Near real-time channel selection for Distributed Acoustic Sensing technology

TL;DR

This paper introduces ORION, a fast channel selector for Distributed Acoustic Sensing data that reduces data volume while maintaining seismic monitoring accuracy, enabling near real-time analysis.

Contribution

The paper presents a novel spatial clustering and waveform attribute-based method for efficient DAS channel selection, improving real-time seismic monitoring capabilities.

Findings

ORION effectively reduces DAS data volume.

Selected channels preserve azimuthal coverage.

Improved hypocenter accuracy in seismic analysis.

Abstract

Distributed Acoustic Sensing (DAS) technology is advancing seismic monitoring by providing dense observations near earthquake sources. However, the resulting data volumes often limit real-time processing capability, with most seismological applications focusing on retrospective analysis of seismic sequences. To address this challenge, we introduce ORION, a fast and versatile selector of high-quality DAS channels that efficiently reduces the amount of data to analyze. The method first adopts spatial clustering to identify cable segments with similar geometrical attributes (e.g, azimuth), and then performs channel selection within each section using waveform attributes (e.g., signal-to-noise ratio); this approach enables spatial sub-sampling while preserving azimuthal coverage. We demonstrate the flexibility of the selector across several cable geometries. Finally, we analyze a seismic sequence using ORION-selected channels and compare the source locations with those from a more conventional uniform distribution of channels along the cable, showing improvements in hypocenter accuracy.