Spatial Probabilistic Quantification of Enhanced Earthquake Detection Using a Machine Learning Picker

Jaehong Chung, Ian W. McBrearty, Yongsoo Park, & Gregory C. Beroza

Published September 8, 2024, SCEC Contribution #13526, 2024 SCEC Annual Meeting Poster #006 (PDF)

Recently, machine learning pickers have surpassed traditional pickers in terms of both accuracy and the quantity of earthquake detections; however the spatial detection variability of machine learning pickers has not yet been fully explored. This study investigates the spatial probability of earthquake detection to quantify the improvements in a machine learning catalog compared to a traditional catalog. We derived the spatially varying probabilistic magnitude of completeness from both catalogs and analyzed their differences. We also compare the spatial detection probabilities of P- and S-waves between the two catalogs to investigate the strengths and weaknesses of the machine learning picker.

Citation
Chung, J., McBrearty, I. W., Park, Y., & Beroza, G. C. (2024, 09). Spatial Probabilistic Quantification of Enhanced Earthquake Detection Using a Machine Learning Picker. Poster Presentation at 2024 SCEC Annual Meeting.

Related Projects & Working Groups
Seismology