We investigate how faults are re-activated under in-situ conditions and determine the energy budget of this process. We search to minimize radiated seismic energy and maximize permeability for safe and efficient heat mining. We use underground research facilities in Germany, France and Sweden.

We analyze a wide range of geophysical, geological, and engineering data to derive information about the contemporary crustal stress field. This information is compiled into a unique global database and used to develop geomechanical models with quantified uncertainties, enabling stress predictions and its changes due to both natural and anthropogenic processes.

We develop seismic ground motion models and tools to improve seismic hazard assessment, incorporating advanced methodologies to characterize ground shaking across different regions. Additionally, we investigate the origins of epistemic and aleatory variability in ground motion, aiming to better understand its physical and statistical sources to refine hazard predictions.

We undertake research to advance the development of probabilistic methodologies for seismic hazard and risk analysis and apply these into practical application in site-specific, national and international scale. Such models underpin several of society’s most effective methods to mitigate risk from earthquakes, acting as inputs to earthquake resistant building codes and as a basis for determining the risk to insurance portfolios.

Growing cities and changes in urban developments changes the risk, that is the impact of natural hazard on infrastructure, in complex ways. For example, it is not scaling with population size but depends on changing building stock, resilience measures, and also city structure. Our focus is to quantify these changing risks.

In recent years, there has been a growing recognition that natural hazards can interact and occur simultaneously, leading to compounding effects that exceed the combined impact of individual hazards. Multi-hazard approaches aim to address the limitations inherent to single-hazard models, which fail to account for the interactions between different hazards and their potential compounding or cascading effects.