The 2022 Outstanding Student and PhD candidate Presentation (OSPP) Award is awarded to Jannes Münchmeyer for the poster/PICO entitled:
Which picker fits my data? A quantitative evaluation of deep learning based seismic pickers (Münchmeyer, J.; Woollam, J.; Rietbrock, A.; Tilmann, F.; Lange, D.; Bornstein, T.; Diehl, T.; Giunchi, C.; Haslinger, F.; Jozinović, D.; Michelini, A.; Saul, J.; Soto, H.)
Click here to download the poster/PICO file.
Jannes Münchmeyer is a PhD Student at GFZ Potsdam and the Humboldt University Berlin under the supervision of Prof. Frederik Tilmann and Prof. Ulf Leser. His research focuses on the detection and real-time assessment of earthquakes using machine learning methods. He developed deep learning methods for earthquake early warning and the real-time estimation of earthquake magnitude and location. Furthermore, he proposed a probabilistic framework for the study of earthquake rupture predictability and applied it to show that earthquake ruptures cannot be assessed precisely during their initial growth phase.
The research presented at EGU22 concerned a quantitative comparison of deep learning based seismic phase pickers. To conduct this study, Jannes and his collaborators built SeisBench: A framework for machine learning in seismology. In the study, he showed how previously published pickers perform when trained on diverse datasets and to which extent they are transferable across datasets. The study ends with specific advise on the optimal pickers for different application scenarios.