Repeat-pass interferometric SAR (InSAR) is widely used for a variety of application scenarios, such as terrain displacement and subsidence monitoring or measuring the state of infrastructures. In this context, the development of effective algorithms to detect temporal and spatial changes in the radar targets becomes of paramount importance. Typically, state-of-the-art methods only return the spatial, temporal, or both locations of the occurred changes without any information about the causes. In this article, we present a novel change detection method able to infer not only whether a target has changed and when but also the reason why a change is detected, defining the concepts of definitive and temporary changes (TCs). This is done by jointly exploiting four radar amplitude images and the corresponding six interferometric coherences computed at different temporal baselines. To this aim, we propose a new deep learning (DL)-based framework based on a fully convolutional neural network (CNN) called deep analysis for nonstable InSAR targets network (DANI-NET). The network design and training strategy are driven by explainable AI (XAI) principles. Here, we rely on the development of fully synthetic training and testing datasets by following a robust statistical derivation, which allows for a full understanding of the network outcomes. We evaluate the DANI-NET performance on an independent synthetic dataset and we compare it to the state-of-the-art permutational change detection (PCD), a nonparametric statistical approach, achieving extremely competitive results. Moreover, we also provide a feature analysis on the prediction explainability using the SHAP method. Finally, we apply DANI-NET to two real-case scenarios, by considering a Sentinel-1 repeat-pass dataset acquired over Iceland during the 2023–2024 Sundhnúkur eruptions and a TanDEM-X multitemporal stack acquired over an open-pit mining site. We validate the method over the Iceland dataset, where we compare the predicted lava field extension with external reference measurements. In both cases, DANI-NET produces high-quality results and adds the possibility of investigating the nature of the changes caused by either natural or man-induced phenomena.

Developing algorithms to detect temporal and spatial changes in radar targets is paramount. This paper specifically addresses the temporal change detection aspect, introducing a rapid non-parametric Coherent Change Detection (CCD) algorithm named Fast-Permutational Change Detection (F-PCD). The F-PCD identifies temporal Change Points (CPs) in a radar target by recognizing block structures in the coherence matrix, showing great robustness against non-stationary noise sources that generally affect the performance of the standard approaches. Moreover, the F-PCD is characterized by an accelerated inference process, ensuring efficiency without substantial performance loss. The F-PCD algorithm can be applied to different scenarios, for example, where DEM changes happen, e.g., mining sites, volcano eruptions, and earthquakes. For this reason, an example of the F-PCD application on an active open-pit mining site is presented to validate its effectiveness. Moreover, its generalization capability is demonstrated by a multi frequency-geometry analysis conducted on the same mining site. Finally, fully exploiting the F-PCD outcomes contributes to a broader understanding of temporal changes in SAR data and introduces new perspectives for interpreting InSAR datasets.

Interferometric Synthetic Aperture Radar (InSAR)-derived surface displacement time series enable a wide range of applications from urban structural monitoring to geohazard assessment. With systematic data acquisitions becoming the new norm for SAR missions, millions of time series are continuously generated. Machine Learning provides a framework for the efficient mining of such big data. Here, we focus on unsupervised mining of the data via clustering the similar temporal patterns and data-driven displacement signal reconstruction from the InSAR time series. We propose a deep Long Short Term Memory (LSTM) autoencoder model which can exploit temporal relations in contrast to the commonly used shallow learning methods, such as Uniform Manifold Approximation and Projection (UMAP). We also modify the loss function to allow the quantification of uncertainties in the time series data. The two approaches are applied to the Lazufre Volcanic Complex located at the central volcanic zone of the Andes and thereby compared.

We reveal the presence of a peculiar physical signal which compromises the accuracy of Earth surface deformation estimates for distributed scatterers [1]. The observed signal is short-lived and decays with the temporal baseline; however, it is distinct from the stochastic noise attributed to temporal decorrelation. To indicate its nature, this physical effect is referred to as fading signal. Designing a simple approach in the evaluation of distributed scatterers, we reveal a prominent bias in the deformation velocity maps. The bias is the result of propagation of small phase error through the time series. We further discuss the role of the phase estimation algorithms in significant reduction of the bias and put forward the idea of a unified analysis-ready InSAR product for achieving high-precision deformation monitoring.

