A multitask encoder–decoder to separate earthquake and ambient noise signal in seismograms

Abstract

Summary Seismograms contain multiple sources of seismic waves, from distinct transient signals such as earthquakes to continuous ambient seismic vibrations such as microseism. Ambient vibrations contaminate the earthquake signals, while the earthquake signals pollute the ambient noise’s statistical properties necessary for ambient-noise seismology analysis. Separating ambient noise from earthquake signals would thus benefit multiple seismological analyses. This work develops a multi-task encoder-decoder network named WaveDecompNet to separate transient signals from ambient signals directly in the time domain for 3-component seismograms. We choose the active-volcanic Big Island in Hawai’i as a natural laboratory given its richness in transients (tectonic and volcanic earthquakes) and diffuse ambient noise (strong microseism). The approach takes a noisy 3-component seismogram as input and independently predicts the 3-component earthquake and noise waveforms. The model is trained on earthquake and noise waveforms from the STandford EArthquake Dataset (STEAD) and on the local noise of seismic station IU.POHA. We estimate the network’s performance by using the Explained Variance (EV) metric on both earthquake and noise waveforms. We explore different neural network designs for WaveDecompNet and find that the model with Long-Short-Term-Memory (LSTM) performs best over other structures. Overall, we find that WaveDecompNet provides satisfactory performance down to a Signal-to-Noise-Ratio (SNR) of 0.1. The potential of the method is 1) to improve broadband SNR of transient (earthquake) waveforms and 2) to improve local ambient noise to monitor the Earth’s structure using ambient noise signals. To test this, we apply a Short-Time-Average to a Long-Time-Average (STA/LTA) filter and improve the number of detected events. We also measure single-station cross-correlation functions of the recovered ambient noise and establish their improved coherence through time and over different frequency bands. We conclude that WaveDecompNet is a promising tool for a broad range of seismological research.

Multi-task encoder with attention mechanism to remove noise from GPR signal

“…For the task of signal denoising of ground penetrating radar, we choose the method proposed by Jiuxun Yin [11] which is proposed for processing seismic data. According to the characteristics of GPR noise, we propose a new multi-task encoder ADAE with integrated attention mechanism.…”

Improved Observations of Deep Earthquake Ruptures Using Machine Learning

“…For subsequent validation of the source characteristics, we select the raw, noisy P waves with SNR > 2 (as defined in Equation 1) and extract the denoised P waves through DenoTe. This ensures that the post-processing analysis is only selecting data that could have been included in previous analysis and should limit the effect of artifacts generated by the model (though these were minimal when using the WaveDecompNet kernel Yin et al (2022)).…”

“…This ensures that the post‐processing analysis is only selecting data that could have been included in previous analysis and should limit the effect of artifacts generated by the model (though these were minimal when using the WaveDecompNet kernel Yin et al. (2022)).…”

“…Yin et al. (2022) combined two‐branch encoder‐decoder and recurrent neural networks to compose the WaveDecompNet, which has been proven effective in reconstructing local earthquake and noise waveforms. Yin et al.…”

“…Yin et al. (2022) demonstrated that even the clean noise waveforms improved the coherence of noise single‐station cross‐correlations for ambient noise seismology.…”