Landslides can be highly disruptive, leading to damage of infrastructure, blockage and diversion of water courses, disruption of mine operations, and in the worst cases loss of life. Critical failure of a landslide can be preceded by precursor deformation (i.e. slow creep of the slope), the detection of which can allow for the adoption of an appropriate mitigation strategy. Conventional methods for detecting such precursor deformation (such as GPS, inclinometers, piezometers, levelling, terrestrial LiDAR and photogrammetry), are expensive, cover a limited area and are ill-suited to remote or dangerous areas. Satellite remote sensing techniques, utilizing open access data, can enable the monitoring of landslides at lower cost, while maintaining high accuracy.
Here we show how InSAR applied to Sentinel-1 synthetic aperture radar data can be used to monitor precursor deformation associated with the July 2018 Fagraskógarfjall landslide (Eastern Iceland). Up to 2 m of displacement occurred in the year leading up to the landslide (2018), and a lower magnitude of displacement (up to 40 cm/yr) occurred in 2015 and 2016. The displacement shows seasonal trends, moving faster at wetter times of year, or after periods of high rainfall, and slower in drier months. We also show the limitations of the method, with displacement in the four months prior to the landslide being of too high a magnitude for InSAR analysis using Sentinel-1 data. Despite the high magnitude of displacement, it was still possible to map the deformation extent. Finally, we are able to determine the post-collapse stability, further aiding hazard assessments of the slope.
The results of this study serve to highlight the effectiveness of InSAR for the monitoring of landslide hazards and slope stability. With the abundance of open-access data, long term monitoring of landslides and potentially unstable slopes is increasingly feasible, with applications to hazard monitoring, engineering, and open pit mining.

The SAR (Synthetic Aperture Radar) imagery are widely used to monitor mainly vertical displacements impacting the Earth surface and human infrastructures. The main remote sensing technique used to extract sub-centimeter information from SAR imagery is the so-called Differential SAR Interferometry (DInSAR). It is based on the differences of phase information and it is well known that this technique may suffer for lack of coherence among the considered image stack in case of both temporal/seasonal variation of the backscatter (i.e., in the vegetated areas) and large and sudden displacements. Moreover, an additional limitation is related to the orientation of the displacements, since only their component along the line-of-sight can be retrieved; in general, it is difficult to estimate 3D (both horizontal and vertical) displacements.

On the other hand, new SAR satellite, such as COSMO-SkyMed and TerraSAR-X, can acquire imagery with high amplitude resolutions (up to few decimeters) and, thanks to the onboard dual frequency GPS receivers, the satellite orbits are estimated with an accuracy of few centimeters. Several experiments by different research groups prove that TerraSAR-X imagery offers the capability to achieve, in a global reference frame, 3D positioning accuracies in the decimeter range and even better. This technique, called Imaging Geodesy, exploits the slant-range measurements, coming from the amplitude information, taking into account all the involved geophysical phenomena. Moreover, thanks to the high amplitude resolution, the Offset Tracking technique, widely used for glacier surface velocity monitoring, can represent an excellent complementary approach, able to overcome the limitation on the displacement orientation, albeit at a lower spatial accuracy and only for the horizontal components.

In this work, we propose the integration of the Offset Tracking and Imaging Geodesy techniques for full exploitation of SAR amplitude information in case of fast displacement monitoring. We investigated the applicability of the developed method using COSMO-SkyMed data to the Corvara area near to Bolzano (Northern Italy), where a landslide characterized by relevant yearly displacements (up to several meters) is monitored for some years by GPS permanent stations and periodic surveys. A new methodology to exploit the slant-range measurements using only the amplitude information was developed, and the geometrical consistency of the method was verified with a simulation of displacements applied to one of the considered stable Amplitude Permanent Scatter (APS), reaching an accuracy of few centimeters. Specifically, a software based on the Opticks free and open source platform and the OpenCV Library was used to apply the technique and to estimate the displacements. Finally, the same methodology was used to evaluate corner reflector displacements, and the results were compared to the available GPS reference data. The preliminary results indicate that COSMO-SkyMed strip map imagery appears able to guarantee a displacement monitoring with an accuracy of few centimeters using only the amplitude data, provided few (at least one) stable APS’s are available outside the monitored area, to correct the residual biases. These results have to be combined appropriately with those (processing ongoing) coming from the Offset Tracking technique for complete exploitation of the potentialities of the two methodologies. The overall results will be presented and discussed.