This paper deals with the analysis of InSAR performance for large-scale deformation measurement. The study evaluates the use of models, especially numerical weather prediction reanalysis, to mitigate disturbances in SAR interferograms. The impact of such corrections is evaluated and, using GNSS measurements, the predicted error bars are validated on a large Sentinel-1 dataset.

This article investigates the presence of a new interferometric signal in multilooked synthetic aperture radar (SAR) interferograms that cannot be attributed to the atmospheric or Earth-surface topography changes. The observed signal is short-lived and decays with the temporal baseline; however, it is distinct from the stochastic noise attributed to temporal decorrelation. The presence of such afading signalintroduces a systematic phase component, particularly in short temporal baseline interferograms. If unattended, it biases the estimation of Earth surface deformation from SAR time series. Here, the contribution of the mentioned phase component is quantitatively assessed. The biasing impact on the deformation-signal retrieval is further evaluated. A quality measure is introduced to allow the prediction of the associated error with the fading signals. Moreover, a practical solution for the mitigation of this physical signal is discussed; special attention is paid to the efficient processing of Big Data from modern SAR missions such as Sentinel-1 and NISAR. Adopting the proposed solution, the deformation bias is shown to decrease significantly. Based on these analyses, we put forward our recommendations for efficient and accurate deformation-signal retrieval from large stacks of multilooked interferograms.

This article deals with the analysis of InSAR performance for large-scale deformation measurement. The study evaluates the use of models, especially numerical weather prediction reanalysis, to mitigate disturbances in SAR interferograms. The impact of such corrections is evaluated by analyzing short-time baseline phase variograms in order to derive a lower bound for the interferometric accuracy, especially at large distances. The variance is then propagated from single interferograms to deformation rates. Finally, using GNSS measurements, the predicted error bars are validated on a large Sentinel-1 data set.

The work deals with the identification and the characterization of the grounding zone area using InSAR data. The idea is to point towards a methodology that minimizes the role of the operator and provides results with performance that can be mathematically described using input parameters. The approach uses the information of the interferometric phase gradient to follow the path of the grounding zone and fit them using a physical model that describes the ice bending. The approach is tested on more than 300 km grounding zone comparing also the results with existing products.

The capacity of SAR interferometry to measure surface deformation with accuracy of 1 mm/year or better are well known. However this is typically limited to relative motion at short distance. Thanks to several advances in SAR sensor quality, data availability, orbit determination, processing, and calibration of atmospheric delays it is now possible to achieve that accuracy even across large distances of hundreds of kilometers.In this paper we revise the main contributions to the large scale error, considering available mitigation techniques. We provide a first validation for a processing based on Sentinel-1 data, by comparing our results with GNSS stations.For future SAR's operating at lower frequencies, it is vital to consider ionospheric corrections and likely also the influence of moisture variations in natural scatterers. The choice of processing algorithms, though typically not discussed, can also have a significant effect on the quality of the result.

This letter investigates the estimation of the slip distribution of a seismic event using the information provided by interferometric phase gradients. Even if the technique is expected to be suboptimal when compared with an estimation using the unwrapped interferometric phase, such an approach would permit to avoid the solution of phase ambiguities also including the parts of the interferogram that could not be reached by the phase unwrapping otherwise. This specifically addresses the cases where the motion gradients are so strong that particular areas have to be masked out due to unwrapping errors. Aim of this letter is to propose a possible way to include such areas modeling the motion with phase gradients. The rationale of this letter relies on the description of the coseismic motion given by the Okada model that provides both the 3-D surface displacement and the gradient tensor information. Based on the latter, this letter defines an inversion strategy that uses the information extracted by the phase gradients, hence avoiding phase unwrapping. This technique is tested on real test sites and compared with the results obtained using the absolute phase.

