Earthquake Seismogram Denoising Across Time, Time‐Frequency, and Hybrid Domain Approaches
- 1Institute of Geophysics ETH Zurich Zurich Switzerland
Abstract
Deep learning earthquake signal denoisers commonly operate directly on waveform data or on time‐frequency representations, yet the impact of this design choice has not been systematically assessed under controlled conditions. Here we design and benchmark four comparable deep learning architectures to assess the effect of input data representation: a time‐domain denoiser, a time‐frequency denoiser–the Earthquake Seismogram (EQS) Denoiser–and two hybrid models processing both domains in parallel or sequentially. We evaluate waveform quality as a direct measure of denoising performance and assess the impact on signal detection and phase picking tasks using standard phase pickers. We further examine how denoising enhances signal onsets using a dedicated first‐motion polarity model, benchmark the best‐performing model against state‐of‐the‐art denoisers, and evaluate its generalization to out‐of‐distribution data. Our results show that the time‐domain model produces the most accurate waveforms at high signal‐to‐noise ratios (SNRs), but fails to detect event signals at low SNRs. The time‐frequency denoiser captures event signals down to lower SNRs but can introduce artifacts such as smeared onsets due to the time‐frequency binning. Hybrid models mitigate shortcomings of single‐domain approaches. The hybrid sequential model EQShyb, which extends the EQS architecture, demonstrates strong signal detection at low SNRs while preserving sharp onsets, improving phase picking and extending first‐motion polarity determination to lower SNRs compared to raw or bandpass‐filtered data. Overall, EQShyb surpasses benchmark models across every waveform and downstream task metric for both in‐ and out‐of‐distribution data, demonstrating that hybrid domain processing is a powerful strategy for accurate seismic denoising.
Related
- Deep-learning-based denoising and interpolation of small displacements caused by fault slip using InSAR and GNSS data: Application to the 2019 Mw 6.2 Hyuga-nada earthquake
- Broadband Ground-Motion Synthesis by Conditioning Denoising Diffusion Models on Low-Frequency Waveforms
- Crustal scale heterogeneity across Bhutan Himalayas: insights from local earthquake shear-wave splitting
- Spatiotemporal Analysis for Frequency-Magnitude Distribution of Earthquakes Across Mainland China: Comparing Classical and b-positive b-values Across Tectonic Regimes
- Fit for purpose evapotranspiration measurements: A multi-site comparison of lower-cost approaches across croplands and shrublands
BibTeX
@article{dahmen2026earthquake,
title = {{Earthquake Seismogram Denoising Across Time, Time‐Frequency, and Hybrid Domain Approaches}},
author = {Nikolaj L. Dahmen},
journal = {Journal of Geophysical Research: Machine Learning and Computation},
year = {2026},
month = oct,
doi = {10.1029/2026jh001403},
url = {https://doi.org/10.1029/2026jh001403},
note = {CC BY},
}
RIS
TY - JOUR
TI - Earthquake Seismogram Denoising Across Time, Time‐Frequency, and Hybrid Domain Approaches
AU - Nikolaj L. Dahmen
PY - 2026
DA - 2026/10/28
JO - Journal of Geophysical Research: Machine Learning and Computation
DO - 10.1029/2026jh001403
UR - https://doi.org/10.1029/2026jh001403
AB - Deep learning earthquake signal denoisers commonly operate directly on waveform data or on time‐frequency representations, yet the impact of this design choice has not been systematically assessed under controlled conditions. Here we design and benchmark four comparable deep learning architectures to assess the effect of input data representation: a time‐domain denoiser, a time‐frequency denoiser–the Earthquake Seismogram (EQS) Denoiser–and two hybrid models processing both domains in parallel or sequentially. We evaluate waveform quality as a direct measure of denoising performance and assess the impact on signal detection and phase picking tasks using standard phase pickers. We further examine how denoising enhances signal onsets using a dedicated first‐motion polarity model, benchmark the best‐performing model against state‐of‐the‐art denoisers, and evaluate its generalization to out‐of‐distribution data. Our results show that the time‐domain model produces the most accurate waveforms at high signal‐to‐noise ratios (SNRs), but fails to detect event signals at low SNRs. The time‐frequency denoiser captures event signals down to lower SNRs but can introduce artifacts such as smeared onsets due to the time‐frequency binning. Hybrid models mitigate shortcomings of single‐domain approaches. The hybrid sequential model EQShyb, which extends the EQS architecture, demonstrates strong signal detection at low SNRs while preserving sharp onsets, improving phase picking and extending first‐motion polarity determination to lower SNRs compared to raw or bandpass‐filtered data. Overall, EQShyb surpasses benchmark models across every waveform and downstream task metric for both in‐ and out‐of‐distribution data, demonstrating that hybrid domain processing is a powerful strategy for accurate seismic denoising.
SN - 2993-5210
ER -