EPick: Attention-based multi-scale UNet for earthquake detection and seismic phase picking

Abstract

Earthquake detection and seismic phase picking play a crucial role in the travel-time estimation of P and S waves, which is an important step in locating the hypocenter of an event. The phase-arrival time is usually picked manually. However, its capacity is restricted by available resources and time. Moreover, noisy seismic data present an additional challenge for fast and accurate phase picking. We propose a deep learning-based model, EPick, as a rapid and robust alternative for seismic event detection and phase picking. By incorporating the attention mechanism into UNet, EPick can address different levels of deep features, and the decoder can take full advantage of the multi-scale features learned from the encoder part to achieve precise phase picking. Experimental results demonstrate that EPick achieves 98.80% accuracy in earthquake detection over the STA/LTA with 80% accuracy, and for phase arrival time picking, EPick reduces the absolute mean errors of P- and S- phase picking from 0.072 s (AR picker) to 0.030 s and from 0.189 s (AR picker) to 0.083 s, respectively. The result of the model generalization test shows EPick’s robustness when tested on a different seismic dataset.

Earthquake monitoring using deep learning with a case study of the Kahramanmaras Turkey earthquake aftershock sequence

“…In this part, the pre-trained DynaPicker on the seismic data with different time duration is further evaluated on continuous seismic data. Moreover, the model is compared with EPick (Li et al, 2022a), a simple neural network that incorporates an attention mechanism into a U-shaped neural network. Here, the pre-trained and saved model of EPick is directly used without retraining.…”

“…Thus, the use of deep learning has been widely embraced in first-motion polarity identification of earthquake waveforms (Chakraborty et al, 2022a), seismic event detection (Perol et al, 2018;Mousavi et al, 2019b;Fenner et al, 2022;Li et al, 2022b), earthquake magnitude classification and estimation (Chakraborty et al, 2021(Chakraborty et al, , 2022b, and seismic phase picking (Ross et al, 2018;Zhu and Beroza, 2019;Mousavi et al, 2020;Li et al, 2021aLi et al, , 2022a. Stepnov et al (2021) stated that seismic phase picking approaches can be roughly divided into two main streams: continuous seismic waveform-based and small window-format-based methods.…”

Recent advances in earthquake seismology using machine learning

“…Although the specific problem of "prediction inconsistency" has been reported (Park et al 2023), DL-based pickers outperform traditional algorithms and achieve picking accuracies similar to those of skilled analysts (Mousavi and Beroza 2023), and the published models have been used in many studies. Similar to event detection/classification problems, stateof-the-art modules and architectures, such as RNN (Zhou et al 2019), attention (Liao et al 2021(Liao et al , 2022aLi et al 2022a), transformer , and edge convolutional module (Feng et al 2022b), were continuously incorporated into the models to improve their performance. In this approach, a model takes seismic waveforms as inputs and outputs the arrival times as scalar values (Ross et al 2018a) or a time series of probability values with a peak at a picked arrival time (Zhu and Beroza 2019;.…”

Seismic Event Detection in the Copahue Volcano Based on Machine Learning: Towards an On-the-Edge Implementation

“…The activation threshold, also known as the trigger, plays a crucial role in determining which events are recorded and which ones are not. Furthermore, experts should set the parameters of STA/LTA methods in such a way that they cannot take advantage of the prior knowledge of previous picks, since each measurement is treated individually [9].…”