Inversion-based Time Series Analysis of PS-InSAR Data: Uncovering the Origins of Subsidence and Annual Fluctuations in Southern Hesse, Germany

Abstract

The uplift and subsidence of the earth's surface can be attributed to many different processes. In urban regions in particular, it is important to understand which ground movements occur, whether they pose a risk to infrastructure and whether countermeasures can be taken. While events such as earthquakes, sinkholes and landslides have abrupt and visible effects, slow ground movements such as slope instabilities and tectonic movements are difficult to detect and it can take decades for visible damage to occur. Remote sensing, especially InSAR and Persistent Scatterer InSAR, provides high spatial and temporal coverage for monitoring these processes. The state of Hesse in central Germany is confronted with various ground movements, including former open-cast lignite mines, active salt mining and landslide-prone geological units. Our study aims to explore previously unknown ground movements in urban regions using remote sensing, analyse detected areas, determine causes, assess risks and anticipate future developments. We use Persistent Scatterer Interferometry (PS-InSAR) data from the Ground Motion Service Germany (BBD) to analyse ground motion patterns. With the help of a Ground Motion Analyser (GMA), time series analysis and external data, we identify regions with significant ground movements throughout the federal state. A case study in Frankfurt am Main, located on the northern edge of the Upper Rhine Graben, shows subsidence that was probably caused by groundwater extraction during the construction of buildings. Another case study in Crumstadt, in the centre of the northern Upper Rhine Graben, shows pronounced seasonal fluctuations, possibly related to temperature and activity in an underground gas reservoir. In both cases, an analysis of external data such as groundwater levels, climate data, construction activity, mining activities, hydrogeological and geological conditions must be taken into account. The ground movements caused by the various possible causes can sometimes be very similar, so a solid external database is particularly important. With the help of the results, the ground movements measured by remote sensing can be linked both qualitatively and quantitatively with the regional conditions. Our results thus contribute to understanding and mitigating the effects of ground movements and underline the importance of analysing both time-varying movements and linear velocities in parallel with external data.