PRIME-DP: Pre-trained Integrated Model for Earthquake Data Processing

TL;DR

PRIME-DP is a versatile pre-trained seismic waveform model capable of multi-task processing, including phase picking and polarization classification, with high accuracy and adaptability to various seismic analysis tasks.

Contribution

It introduces PRIME-DP, a multi-task pre-trained seismic model that can be fine-tuned for different seismic data processing tasks without changing architecture.

Findings

Achieves over 85% recall for Pg/Sg phase picking.

Attains over 80% accuracy in P polarization classification.

Reaches 95.1% accuracy in event classification after fine-tuning.

Abstract

We propose a novel seismic wave representation model, namely PRIME-DP (Pre-trained Integrated Model for Earthquake Data Processing), specifically designed for processing seismic waveforms. Most existing models are designed to solve a singular problem. Unlike these models, PRIME-DP is capable of multi-task single station seismic waveform processing, including Pg/Sg/Pn/Sn phase picking and P polarization classification. Moreover, it can be fine-tunned to various tasks, such as event classification without architecture modifications. PRIME-DP can achieve a recall rate of over 85% for Pg and Sg phases on continuous waveforms and achieves over 80% accuracy in P polarization classification. By fine-tuning classification decoder with NeiMeng dataset, PRIME-DP achieves 95.1% accuracy on event.