Deep learning phase pickers: how well can existing models detect hydraulic-fracturing induced microseismicity from a borehole array?

Lim, Cindy S.Y. ORCID: https://orcid.org/0000-0003-1977-5861; Lapins, Sacha ORCID: https://orcid.org/0000-0002-0113-0240; Segou, Margarita ORCID: https://orcid.org/0000-0001-8119-4019; Werner, Maximilian J. ORCID: https://orcid.org/0000-0002-2430-2631. 2025 Deep learning phase pickers: how well can existing models detect hydraulic-fracturing induced microseismicity from a borehole array? Geophysical Journal International, 240 (1). 535-549. 10.1093/gji/ggae386

Before downloading, please read NORA policies.

Preview

Text (Open Access Paper) ggae386.pdf - Published Version Available under License Creative Commons Attribution 4.0. Download (4MB) | Preview

Abstract/Summary

Deep learning (DL) phase picking models have proven effective in processing large volumes of seismic data, including successfully detecting earthquakes missed by other standard detection methods. Despite their success, the applicability of existing extensively trained DL models to high-frequency borehole data sets is currently unclear. In this study, we compare four established models [Generalized Seismic Phase Detection (GPD), U-GPD, PhaseNet and EQTransformer] trained on regional earthquakes recorded at surface stations (100 Hz) in terms of their picking performance on high-frequency borehole data (2000 Hz) from the Preston New Road (PNR) unconventional shale gas site, in the United Kingdom (UK). The PNR-1z data set, which we use as a benchmark, consists of continuously recorded waveforms containing over 38 000 seismic events previously catalogued, ranging in magnitudes from −2.8 to 1.1. Remarkably, all four DL models can detect induced seismicity in high-frequency borehole data and two might satisfy the monitoring requirements of some users without any modifications. In particular, PhaseNet and U-GPD demonstrate exceptional recall rates of 95 and 76.6 per cent, respectively, and detect a substantial number of new events (over 15 800 and 8300 events, respectively). PhaseNet’s success might be attributed to its exposure to more extensive and diverse instrument data set during training, as well as its relatively small model size, which might mitigate overfitting to its training set. U-GPD outperforms PhaseNet during periods of high seismic rates due to its smaller window size (400 samples compared to PhaseNet’s 3000-sample window). These models start missing events below $M_w$ −0.5, suggesting that the models could benefit from additional training with microseismic data-sets. Nonetheless, PhaseNet may satisfy some users’ monitoring requirements without further modification, detecting over 52 000 events at PNR. This suggests that DL models can provide efficient solutions to the big data challenge of downhole monitoring of hydraulic-fracturing induced seismicity as well as improved risk mitigation strategies at unconventional exploration sites.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.1093/gji/ggae386
ISSN:	0956-540X
Date made live:	24 Jan 2025 15:24 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/538801

Actions (login required)

View Item

Document Downloads

More statistics for this item...