This paper deals with precise absolute geolocation of point targets by means of a pair of high-resolution synthetic aperture radar (SAR) acquisitions, acquired from a satellite. Even though a single SAR image is a 2-D projection of the backscatter, some 3-D information can be extracted from a defocussing analysis, depending on the resolution, thanks to orbital curvature. A second acquisition, observing the same scene under a different look angle, adds stereogrammetric capability and can achieve geolocation accuracy at decimeter level. However, for the stereogrammetric analysis to work, it is necessary to match targets correctly in the two images. This task is particularly difficult if it has to be automatic and targets are dense. Unfortunately, the defocussing-based geolocation is not sufficient for reliable target matching: the limiting factor is the unknown tropospheric delay that can cause geolocation errors of several meters in the elevation direction. However, observing that the tropospheric phase screen displays a low-pass character, this paper shows how to identify statistically the local atmospheric disturbances, therefore dramatically improving the score of successful matching. All steps involved exploit peculiar radar image characteristics and, thanks to this, avoid generic point cloud matching algorithms. The proposed algorithm is shown at work on a pair of TerraSAR-X staring spotlight images.

Tandem-L is a proposal for an innovative L-band SAR mission for the systematic observation of dynamic processes on the Earth's surface. The mission concept is based on two SAR satellites flying in close formation featuring latest digital beamforming techniques in combination with a large deployable reflector for increasing the swath width and imaging resolution. This enables innovative operation modes such as polarimetric SAR tomography for determining the vertical structure of vegetation and ice. With novel imaging and processing techniques and the vast recording capacity of up to 8 Terabyte/day, it will provide vital information for solving pressing scientific questions in the biosphere, geosphere, cryosphere, and hydrosphere and will make an essential contribution for a better understanding of the Earth system and its dynamics. This paper gives an overview of the Tandem-L mission project which has successfully passed the intermediate system requirements review of Phase B1.

This letter aims to discuss a general framework that allows the direct interpretation of the wrapped differential synthetic aperture radar interferometry phase in terms of surface strain S and rotation R components. The methodology is demonstrated showing the estimation of strain and rotation components of a glacier flow using three TerraSAR-X interferometric geometries (ascending right-looking, descending right-looking, and descending left-looking). Finally, since the left-looking geometry can be difficult to obtain on a regular basis, the surface parallel flow assumption is extended to the phase gradients inversion in order to reduce the amount of necessary geometries from three to two.

This paper presents a novel approach to determine 3D absolute geolocation of point targets using just two long-aperture SAR acquisitions. Moreover, the Zenith Path Delay of both acquisitions is obtained. First results presented here show that the absolute positioning accuracy reaches the sub-meter level.

Tandem-L is a highly innovative SAR satellite mission for the global observation of dynamic processes on the Earth's surface with hitherto unknown quality and resolution. Thanks to its novel imaging techniques and its unprecedented acquisition capacity, Tandem-L will deliver urgently needed information for the solution of pressing scientific questions in the areas of the biosphere, geosphere, cryosphere and hydrosphere. The feasibility of Tandem-L has been analyzed and confirmed in the scope of a phase A study, which has been conducted in close cooperation between the German Aerospace Center (DLR) and the German space industry. This paper provides an overview of the Tandem-L mission concept and summarizes the actual development status.

A conventional interferometric synthetic aperture radar (SAR) system provides 1-D line-of-sight motion measurements from repeat-pass observations. Two-dimensional motions may be measured by combining two observations from ascending and descending geometries. The third motion component may be retrieved by adding a third geometry and/or by integrating along-track measurements although with much reduced precision compared to the other two components. Several options exist to improve the accuracy of retrieving the third motion component, such as combining left- and right-looking observations or exploiting recently proposed innovative SAR acquisition modes (BiDiSAR and SuperSAR). These options are, however, challenging for future SAR systems based on large reflector antennae, due to lack of capability to electronic beam steering or frequent toggle between left- and right-looking modes. Therefore, in this letter, we assess and compare the realistic acquisition scenarios for a reflector-based SAR in an attempt to optimize the achievable 3-D precision. Investigating the squinted SAR geometry as one of the feasible scenarios, we show that a squint of 13.5° will yield comparable performance to the left-looking acquisition, while further squinting outperforms this or other feasible configurations. As an optimum configuration for 3-D retrieval, the squinted acquisition is further elaborated: the different acquisition plans considering a constellation of two satellites as well as the challenges for data processing are addressed.

