Conference Agenda

Since the end of 2022, a new release of Bodenbewegungsdienst Deutschland (BBD) provided September 2022 by Federal Institute for Geosciences and Natural Resources (BGR) and the first release of the European Ground Motion Service (EGMS) as part of the Copernicus Land Monitoring Service are available. Both services are based on InSAR displacement estimations generated from Sentinel-1 data and cover the whole area of Germany. Although for Germany both products were processed by GAF AG with software developed by Earth Observation Center, which is part of German Aerospace Center (DLR) there are several differences regarding processing. These differences concern e.g. calibration, covered time span, use of DS (EGMS) or not (BBD), criteria for point selection, default displacement model, temporal sampling and raster size used for vertical displacements.
It suggests itself to ask, if there are differences in performance between BBD and EGMS and how well do the two new releases perform compared to other geodetic techniques. For a study commissioned by the surveying authorities of the state of Baden-Württemberg (Landesamt für Geoinformation und Landentwicklung Baden-Württemberg (LGL)), we investigated the performance of BBD and EGMS and validated them against levelling and GNSS data. Areas near the Hambach surface mine, at the cavern storage field Epe and at the SAPOS stations located in and near Baden-Württemberg were selected as test cases. In addition, an assessment of the coverage of the train tracks of Deutsche Bundesbahn, the motorways and federal roads in northern Baden-Württemberg between Karlsruhe and Stuttgart will be given.

The surroundings of the Hambach surface mine show significant linear displacements caused by lowering the ground water table. We compared 14805 points of EGMS with nearby points of BBD. Both services detected essentially linear displacements. Their results show good agreement, as can be expected when the actual displacement is compatible with the displacement models used for processing.
Due to gas storage at Epe, nonlinear displacements occur that are not compatible with either of the displacement models of both services. As anticipated, significant differences between the results of BBD and EGMS are observed in the area of strongest displacement. In addition, levelling data from yearly campaigns from 2015 to 2021 at 615 measurement points (304 useable for BBD, 453 useable for EGMS) were provided by Salzgewinnungsgesellschaft Westfalen (SGW), the operator of the cavern field. Comparison between levelling and InSAR results likewise show a moderate (BBD) or bad agreement (EGMS) in the area of strongest displacement. This is partly due to the additional points selected by EGMS.
As third test case, time series of 32 GNSS stations were compared to nearby points of BBD and 36 (32 plus 4 French or Swiss stations) to nearby points of EGMS. In this case, beside vertical displacements also displacements in East-West direction and LoS were compared. The overall agreement between GNSS and InSAR results from both services is good.
Our comparison shows that the products of BBD and EGMS are of similar quality. For the area of strongest displacement over the cavern storage field at Epe a displacement models adapted to the phenomenon would be needed. The results obtained with the all-purpose models of the services do not agree well with levelling in this area. Compared to GNSS, BBD and EGMS both show good agreement.
Finally, the assessment of the coverage of the train tracks of Deutsche Bundesbahn, the motorways and federal roads in northern Baden-Württemberg between Karlsruhe and Stuttgart will be given shows that a better coverage is obtained with EGMS, presumably because of the use of DS.

The European Ground Motion Service (EGMS) is part of the Copernicus Land Monitoring Service (CLMS) managed by the EEA (European Environment Agency) [1]. EGMS is based on the full resolution InSAR processing of ESA Sentinel-1 (S1) acquisitions over Europe (Copernicus Participating states) [2]. The first release or Baseline includes ground motion timeseries between 2015 and 2020. Yearly updates of this open dataset will be released every 12 months.

The EGMS employs persistent scatterer (PS) and distributed scatterer (DS) in combination with a Global Navigation Satellite System (GNSS) model to calibrate the ground motion products. This public dataset consists of three products levels (Basic, Calibrated and Ortho). The Basic and Calibrated product levels are full resolution (20x5m) Line of sight (Los) velocity maps coming from ascending/descending orbits. The Ortho product offers horizontal (East-West) and vertical (Up-Down) anchored to the reference geodetic model resampled at 100x100m. Since Interferometric Synthetic Aperture Radar (InSAR) data production involves the application of thresholds and filters to remove unwanted phase artefacts the results may contain systematic effects, outliers or simply measurement noise.

