We live in times of increasing vulnerability to extreme natural hazards, and growing conscience exists that our society remains insufficiently prepared to confront high-impact, low-probability natural catastrophes due to insufficient understanding of such events and their impacts. The key to accurately estimate the probability of an extreme natural hazard to occur in the near future lies in our ability to accurately date deposits of events such as tsunamis, storm floods, landslides, and earthquakes that occurred in the past, to reconstruct recurrence patterns. A number of different dating methods have been developed and are in use (e.g. radiometric, surface exposure, optically stimulated luminescence, etc.), but all have their particular limitations (e.g. with respect to the datable age range, the deposit's material, or external environmental factors) and there is no single method that can date all natural hazard deposits; and some cannot by dated by any of the available methods.

I am developing and testing a novel but promising dating method for natural hazards, based on magnetic records in rocks that are acquired post-emplacement of the rock during a natural hazard event: Microscopic magnetic inclusions in rocks normally carry a magnetization that is aligned with the geomagnetic north at the time of their formation. After a rock/boulder is deposited by a natural disaster, the orientation of this magnetization is no longer aligned northward. Over time, some of these grains re-orient their magnetization vector northward through a slow viscous process: a viscous remanent magnetization (VRM). Determining the progress of this VRM allows to estimate the time passed after deposition.

Case studies of this method during my PhD gave promising results and showed that “VRM geochronology” can be a viable order-of-magnitude dating tool. My current research focusses on improvements of the method, in particular with respect to rocks with complex magnetic mineralogies, the application of the method on a larger variety of deposit types (e.g. conglomerates), as well as on carrying out a larger number of case studies to assess the performance of the method.