The European Ground Motion Service (EGMS) is the most recent addition to the product portfolio of the Copernicus Land Monitoring Service. The EGMS is funded by the European Commission in the frame of the Copernicus Programme and it is implemented under the responsibility of the European Environment Agency. The Service provides consistent, regular, standardized, harmonised and reliable information regarding natural and anthropogenic ground motion phenomena over the Copernicus Participating States and across national borders, with millimetre accuracy. The EGMS is based on the multi-temporal interferometric analysis of Sentinel-1 radar images at full resolution. Global navigation satellite systems (GNSS) data are used as calibration of the interferometric measurements. The EGMS distributes three levels of products: (i) basic, i.e. line of sight (LOS) velocity maps in ascending and descending orbits referred to a local reference point; (ii) calibrated, i.e. LOS velocity maps calibrated with a geodetic reference network so that measurements are no longer relative to a local reference point and (iii) ortho, i.e. components of motion (horizontal and vertical) anchored to the reference geodetic network. Data are available and accessible for all and for free through a dedicated viewer and download interface.
EGMS is an unprecedented opportunity to study geohazards and human-induced deformation over Europe, such as slow-moving landslides, subsidence due to groundwater exploitation or underground mining activities, volcanic unrests, and many more. These data can serve a wide spectrum of users interested in ground motion data for geohazards mapping and monitoring. This presentation will offer a first look at the products distributed by EGMS through relevant case studies in different environmental contexts of Europe. Landslides along the Alpine Arc and in the rocky slopes of Scandinavian fjords, subsidence in alluvial plains in Spain and Italy, mining-induced deformation in Poland and Germany are some of the examples that will be presented. The interferometric data will be analyzed to provide an interpretation under geoscientific aspects of the measured ground motion and to show how the EGMS products can be successfully used for geohazards-related studies.

The Lisbon Metropolitan Area, with an extension of 4,390 km2 and located in the centre of Portugal, is well-known for its significant landslide and subsidence phenomena, among other geohazards. Also, the LMA is comprehended by 18 urban and rural municipalities, and whose population is over 2.8 million inhabitants. The main aim of this study is detecting and analysing ground deformations associated to these processes by means of A-DInSAR techniques. For this, the following methodology was : i) selection and processing of 48 SAR images in ascending trajectory, provided by Sentinel-1A between January 2018 and April 2020, by means of P-SBAS technique implemented in the European Space Agency (ESA)´s GEP service; ii) obtaining Line-of-Sight (LOS) mean deformation velocity map (mm year-1) and deformation time series (mm); and application of ADA (Active Deformation Areas) post-processing procedure to detect local areas with outstanding deformations; iii) validation and interpretation of A-DInSAR results and identified ADA through field surveying and geological settings. The results show a LOS velocity (VLOS) ranging between -38.0 and 18.9 mm year-1, and an accumulated ground displacement between -74.7 and 40.1 mm. Moreover, 592 ADA were identified, which 492 ADA were selected and analysed. It has been possible to differentiate local sectors with recent deformation related to landslide incidence, with maximum VLOS of 25.5 mm year-1 and urban-industrial subsidence due to aquifers over-exploitation, with maximum VLOS of -38.0 mm year-1. This study represents an important contribution to improve the knowledge about ground motions in the Lisbon Metropolitan Area. In addition, this work corroborates the reliability and usefulness of the GEP service and the ADA methodology as powerful tools to study geological hazards at both regional and local scales. Future research will involve improving the interpretation of the A-DInSAR dataset, by means of re-processing in descending trajectory, and the estimation of the VSLOPE from VLOS measurements obtained in this work.

The Ostrava region has been the heartland of black coal mining in the Czech Republic for centuries. Though coal production has plummeted over the last decades, its geotechnical impacts represented by large scale ground subsidence and related risks will remain affecting the landscape in the future. Recultivations and development zones planning in the area need to consider evidence of both existing and future patterns of the geohazard.
In the frame of the CURE project, the Gisat team implemented retrospective ground motion mapping from time series of Sentinel-1 imagery using persistent scatterers interferometric technique. Custom algorithms for correction of abundant phase unwrapping errors (often associated with non-linear displacement trends in the time domain) have been developed and implemented into automated post-processing workflows. A custom algorithm to recognize non-linear motion was used to distinguish low-coherent points due to non-linear displacement from noise-only points, allowing to also monitor areas subjected to inconstant deformation trend dynamics. Spatial clusters of temporally specific patterns such as motion acceleration or deceleration have been detected by the mining interferometric time series using the algorithm. The multi-pass constellation of input SAR data allowed decomposition of the motion vector in radar geometry to both vertical motion fields. Motions with comprehensive horizontal components have been strongly affecting some localities and increasing risks related to surface angular strains, which is a crucial factor to be considered for new buildings constructions in the undermined areas. In addition, the implemented workflow provides automated tools to derive vertical and horizontal strains related to detected ground deformations as a baseline for quantification of surface faulting risks for dwellings and infrastructure.
Results show a large extent and severity of ground deformation phenomena in the region. Subsidence affects both industrial fields and abandoned mines zones, which are supposed to undergo recultivation in near future, and also infrastructure, villages and urban settlements within and around the subsidence bowls. In addition, flood risk aggravation should be expected in flood-prone sectors affected by subsidence. The tool provides metrics based on existing flood hazard maps, DSM and projected subsidence rates. The developed service and analytical workflow chain has been tailored for automation and operational monitoring and envisions complementarity to the upcoming European Ground Motion Service.