The subject of this abstract is to describe the independent validation of this continental scale ground motion timeseries dataset. The goal is to assess that the EGMS products are consistent with user requirements and product specifications, covering the expected range of applications. Information on validation is of great interest to the end users since it indicates which phenomena the EGMS can capture, which are the possible fields of application, and the constraints in the applicability of the EGMS products. To evaluate the fitness of the EGMS ground motion data service seven reproducible validation activities (VA) have been developed gathering validation data from different sources across 12 European countries.

• VA1 – Point density check performed by Sixense. This activity evaluates the point density consistency across the different land cover classes defined in CLC Urban Atlas 2018 (high resolution land cover layer).

• VA2 – Comparison with other ground motion services carried out by NGI. This activity checks the performance of the continental ground motion service against the quality controlled and validated regional initiatives.

• VA3 – Comparison with inventories of phenomena/events performed by BRGM. This activity compares the EGMS data with the information provided by inventories (points locating phenomena, polygons representing the geometry of the phenomena, expected velocity or qualitative characteristics of the motion, dates of events or damages).

• VA4 – Consistency check with ancillary geo-information carried out by NGI. This task makes use of national inventories of geomorphological, geotechnical and geological data together with expert judgement and automated procedures to discover active deformation areas on the EGMS timeseries datasets.

• VA5 – Comparison with GNSS data performed by TNO. The goal of this activity is to perform a validation of the geocoding of the EGMS products together with ground motion timeseries comparison of GNSS measurements.

• VA6 – Comparison with insitu monitoring data performed by GBA. The objective of this task is to evaluate the insitu measurements coming from GPS campaigns, levelling data, extensometers, piezometers, inclinometers, geodetic monitoring, and tilt meters against the EGMS ground motion data.

• VA7 – Evaluation XYZ and displacements with Corner Reflectors performed by TNO. This activity aims to evaluate the precision of the EGMS timeseries (location, height and observed motion).

The EGMS Validation system environment developed and maintained by Terrasigna includes all the necessary elements to perform all the validation tasks from data collection and description to execution of the different methodologies. The objective of this portable cloud-based system is to guarantee reproducibility of all the validation activities:

• A web-based validation data upload tool where scientists can upload their validation data and EGMS subsets.

• A validation data catalogue (based on OGC CSW) where all validation sites data is properly described and georeferenced to ensure reproducibility.

• JupyterHub notebook environment where scientists can develop their validation scripts (Python/R). These notebooks produce graphs and figures to be included in the yearly validation reports.

References

[1] Crosetto, M.; Solari, L.; Mróz, M.; Balasis-Levinsen, J.; Casagli, N.; Frei, M.; Oyen, A.; Moldestad, D.A.; Bateson, L.; Guerrieri, L.; Comerci, V.; Andersen, H.S. The Evolution of Wide-Area DInSAR: From Regional and National Services to the European Ground Motion Service. Remote Sens. 2020, 12, 2043. https://doi.org/10.3390/rs12122043

[2] Costantini, Mario & Minati, F. & Trillo, Fritz & Ferretti, Alessandro & Novali, Fabrizio & Passera, Emanuele & Dehls, John & Larsen, Yngvar & Marinkovic, Petar & Eineder, Michael & Brcic, Ramon & Siegmund, Robert & Kotzerke, Paul & Probeck, Markus & Kenyeres, Ambrus & Proietti, Sergio & Solari, Lorenzo & Andersen, Henrik. (2021). European Ground Motion Service (EGMS). 10.1109/IGARSS47720.2021.9553562.

The use of Interferometric Synthetic Aperture Radar (InSAR) for ground motion detection and monitoring is rapidly increasing, many locations, particularly urban areas around the world, have been studied using different types of satellite data (e.g., ERS-1/2, Sentinel-1, TerraSAR-X), where the rate and distribution of ground movements have been reported. Focus on wide-area deformation monitoring are also increasing, and numerous national services have been established across Europe, including InSAR-Sweden, InSAR Norway, BodenBewegungsdienst Deutschland-Germany, Danish Ground Motion Service, Dutch Ground Motion Service, and the Sentinel-1 Monitoring Services-Italian Regions. Now, the first InSAR monitoring program at a continental scale, the European ground motion services (EGMS), is available (https://egms.land.copernicus.eu/). Thanks to the availability of Copernicus Sentinel-1 satellites images, which cover relatively large areas with a 12-day revisit time. The EGMS is based on the multi-temporal interferometric analysis of Sentinel-1 satellite images and currently covers the period between February 2015 to December 2021 (the first update) and is planned to be updated annually. The EGMS provides ground motion information at three main levels: the basic product, which provides the displacement motion along the line of site (LOS) and is referred to a local reference point; the calibrated product, which is similar to the basic one and is referenced to derived GNSS data model, making absolute InSAR measurement; and the ortho products, which use the calibrated product to generate vertical and east-west displacement by combining ascending and descending measurements.

