, derived using 64 S1-A images comprising the period from March 2015 to May 2017, are analyzed in the last section of the paper. Finally, the deformation velocity map and the time series are described, in particular over the urban areas, where the proposed approach yields the best results.

This paper reports the results of two terrestrial surveys aimed at monitoring two Alpine glaciers located in Italy and Spain respectively, and carried out using a Ground Based SAR interferometer. Although the monitoring of glaciers based on this technique does not represent a novelty, these two case studies are peculiar, due to the characteristics of the Alpine glaciers among which the dominant role of the atmospheric phase screen (APS) on the radar signal propagation. These kind of glaciers, with their climate and geographical features, often demand a detailed analysis of the acquired data, at small temporal (less than an hour), and spatial (a few square meters) scale. The meteorological conditions, which affect the dynamics of the glaciers, deeply influenced the backscattering behavior, demanding a careful analysis of the amplitude of the radar signal, to characterize the surface, and of the interferometric phase, to evaluate the role of the APS. Only after the correction of the APS, a final accuracy of a few millimeters/day was attained in the daily velocity of the glacier in both cases.

In the last decade, Persistent Scatterer Interferometry (PSI) and SAR tomography (TomoSAR) have been used for reconstructing the elevation profile of a scene, starting from a set of co-registered Synthetic Aperture Radar (SAR) images. The possible advantage of TomoSAR over classical interferometric methods consists in the potential capability of improving the detection of single scatterers presenting stable proprieties over time (Persistent Scatterers or PS), as well as to enable the detection of multiple scatterers interfering within the same range-azimuth resolution cell. In urban environment, when only single dominant scatterers are present in each range-azimuth resolution cell, both methods can be exploited to estimate the altitude, deformation rate and thermal expansion of a subset of reliable scatterers, which are selected on the basis of different criteria. This paper is focused on a performance analysis of the two class of methods, using the results obtained in urban environment on simulated and real TerraSAR-X data. A concise description of both techniques, along with a discussion on their potential capabilities in selecting the most reliable scatterers, is given.

This paper describes some of the key features of the Persistent Scatterer Interferometry chain of the Geomatics (PSIG) Division of CTTC. The paper firstly provides an overview of the entire PSI chain. It then focuses on the first part of the chain, which provides the input data for the estimation of the Atmospheric Phase Screen (APS). In this part, the so-called Cousin Persistent Scatterers (CPSs) are sought, which are Persistent Scatterers (PSs) characterized by a moderate spatial phase variation that ensures a correct phase unwrapping. The main output of this part of the chain is a set of correctly unwrapped and temporally ordered phases, which are computed on CPSs that cover homogeneously the area of interest. In order to do so a procedure to check the consistency of phase unwrapping is used. The paper describes the used algorithms and illustrates their performances using a set of TerraSAR-X StripMap images over the metropolitan area of Barcelona.

Deformation monitoring using ground-based synthetic aperture radar (GB-SAR) data usually exploits the interferometric phases. In this letter, a new noninterferometric procedure is proposed, which exploits the geometric content of GB-SAR amplitude imagery and estimates deformation through image matching. This letter describes, step by step, this procedure. In order to achieve acceptable deformation measurement performances, the technique needs special targets, which have to guarantee a good image matching quality. If the available natural targets are insufficient, artificial corner reflectors are required. The new approach overcomes some of the main limitations of GB-SAR interferometry; it yields aliasing-free deformation estimates, is insensitive to atmospheric effects, and provides 2-D displacement measurements while interferometry only has a mono-dimensional measurement capability. Several experiments focusing on the performances of the new procedure are described. The procedure is validated using different scenarios. On a real-size landslide scenario, a mean absolute error over 12 corner reflectors of 0.59 cm is achieved, i.e., 1/85th of the pixel size. These encouraging results can be of interest for several deformation monitoring applications.

This letter focuses on the thermal expansion component of persistent scatterer (PS) interferometry (PSI), which is a result of temperature differences in the imaged area between synthetic aperture radar (SAR) acquisitions. This letter is based on very high resolution X-band StripMap SAR data captured by the TerraSAR-X spaceborne sensor. The X-band SAR interferometric phases are highly influenced by the thermal dilation of the imaged objects. This phenomenon can have a strong impact on the PSI products, particularly on the deformation velocity maps, if not properly handled during the PSI analysis. In this letter, we propose a strategy to deal with the thermal dilation phase component, which involves further developing the standard two-parameter PSI model (deformation velocity and residual topographic error) with a third unknown parameter called the thermal dilation parameter, which is estimated for each PS. The map obtained from plotting this parameter for all PSs of a given area is hereafter called thermal map. This letter describes the proposed model and outlines the issue of parameter estimability. In addition, the potential of exploiting the thermal maps is analyzed by illustrating two examples of the Barcelona (Spain) metropolitan area. Thermal maps provide two types of information: The first one is the coefficient of thermal expansion of the observed objects, while the second one, which is related to the pattern of the thermal dilation parameter, gives information about the static structure of these objects. Two important aspects that influence the exploitation of thermal maps are discussed in the last section of this letter: the line-of-sight nature of the derived estimates and the achievable precision in the estimation of the coefficient of thermal expansion.

The DInSAR technique enables to determine with precision the surface displacements, using a combination of multiple interferograms. The DInSAR processing steps generate different kinds of errors, which propagate in the entire chain. This work is focused on a particular type of error generated during the DInSAR processing: the unwrapping related errors. The errors generated during the unwrapping process and the use of a procedure to automatically detect and correct them are presented in this work. By an iterative process and exploiting the SVD least squares method for outliers rejection, this procedure determines the phases values associated with each SAR image, starting from a stack of interferograms. It works on previously selected pixels and provides good results with high observation redundancy. The effectiveness of the procedure is illustrated by using ERS SAR data acquired over Barcelona (Spain).