Hazards involving ground deformation such as landslides, mudflows, avalanches, volcanic eruptions, ground subsidence and collapse, and infrastructure failures have been normal occurrences globally, often leading to significant human and economic losses. Such scenarios require effective monitoring of movements, which can provide insight into mechanisms and triggering factors of hazardous events or act as the basis for mitigating risks, e.g. understanding maintenance and remedial measures, early warning and rapid decision-making for countermeasures or evacuation. Ground-based synthetic aperture radar (GBSAR), together with interferometry, represents a powerful tool for deformation monitoring. Due to its inherent flexibility, a typical GBSAR system can be performed in either continuous or discontinuous mode for different scenarios, depending on the rates of change or the practical environment for instrument deployment. The goal of this research is to develop a framework to effectively utilise GBSAR for deformation monitoring in all modes, with the emphasis on accuracy, robustness, and real-time capability.

To achieve this goal, advanced interferometric SAR (InSAR) processing algorithms have been proposed to address existing issues in conventional interferometry for GBSAR deformation monitoring. The proposed InSAR algorithms include: (i) a new non-local method termed as “MIAS” (multi-temporal interferometry based on amplitude similarity) for the accurate estimation of coherence and interferometric phase, which can largely mitigate the coherence estimation bias in conventional “boxcar” estimators and avoid overestimating the decorrelated area without the cost of resolution; (ii) a new criterion for the selection of coherent pixels from a redundant network of interferograms, which enables the selection of not only qualified partially coherent pixels but also all persistent scatterers and makes the most of redundant observations, thus, maximizing the density of selected pixels and optimizing the reliability of time series analysis; and (iii) a new model for the correction of atmospheric and geometric (repositioning) errors jointly.

On the basis of the proposed algorithms, two complete interferometric processing chains have been developed: a novel chain for real-time continuous GBSAR deformation monitoring (RT-GBSAR) and a new chain for discontinuous GBSAR deformation monitoring. The challenges in processing continuous GBSAR data have been properly resolved by the proposed RT-GBSAR chain with three unique merits: (i) low requirement of computational random-access memory and capability of processing infinite images; (ii) insights into the evolution of surface movements through temporarily-coherent pixels; and (iii) real-time processing of continuous GBSAR data. The proposed discontinuous GBSAR chain integrates additional automatic coregistration of images and correction of repositioning errors between different data collection campaigns, making it a complementary tool to the RT-GBSAR for deformation monitoring.

The proposed chains and algorithms have been fully implemented and integrated into a bespoke GBSAR data processing software package. A number of deformation monitoring applications have been completed using this package, including three continuous (i.e., a sand dune, a metro bridge, and a coastal cliff) and one discontinuous (a hillside). Significant deformation signals are detected from the three continuous applications and no deformation from the discontinuous application. The achieved results are justified quantitatively via a defined precision indicator for the time series estimation and validated qualitatively via a priori knowledge of these observing sites. Successful applications have demonstrated the effectiveness and feasibility of the presented techniques for high-accuracy measurement of surface deformation. We believe the developed software package will greatly facilitate all deformation monitoring scenarios to which GBSAR is potentially suited.

Detecting slope instabilities in abandoned mining areas is critical to reduce their possible impact. This work presents a methodology to accurately identify and measure mining-induced deformation areas using Sentinel-1 TOPS data. The method consists of spatially analyzing the measured deformation rates based on a set of GIS-mapped features, e.g. tailing dumps, wells, faults, etc. It is applied in La Union (Murcia, SE Spain), a historic metal mining area. In contrast to previous studies conducted in La Union, mainly aimed at mapping and monitoring ground subsidence using ERS and ENVISAT data, this paper focuses on assessing slope instability using Sentinel-1A data, which provides higher coherence and temporal sampling (12-day revisit time). Differential Interferometry (DInSAR) analysis is performed on a stack of 37 C-band Sentinel-1A images acquired from 30 October 2014 to 3 July 2016. Thanks to the enhanced capabilities of Sentinel-1 over non-vegetated areas, due to its high spatial and temporal resolution, a higher Persistent Scatterers (PSs) density than that provided by ERS and ENVISAT is obtained. As a second step, statistical analysis of the obtained line-of-sight (LOS) displacements is carried out for 1717 GIS-mapped features. Additionally, identical analysis is performed for LOS displacements projected along the steepest slope, after applying the corresponding corrections. The results show that maximum LOS deformation rates of 8 cm/year and maximum down-slope deformation rates of 7,5 cm/year, affects an area near the “Tomasa” tailing dump where displacements are mostly horizontal, consistent with a planar-landslide behavior. Finally, a time series analysis is carried out to study the temporal evolution of both the LOS and the down-slope displacements. For the validation of the satellite data, the deformation rates retrieved with DInSAR are compared with Differential GPS (DGPS) data acquired during field campaigns. The whole methodology is then demonstrated and verified by comparing the results with an Active Deformation Areas (ADAs) dataset obtained by means of a non-commercial C++ software called ADAFinder (Barra et al. 2017), that automatically delineates areas with similar deformation behavior by discriminating the more reliable deforming areas making use of the deformation rates and the time series. The proposed methodology can be applied to mining-related slope instabilities worldwide thanks to the global availability of Sentinel-1 data. The aim of the method is to simplify the InSAR-derived information in order to be exploited by non-expert final users.