This study compares previous Persistent Scatterer Interferometry (PSI) study results with the EGMS in terms of vertical and E-W movements components in Gävle city in Sweden, where Gido et al. (2020) studied active ground subsidence using Sentinel-1 data collected between 2015 and 2020. The PSI technique was used to estimate the subsidence rate for Gävle city, and the results were validated with a long record of precise leveling data and correlated with geological observations. The study compares the vertical and E-W displacement time-series at some deforming locations using combined ascending and descending data for both PSI results. Although the number and imaging dates of Sentinel-1 data and the parameters used for PSI processing are not entirely the same, the compared results demonstrate a good agreement between corresponding study on the localization and rate of displacement in the city in the last 5 years. It is worth mentioning that we have previously done similar validation work for GMS of Sweden looking at the LOS rates, Nilfouroushan et al. (2023), and in this study, we will focus on vertical and east-west displacement rates. The existence of National Ground Motion Services in different countries provides an opportunity to compare and cross-check the new EGMS.

The present work was born with the intention of combining two main activities: the exploitation of the incredible opportunity provided by the EGMS (European Ground Motion Service) initiative with respect to the availability of ground motion data, and one of the activity carried out by ISPRA and Ministry of Culture – (General Directorate for the Safety of Cultural Heritage), in the general framework of the implementation of the first “Extraordinary National Plan for Monitoring and conservation of Italian Cultural Heritage ” (NPMCH).

EGMS is the largest wide-area A-DInSAR service ever created, provided consistent, updated, standardized, reliable information regarding natural and anthropogenic ground motion phenomena; based on Sentinel-1A and 1B SAR data, processed at full resolution and the ground motion is estimated using an A-DInSAR approach aimed to derive deformation maps and time series.

The NPMCH is aimed at the monitoring, conservation, and proactive protection of cultural heritage, and specifically on its protection against the impacts of different hazards, both anthropogenic and natural, including climate-induced extreme events.

Starting from this purpose, two case studies have been selected and carried out: the archaeological area of the Phlegrean Fields and the ancient port of Classe in Ravenna city.

For both, aim target was to evaluate the potential ground deformation affecting the archaeological areas using both the EGMS (data and products) and the high-definition Cosmo-SkyMed data, coming from Italian Space Agency mission.

The archaeological area of the Phlegrean Fields, a coastal region in southern Italy located in an active caldera near Naples, is therefore an area prone to potential ground deformation phenomena. More in detail, a specific Interferometric Synthetic Aperture Radar (InSAR) analysis has been implemented, focusing on the period between 2016 and 2020 using Sentinel-1 SAR data to generate ground displacement measurement points (Persistent Scatterers with times series) and to analyze their spatial distribution and correlation with slope instability and archaeological remains damages.

First result shows significant deformation patterns in the area, with vertical uplift rates up to 50 mm/year in the central volcanic area (Pozzuoli). The analysis yields numerous but not exhaustive information about the presence of small-scale landslide phenomena in the surroundings of the Roman Thermae of Baia. Then an InSAR analysis using high-definition Cosmo-SkyMed SAR data has been performed, to derive information on small scale landslides by comparing the time series made with CSK and SENTINEL data. The CSK data are in X-band (wavelength 3.1 cm) and have a spatial resolution of 3 meters, much precise than Sentinel-1 (20 meters), with the ability to detect even smaller displacements affecting archaeological structures (e.g. walls, roof, caves and rock structures). The dataset consists of Images (57) descending and (60) ascending scenes in the period from 2017 to 2021. Data processing has been performed using the Interferometric synthetic aperture radar Scientific Computing Environment (ISCE), the Stanford Method of Persistent Scatterers (StaMPS) and TRAIN Toolbox for Reducing Atmospheric InSAR Noise. Moreover, results data have been calibrated by local GNSS network data.

CSK data results provide useful elements to confirm current uplift trend in the entire Phlegraean Fields area in accordance with Sentinel data. After a recent extraordinary clearing of the slope from vegetation, the overall stability condition was better clarified. InSAR analysis provides very useful information to detect and monitor ground displacements, thus offering to archaeological site managers a powerful tool for the prevention of ground related damage of cultural heritage.

