arXiv:2605.22837v1 Announce Type: cross Abstract: Numerous studies have shown that the machine-learning picker PhaseNet produces accurate P and S picks on local earthquake signals, but its performance can degrade sharply on teleseismic signals. To address this limitation, we present a reproducible MsPASS workflow that (i) enables scalable data preparation and management for large seismic archives and (ii) supports standardized PhaseNet training and inference. We assembled a control dataset of 1.6 million waveforms linked to teleseismic P-wave picks made by analysts at the USArray Array Network
The proliferation of AI in scientific domains necessitates continuous evaluation and refinement of models like PhaseNet for improved real-world performance and data processing efficiency.
Improving AI performance on challenging datasets like teleseismic signals broadens its applicability in critical scientific fields such as seismology, leading to more accurate predictions and understandings of natural phenomena.
The ability to accurately apply machine learning models to complex teleseismic data means more efficient and precise analysis of global seismic events, which was previously a known limitation.
- · Geophysical research institutions
- · Seismologists
- · AI/ML developers
- · Traditional manual data analysis methods
PhaseNet accuracy on teleseismic data improves, leading to better seismic event detection.
Enhanced global earthquake monitoring and early warning systems become more feasible.
Deeper understanding of Earth's internal structure and processes through more granular seismic data analysis.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG