Details

MarDATA in Antarctica

In the WG Industrial Mathematics, models for the automated analyses of earthquake events in Antarctica were developed based on machine learning methods.

The Alfred-Wegener-Institute (AWI) operates the Neumayer III research station in Antarctica, where the geophysical observatory has been responsible for earthquake monitoring for more than 20 years and records seismological data for this purpose. Numerous earthquake signals are recorded every day, which have so far been manually evaluated and 'picked' by data analysts at the station, i.e. the exact time of arrival of the earthquakes is determined on the basis of the seismograms.

In preparation for her PhD in the MarDATA Helmholtz School for Marine Data Science in cooperation with the University of Bremen, the AWI in Bremerhaven, the GEOMAR in Kiel and the Christian-Albrechts-University Kiel, Louisa Kinzel developed and tested machine learning models in her master's thesis at ZeTeM to automate the process of earthquake analyses.

In December/January, the research ship Polarstern brought the researchers and the wintering crew to Neumayer Station. The two scientists Timo Dornhöfer and Lorenz Marten, who are responsible for the geophysical observatory, are now working on implementing the newly developed models on site and testing them during operation to support them in their routine task of earthquake picking.

So the ZeTeM has now reached into the eternal ice of Antarctica with its novel developments in machine learning!

 

[Translate to English:]
[Translate to English:]