The coastal area of Ravennna is historically affected by both natural and anthropogenic subsidence processes at different scale, from regional to local. First results performed trough SBAS processing of Cosmo Sky-Med dataset and calibration with GNSS regional network, in the time interval between years 2018 and 2022, confirmed the general ground subsidence affecting the Ravenna area of about 5 mm/yr, as measured by the local GNSS station. Any differential displacements affects the archaeological area of the Port of Classe, while few local settling have been highlighted on recent commercial building in the city's suburbs, This results are in accordance also with the measurements obtained by the Copernicus EGMS in the same time interval, coming from Sentinel-1 data.

The main results of this study have highlighted the importance of the EGMS service for preliminary studies at medium resolution. The anomalies highlighted at the sub-regional and municipal scale must then be detailed in both spatial and temporal resolution in order to be correctly interpreted, validated and calibrated directly in situ.

The last decades have seen a growing need for sophisticated tools that enable a constant and reliable monitoring of ground-motion phenomena, as part of more and more integrated risk assessment and management workflows. The exposure of the built environment to geohazards has increased, due to the rapid urbanization, man-induced environmental transformations leading to higher hydrogeological risk, and global climate change. The availability of satellite Synthetic Aperture Radar (SAR) datasets with increasing spatial and temporal coverage, with decreasing temporal intervals between two subsequent acquisitions, such as the ones collected by the Copernicus Sentinel-1 constellation, gives the opportunity to analyse and monitor ground-surface deformation phenomena, of natural origin or man-induced.

Ground-motion phenomena have shown to be well studied using satellite radar interferometry [1] and have seen relevant developments in terms of accuracy and coverage with the introduction of techniques based on persistent scatterers (Persistent Scatterer Interferometric SAR, PSInSAR) [2]. The data availability has been accompanied by a development of increasingly performing PSI algorithms and processing chains [3]. Up to the present moment, the monitoring of ground motion phenomena has been mainly performed using the PSInSAR technique on a local scale, by developing advanced processing chains that adapt to one particular case study. However, a major challenge consists of developing robust processing architectures that can detect hazards and actualize the available information on more extended areas [4-5].

There is now the capability to monitor entire countries and a pan European Ground Motion Service (EGMS) has been recently activated. The European Ground Motion Service (EGMS) is the most recent addition to the product portfolio of the Copernicus Land Monitoring Service. The Service is funded by the European Commission in the frame of the Copernicus Programme. It is implemented under the responsibility of the European Environment Agency [6-7].The EGMS distributes three levels of products: (i) Basic, consisting of line-of-sight (LOS) velocity maps in ascending and descending orbits referred to a local reference point; (ii) Calibrated, which is obtained by correcting the Basic product data using a model derived from Global Navigation Satellite Service (GNSS) data as reference; (iii) Ortho, containing the vertical and horizontal (East-West) displacements computed from the Calibrated data. The available EGMS data refer to the period ranging from 2015 to 2020. Both Basic and Calibrated products are derived from full resolution (~4 by 14 m) Sentinel-1 radar images. Ortho product is resampled on a regular grid with 100 by 100 m cells [6-7].

This work focuses on the development of an automatic routine to extract the areas affected by ground motion phenomena over wide areas using the EGMS Basic product datasets (due to their higher resolution), with the aim of building a database of active deformation areas (ADAs). The results shown in this abstract were obtained applying the developed processing routine over Spain, while the production of a pan European database is an ongoing activity.

ADAs may have different causes, such as landslides, sinkholes, subsidence and volcanic activity, and rigorous scheme for their detection should involve the evaluation of the pixel displacement time series and average velocity (contained in the EGMS datasets). The ADA Finder tool has been previously developed [8-9] with the aim of easing the management, use and interpretation of PSInSAR results, consisting of an ADA detection algorithm based on few spatial and statistical parameters of the pixel displacement time series. The ADA Finder tool first removes outliers and isolated PS points, then a velocity threshold is applied to eliminate points that are considered as stable. In this work we set the value of this threshold at 5 mm/year, considering that the average noise level for the velocity values of the EGMS Basic product is about 2 mm/year [8]. Then, the detected points whose distance is lower than 40 m are grouped together into one polygon defining a new ADA. The final stage computes a quality index (QI) for each detected ADA, with values ranging from 1 (reliable ADAs) to 4 (very noisy ADAs). The QI values are computed accounting for the spatio-temporal correlation properties of the displacement values associated with the points forming the ADA under analysis [8]. In this work, only the ADAs whose QI is equal to 1 or 2 are considered. The ADA Finder output consists of polygons associated with the detected ADAs, together with their QI, and few relevant statistical parameters (mean, maximum and minimum displacement velocity, number of PS points).