The differential ionospheric path delay is a major error source in L-band interferograms. It is superimposed to topography and ground deformation signals, hindering the measurement of geophysical processes. In this paper, we proceed toward the realization of an operational processor to compensate the ionospheric effects in interferograms. The processor should be robust and accurate to meet the scientific requirements for the measurement of geophysical processes, and it should be applicable on a global scale. An implementation of the split-spectrum method, which will be one element of the processor, is presented in detail, and its performance is analyzed. The method is based on the dispersive nature of the ionosphere and separates the ionospheric component of the interferometric phase from the nondispersive component related to topography, ground motion, and tropospheric path delay. We tested the method using various Advanced Land Observing Satellite Phased-Array type L-band synthetic aperture radar interferometric pairs with different characteristics: high to low coherence, moving and nonmoving terrains, with and without topography, and different ionosphere states. Ionospheric errors of almost 1 m have been corrected to a centimeter or a millimeter level. The results show how the method is able to systematically compensate the ionospheric phase in interferograms, with the expected accuracy, and can therefore be a valid element of the operational processor.

Interferometric synthetic aperture radar (InSAR) measurements are merely sensitive to the deformation along the Line of Sight (LOS) direction of the sensor. To improve the geometrical sensitivity and retrieve the three-dimensional deformation, the integration of InSAR from non-coplanar acquisitions as well as fusion with resolution-scale SAR image shift measurements has become a standard approach. Using different statistical measures, we assess and compare the influence of different image acquisition strategies as well as data fusion on the performance of InSAR in 3D deformation retrieval. Integrating nominal InSAR acquisitions, i.e. a set of measurements from ascending and descending tracks acquired from right-looking geometry, a strong correlation between the retrieved 3D parameters in the local vertical-north plane is observable. This correlation is sought to be decreased by non-nominal acquisitions; i.e. left-looking or squinted observations. These acquisition strategies are discussed for consideration in the future L-band mission Tandem-L.

The work presented exploits the long Synthetic Aperture Radar (SAR) of a single TerraSAR-X Staring Spotlight (ST) acquisition to derive absolute heights. Here, the slight azimuth defocussing effect due to height mismatch between true height and the height assumed in SAR focusing is analyzed. The impact is almost negligible for most of acquisition modes. In contrast, spaceborne modes with very long aperture, such as TerraSAR-X ST acquisition mode, present sensibility that can be used to retrieve absolute heights. The accuracy depends on incidence angle, orbit type and mainly on Signal to Clutter Ratio (SCR). A result over real data is presented to demonstrate that absolute heights can be retrieved with an accuracy of few meters using a single TerraSAR-X ST acquisition.

The work presented in this letter exploits the long synthetic aperture radar (SAR) of a single TerraSAR-X Staring Spotlight (ST) acquisition to derive absolute heights. Here, the slight azimuth defocusing effect due to height mismatch between the true height and the height assumed in SAR focusing is analyzed. The impact is almost negligible for most of acquisition modes. In contrast, spaceborne modes with very long aperture, such as the TerraSAR-X ST acquisition mode, present sensibility that can be used to retrieve absolute heights. The accuracy depends on incidence angle, orbit type, and mainly on signal-to-clutter ratio. Two different results are presented to demonstrate that absolute heights can be retrieved with accuracy of few meters using a single TerraSAR-X ST acquisition.

Ionosphere irregularities along the synthetic aperture generate shifts and blurring that cause decorrelation. In this paper it is shown how, by partially focusing SAR images to the height of the ionosphere, it is possible to reduce the ionospheric azimuth effects and increase the coherence. This permits, even in case of turbulent ionosphere, to obtain better accuracies when separating the deformations phase from the ionospheric phase using the delta-k split-band interferometry method.

Spectral estimation is considered in the paper as an additional instrument towards a better understanding of the physical phenomena behind the layover scattering decomposition. A super-resolution technique is employed to derive the fringe frequencies characterizing the layover portion. Due to the limited estimation support, only the dominant frequency is found to be reliable information. The non-linear relationship with slopes is employed to derive a principal slope map. A bistatic interferometric scenario is tested. It is found that for the majority of the detections the facade contribution is the prevailing one due to the presence of targets with a high backscattered signal return at the vertical slope. The number of layover contributors is assessed prior to the spectral estimation. It has been estimated that the signal return is dominated by a single contribution for the majority of the layovers.