Sometimes everything is not what it seems in DInSAR results. Many ground instabilities detected by DInSAR techniques are clear cases of active slope movements, artificial fill compaction or subsidence induced by mining, water pumping or rock dissolution. However, some cases present a complexity that makes the interpretation of detected movements difficult. Furthermore, the slope movements detected by DInSAR have different characteristics that determine their possible future behavior and, therefore, the danger they can pose to infrastructures or population. For this reason, the characterization of active landslides and other ground instabilities will have a great importance to manage their associated risk. Several years of research in Andalusia (S Spain) offered relevant case studies where the interpretation of ground movements was not straightforward. These cases are being studied from different points of view to better understand their origin. Here, we describe two examples of areas in motion with particular characteristics where an initial interpretation based on general rules fail in providing a reliable explanation of the detected movements. The first example is a close depression, the Zafarraya Polje, created by tectonic and dissolution processes where slight displacements were detected during the 2003-2008 period. An initial interpretation following general rules led to a typical compaction of surficial sediments induced by groundwater withdrawal. A second interpretation has pointed out to an active tectonic origin and the most elaborated theory on its origin is based on rock massif compaction. The second example is an urban state showing severe pathologies in some buildings and slight movements identified by DInSAR. The previous interpretation of the movements was an active slope movement impacting all area but a subsequent thorough evaluation suggests a more complex situation that combine sliding and compaction of artificial fillings. We also show some other areas in motion without a straightforward interpretation where there is not a clear diagnosis of the movements’ origin. In the presented cases, the interpretation can determines solutions or measures to be taken regarding the ground instabilities. They are good examples to illustrate the need of integrating a deep knowledge of the terrain with DInSAR results in order to improve the assessments carried out in DInSAR studies.

Decades after the end of coal mining in the Province of Limburg, the Netherlands, lingering effects are still being experienced at the surface, at times in the form of sinkholes. An illustration of such a hazard was the ‘t Loon event in 2011 when a large shopping mall became unstable without actually collapsing due to the development of a sinkhole in the lower laying parking garage.
An interesting aspect, however, has been detected by past radar satellite missions over ‘t Loon, accelerating surface deformation months before the sinkhole formation was observed by a single PS point. In principle, this information may be used, however, due to a large number of InSAR data points (PS/DS) covering the entire area of the mining concessions (~240 km2), it is difficult to identify the relevant data points. On the other hand, relying on a single PS for sinkhole signature is not enough.
Nevertheless, since no other data sets are known to be suitable for setting up an early warning system, we need to overcome this hurdle.

Two main sorts of surface displacements have been identified in the area due to past radar satellite missions: one of large wavelength uplift signature covering the whole former mining concessions, and the other being scattered single data-points subsiding which might be an indication of sinkhole formation.
The uplift signal is attributed to the cessation of minewater pumping from the mines which leads to flooding of the mines and subsequent decompaction of the zone of disturbed rock above coal panels. This decompaction is featured by a large amount of spatially correlated data points uplifting. The subsiding single data-points, which could be indicators of sinkholes, due to their localized nature are more difficult to identify.

In this study we aim to: understand the these two sources of surface displacements and to significantly reduce the search space for sinkhole detection. To this extent, we exploit surface displacements by satellite geodesy together with georeferenced historical mining maps, hydro-geological data (piezometers), geological subsurface data and combine these with the location of infrastructure at the surface.
We exploit ~28 years of InSAR products processed at the request of the Dutch Minister for Economic Affairs and Climate and derived from five satellite missions: ERS, Envisat, Radarsat, TerraSAR-X and Sentinel. The detected regional uplift signal by ESA’s first radar missions (ERS/Envisat) is ongoing and confirmed by recent SENTINEL and Radarsat datasets.
For the regional uplift signal with a rate of displacement of about 5mm/year, we focus on piezometer data to access the correlation between the mine water rise and the uplift at the surface in time and space. The goal is to assess the physical relationship between the water rise and the surface displacements to be able to predict the current and future ground and mine water levels.
For the sinkhole reduction of search space, we first identify subsurface mining configurations which are similar to those where sinkholes occurred in the past. This identification is done using geological information (logs, wells, mining maps and upward drillings). Finally, we spatially correlate the identified subsurface mining configurations which are more prone to sinkhole formation with infrastructure at the surface. The end goal of this study is to show that we can reduce the previously defined search space (whole mining concession) which is relevant for ongoing monitoring of mining related hazards.

In this session, we present our results on a multidisciplinary approach that may facilitate the application of satellite data for timely prediction of future sinkholes in the coal mining concession areas.