The ADA Finder tool was employed to each burst of the EGMS Basic product covering the Spanish territory, using a parallel processing routine implemented in Python on a 48 CPU core computer. We observe that the bursts of the EGMS Basic product are associated to the burst of Sentinel-1 data, separately for the ascending and descending orbit trajectories. The total processing time was about 48 hours. The detected ADAs were finally merged together to generate two databases of detected ground deformation areas associated with the Sentinel-1 ascending and descending orbit trajectories. The obtained results are shown in Figure 1, together with the Digital Elevation Model (DEM) values of the NASA SRTM with 90 m resolution.

The total number of detected ADAs is about 3400 and 2200 for the ascending and descending data, respectively, with surfaces ranging from 2000 m² to 29 km². A high density of deformation areas can be noticed in the South-East of Spain, where several case studies are present, such as the ground subsidence affecting the Lorca and Murcia plains, due to intense groundwater exploitation [10], and the landslides occurring in the Granada province [11]. In the full paper version, we aim to present the results of the automatic ADA detection for the whole European territory covered by the EGMS, together with a preliminary statistical analysis of the ADA database.

(a) (b)

Figure 1. Map of detected ADAs covering Spain, using the EGMS Basic product data, ascending (a) and descending (b) Sentinel-1 orbits, overlayed with the 90m SRTM DEM values (gray scale).

References

[1] Massonnet, D., Feigl, K.L., Radar interferometry and its application to changes in the Earth's surface, Reviews of Geophysics, 36(4), 441-500, 1998

[2] Ferretti, A., Prati, C., Rocca, F., Permanent scatterers in SAR interferometry, IEEE Transactions on Geoscience and Remote Sensing, 39(1), 8-20, 2001

[3] Pepe, A., Yang, Y. Manzo, M. and Lanari, R., "Improved EMCF-SBAS Processing Chain Based on Advanced Techniques for the Noise-Filtering and Selection of Small Baseline Multi-Look DInSAR Interferograms," in IEEE Transactions on Geoscience and Remote Sensing, vol. 53, no. 8, pp. 4394-4417, Aug. 2015

[4] Raspini, F., Bianchini, S. Ciampalini, A., Del Soldato, M., Solari, L., Continuous, semi-automatic monitoring of ground deformation using Sentinel-1 satellites, Scientific Reports, vol. 8, article number: 7253 (2018).

[5] Festa, D., Bonano, M., Casagli, N., et al.; Nation-wide mapping and classification of ground deformation phenomena through the spatial clustering of P-SBAS InSAR measurements: Italy case study, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 189, 2022, pp. 1-22

[6] Crosetto, M., Solari, L., Mróz, M., Balasis-Levinsen, J., Casagli, N., Frei, et al. (2020). The evolution of wide-area DInSAR: From regional and national services to the European Ground Motion Service, Remote Sensing, 12(12), 2043

[7] Larsen, Y., Marinkovic, P., et al. (2020). European Ground Motion Service: Service Implementation. Copernicus Land Monitoring Service. https://land.copernicus.eu/usercorner/technical-library/egms-specification-and-implementation-plan

[8] Barra, A., Solari, L., Béjar-Pizarro, M., Monserrat, O., Bianchini, S., Herrera, G., Crosetto, M., Sarro, R., González-Alonso, E., Mateos, R.M., Ligüerzana, S., López, C., Moretti, S. (2017). A methodology to detect and update active deformation areas based on Sentinel-1 SAR images, Remote Sensing, 9, 1002

[9] Tomás, R., Pagán, J. I., Navarro, J. A., Cano, M., Pastor, J. L., Riquelme, A., et al. (2019). Semi-automatic identification and pre-screening of geological–geotechnical deformational processes using persistent scatterer interferometry datasets, Remote Sensing, 11(14), 1675

[10] Bonì R., Herrera, G., Meisina, C., et al, Twenty-year advanced DInSAR analysis of severe land subsidence: The Alto Guadalentín Basin (Spain) case study, Engineering Geology, 198,2015, pp 40-52.

[11] Reyes-Carmona,C., Galve, J.P., Moreno-Sánchez, M., et al. Rapid characterisation of the extremely large landslide threatening the Rules Reservoir (Southern Spain). Landslides 18, 3781–3798 (2021)