There is a need for scattering models that link quantitatively synthetic aperture radar (SAR) interferometric observables to soil moisture. In this paper, we propose a model based on plane waves and the Born approximation, deriving first the vertical complex wavenumbers in the soil as a function of geometrical and dielectric properties and successively the complex interferometric coherences. It is observed that soil moisture behaves on the phase in a similar way as tomography does, breaking the phase consistency in triplets of interferograms. The proposed model is validated with L-band airborne SAR data; preliminary inversion results based on interferogram triplets and coherence magnitudes are presented.

In recent years, Persistent Scatterer Interferometry (PSI) [1], [2] has been widely used for scientific applications and has developed into an operational and commercially rewarding remote sensing technology. Now, ESA's upcoming Sentinel-1 mission allows a continuous, repeated without gap and global mapping of the Earth's surface based on Terrain Observation by Progressive Scans (TOPS). The idea at ESA is to also extend PSI processing to such large coverages, mapping countries and continents. This will support users in subsidence monitoring of volcanic and seismogenic areas, of costal lowland, of landslides in mountainous areas and of mining and ground water regulation on a small scale. For this reason, a wide area product (WAP) for PSI monitoring has been developed at DLR. The paper has three specific objectives. The first objective is to list and describe the updated algorithms. In this context, we illustrate the improvement with respective processing examples. Our second objective is to explain the characteristics of the WAP. The last objective is to provide WAP processing results for seismogenic areas. The test cases of Greece and of Turkey demonstrate WAPs potential, its applicability and use.

The aim of this work is to show two possible applications for Alternating Bistatic TanDEM-X data. The first application is a tomographic processing while the second is a 2D ship velocity estimation. Results over real data are presented for both applications.

A procedure to simulate the effects of the ionosphere on Synthetic Aperture Radar (SAR) images is presented. The propagation delay errors induced by the ionosphere have to be compensated to millimeter level in order to meet the scientific requirements for an L-band mission dedicated to deformation measurements, which are summarized in [1]. The simulator presented in this paper can be used to study the effects of an arbitrary ionospheric state on SAR images and to generate disrupted raw and focused images starting from ionospherefree real SAR images and use them to validate ionosphere estimation methods.

Persistent Scatterer Interferometry (PSI) is a well established SAR technique to monitor the Earth's surface displacements with millimeter accuracy [1]. In the framework of the ESA's Terrafirma project, a PSI based Wide Area Product (WAP) is developed and demonstrated by DLR [2]. The PSI technique is well applicable for the processing of urban areas, because many persistent scatterers (PSs) are available due to man-made features. However, the mapping of areas with an extension of more hundreds of kilometres is the objective of the WAP. Thus rural and mountainous regions also need to be handled by the processor. In such areas, the distribution of PSs is very inhomogeneous and, typically, the point density is very low. In this paper we address two technical difficulties in the development of the WAP. Firstly, we report on the determination and estimation of a reliable reference network of the processing area in order to estimate and compensate the atmospheric phase screen (APS). Secondly, the mosaiking of independent reference networks into a global consistent data set is described. As basic method, the conventional least squares adjustment is used on four hierarchic levels. Consequently, the error propagation through all these levels can be analyzed and evaluated.

One of the main limitations that affects DInSAR measures is the degradation of the phase quality due the changes occurring in the radar resolution cell. Nevertheless it has been observed that in some areas the phase decorrelation process saturates to an amount of coherence normally quite small but not negligible. This behavior can not always be explained with a point target model as far as the interested areas sometimes are not characterized by man-made features and the characteristics change radically with the wavelength. In this paper a method to identify this areas from a relative small amount of SAR acquisitions is proposed. The the results are compared with the residual coherence measured in a test sites acquired in X, C and L band.

Since long it is known that SAR interferometric observables are influenced by soil moisture variations, however there is a lack of scattering models that link interferometric observables and variations of the dielectric properties. In this work we propose a model based on plane waves and Born approximation, deriving first the vertical wavenumbers in the medium as a function of geometrical and dielectric properties and successively the interferometric coherences. It is observed that soil moisture behaves on the phase in a similar way as tomography does, breaking the phase consistency in triplets of interferograms. This property, along with coherence magnitudes, is exploited in an attempt at moisture inversion on real data.