Being the Geological Survey of Austria one of the key role player on creation of the Austria’s Centre for Geo-Monitoring is more and more involved in national and local project concerning the use of Copernicus data for geo-hazard prevention and monitoring. One of the major focus is the use of innovative DInSAR MTI methods applied to landslide and subsidence monitoring.

The data used for this work were obtained for free at the COMET LICS Sentinel-1 portal where is possible to download a number of already processed Differential Interferograms using Sentinel -1.
In this web site is possible to look for ascending (ideal for east facing slopes) and descending (ideal for west facing slopes) in mountainous areas all around the world. Some more missing interferograms were created using the Geohazard TEP infrastructure free available, being our institution accepted to use the service as an early adopter user.
A workflow based on MATLAB scripts was applied to the COMET LICS data and Geohazard TEP in order to assign to each inteferogram a fix reference datum and the tropospheric correction was performed using GACOS – a free service available online. The result of this approach was compared with a MTI product obtained by proceeding a series of SLC Sentinel-1 data directly downloaded from the Copernicus web site. The open source software infrastructure used for the elaboration combines SNAP (python based), StamPS and TRAIN with the adoption of the GACOS corrections.

The application of these two workflows was applied to a study case involving a sudden ground deformation occurred between 2016-2018 in Austria. In this work, the two methods will be described, correlated with hydrologic data and field observation. Further analysis will involve the study of time-series data in order to clarify the ongoing process and to identify the most exposed areas to damages.

Tehran, the capital of Iran, which hosts more than 8 million people, has been experiencing rapid population growth and extensive development in the past few decades. The increasing need for groundwater in this area has been met by extensive groundwater extraction which caused compaction of fine-grained sediments in the sub-surface, showing itself as land subsidence at the surface.
We use an extensive collection of SAR data to study land subsidence and its variations in Tehran. The data is acquired over a 15-year period between 2003 and 2017 by Envisat, ALOS, TerraSAR-X, and Sentinel-1 A/B satellites. We estimate the spatial and temporal behavior of surface displacement using time-series of interferograms.
The results are in good agreement with each other and also with independent, precise leveling measurement in the area. They show that the subsidence mainly occurs in agricultural areas in the southwest of Tehran. The maximum rate of subsidence has slightly decreased from more than 25 cm/yr in 2003 to around 20 cm/yr in 2017. However, our results reveal that the subsidence is spreading towards the urban area of Tehran and posing a significant hazard to the infrastructure and urban development in this region.
Combining the time series of displacement derived from independent SAR datasets, allows us to obtain a long time series of displacement which reveals two main components in the data that are linked to groundwater level changes. The subsidence trend in the time-series is a result of long-term extraction of groundwater and in agreement with the groundwater level measurements. The short-term variations of land subsidence are also following the seasonal variations of groundwater due to discharge/recharge of groundwater.
In the past three decades, no significant recovery in groundwater has been observed. Therefore, most of the long-term subsidence is expected to be inelastic and non-recoverable causing earth fissures, cracks etc. On the other hand, the seasonal ups and downs of subsidence, linked to seasonal variations of groundwater level, is the only elastic deformation we observe in the time series. Comparing the magnitude of elastic to inelastic deformation show that non-recoverable part of deformation is dominant. Therefore, effective groundwater management should be implemented to reduce the effect of non-recoverable compaction of the aquifer.