The persistent scatterer interferometry (PSI) is a well established radar technique to monitor the Earth's displacements with millimetre accuracy. It uses men made features typically made of metal (persistent scatterers) given by chance to form interferometric phase time series spanning many years. Actually, its application is limited to urban areas only because of the high density of usable persistent scatters. In the course of ESA's Terrafirma project, a wide area product (WAP) PSI mapping is demonstrated by DLR. Subject is to map countries and continents based on the PSI technique. The WAP is foreseen to be a standard level 1 product for the future Sentinel-1 mission with its TOPS mode acquisition scenario. However, many technical problems need to be solved in order to extend the PSI mapping area from urban areas to rural and even mountainous regions. This paper reports on the wide area product, the technical challenges and their algorithmic solutions. Also, WAP example data are presented.

For spaceborne SAR (Synthetic Aperture Radar) systems, the dispersive effects of the ionosphere on the propagation of the SAR signal can be a significant source of phase error. While at X-band frequencies the effects are small, current and future P-, Land C-band systems would benefit from ionospheric compensation to avoid errors in topographic retrieval. In this paper the focus is on the effects of the ionosphere on repeat-pass SAR interferometry from Pthrough X-bands and methods for their estimation which are demonstrated on L-band ALOS-PALSAR acquisitions.

Efficient estimation of the interferometric phase and complex correlation is fundamental for the full exploitation of SAR Interferometry capabilities [1]. Particularly when combining interferometric measures arising both from distributed and concentrated point targets, the interferometric phase has to be correctly extracted in order to preserve its physical meaning and respect the homogeneity hypothesis that we assume when performing a coherent averaging [2]. Recently, an amplitude-based algorithm for the adaptive multilooking of InSAR stacks was proposed [3], [4] where it was shown that a comparison of the backscatter amplitude statistics is a suitable way to adaptively group and average the pixels in order to preserve the phase signatures of natural structures in the observed area.

Efficient estimation of the interferometric phase and complex correlation is fundamental for the full exploitation of interferometric synthetic aperture radar (InSAR) capabilities. Particularly, when combining interferometric measures arising both from distributed and concentrated targets, the interferometric phase has to be correctly extracted in order to preserve its physical meaning. Recently, an amplitude-based algorithm for the adaptive multilooking of InSAR stacks was proposed where it was shown that a comparison of the backscatter amplitude statistics is a suitable way to adaptively group and average the pixels in order to preserve the phase signatures of natural structures in the observed area. In this letter, different methods to compare amplitude statistics will be presented, compared through simulation and applied to real data. Based on these, recommendations are made concerning which method to use in practice.

For spaceborne SAR (Synthetic Aperture Radar) systems, the dispersive effects of the ionosphere on the propagation of the SAR signal can be a significant source of phase error. While at X-band frequencies the effects are small, current and future L-band systems would benefit from ionospheric compensation. We consider two ways to estimate the ionospheric delay in SAR signals and evaluate them on L-band ALOS-PALSAR acquisitions.

The application of the Permanent Scatterers (PS) Technique in multi-temporal data-sets, namely the identification and exploitation of sparse coherent targets, has shown that it is possible to estimate and remove interferometric phase components due to atmospheric effects and orbital fringes. So far, the application of the PS technique has been focused on the extraction of the motion field of the area of interest. However, it is also known that PS relative elevations can be estimated with sub-meter precision while smooth errors can be removed using a coarse resolution DEM or the data of the Shuttle Radar Topography Mission (SRTM). In this paper, we describe a new approach combining the PS Technique and standard interferometry to improve the quality of lnSAR DEM's. ERS Tandem interferograms are exploited to increase the number of coherent pixels, while atmospheric effects are estimated and subtracted by means of the sparse PS grid. Prior information and PS elevation are used to reduce the probability of phase-unwrapping errors. Preliminary results are reported and the key-factors for its successful application (e.g. the number of Tandem acquisitions available, PS density) are discussed.