Benchmark on the accuracy and efficiency of several neural network based phase pickers using datasets from China Seismic Network

Seismic phase pickers based on deep neural networks have been extensively used recently,demonstrating their advantages on both performance and efficiency.However,these pickers are trained with and applied to different data.A comprehensive benchmark based on a single dataset is therefore lacking.Here,using the recently released DiTing dataset,we analyzed performances of seven phase pickers with different network structures,the efficiencies are also evaluated using both CPU and GPU devices.Evaluations based on F1-scores reveal that the recurrent neural network(RNN)and EQTransformer exhibit the best performance,likely owing to their large receptive fields.Similar performances are observed among PhaseNet(UNet),UNet++,and the lightweight phase picking network(LPPN).However,the LPPN models are the most efficient.The RNN and EQTransformer have similar speeds,which are slower than those of the LPPN and PhaseNet.UNet++requires the most computational effort among the pickers.As all of the pickers perform well after being trained with a large-scale dataset,users may choose the one suitable for their applications.For beginners,we provide a tutorial on training and validating the pickers using the DiTing dataset.We also provide two sets of models trained using datasets with both 50 Hz and 100 Hz sampling rates for direct application by end-users.All of our models are open-source and publicly accessible.

Seismic phases from the Moho and its implication on the ultraslow spreading ridge

The Moho interface provides critical evidence for crustal thickness and the mode of oceanic crust accretion. The seismic Moho interface has not been identified yet at the magma-rich segments （46°-52°E） of the ultra- slow spreading Southwestern Indian Ridge （SWIR）. This paper firstly deduces the characteristics and do- mains of seismic phases based on a theoretical oceanic crust model. Then, topographic correction is carried out for the OBS record sections along Profile Y3Y4 using the latest OBS data acquired from the detailed 3D seismic survey at the SWIR in 2010. Seismic phases are identified and analyzed, especially for the reflected and refracted seismic phases from the Moho. A 2D crustal model is finally established using the ray tracing and travel-time simulation method. The presence of reflected seismic phases at Segment 28 shows that the crustal rocks have been separated from the mantle by cooling and the Moho interface has already formed at zero age. The 2D seismic velocity structure across the axis of Segment 28 indicates that detachment faults play a key role during the processes of asymmetric oceanic crust accretion.

Identification method of seismic phase in three-component seismograms on the basis of wavelet transform

This paper puts forward wavelet transform method to identify P and S phases in three component seismograms using polarization information contained in the wavelet transform coefficients of signal. The P and S wave locator functions are constructed by using eigenvalue analysis method to wavelet transform coefficient across several scales. Locator functions formed by wavelet transform have stated noise resistance capability, and is proved to be very effective in identifying the P and S arrivals of the test data and actual earthquake data.

General seismic wave and phase detection software driven by deep learning

We developed an automatic seismic wave and phase detection software based on PhaseNet,an efficient and highly generalized deep learning neural network for P-and S-wave phase picking.The software organically combines multiple modules including application terminal interface,docker container,data visualization,SSH protocol data transmission and other auxiliary modules.Characterized by a series of technologically powerful functions,the software is highly convenient for all users.To obtain the P-and S-wave picks,one only needs to prepare threecomponent seismic data as input and customize some parameters in the interface.In particular,the software can automatically identify complex waveforms(i.e.continuous or truncated waves)and support multiple types of input data such as SAC,MSEED,NumPy array,etc.A test on the dataset of the Wenchuan aftershocks shows the generalization ability and detection accuracy of the software.The software is expected to increase the efficiency and subjectivity in the manual processing of large amounts of seismic data,thereby providing convenience to regional network monitoring staffs and researchers in the study of Earth's interior.

On some problems of seismic crustal phase

In this paper, some practical problems in seismological observations are discussed on the identification of seismiccrustal phase, such as: “whether P (Pg) and S(Sg) are direct waves or not?”; “whether S(Sg) and Lg as well asng and P (Pg) are two kinds of waves or identical wave?”;“on seismoiogicai phases in regional travel-timetable”;“on the symbols and identifications of phases in seismological observation repors”;“on the relation between mLg and ML”, and so on. Some confused ideas on these problems are clarified.

Numerical modeling of global seismic phases and its application in seismic phase identification

Earthquake data include informative seismic phases that require identification for imaging the Earth's structural interior.In order to identify the phases,we created a numerical method to calculate the traveltimes and raypaths by a shooting technique based upon the IASP91 Earth model,and it can calculate the traveltimes and raypaths for not only the seismic phases in the traditional traveltime tables such as IASP91,AK135,but also some phases such as pPcP,pPKIKP,and PPPPP.It is not necessary for this method to mesh the Earth model,and the results from the numerical modeling and its application show that the absolute differences between the calculated and theoretical traveltimes from the ISAP91 tables are less than 0.1 s.Thus,it is simple in manipulation and fast in computation,and can provide a reliable theoretical prediction for the identification of a seismic phase within the acquired earthquake data.

DiTing:A large-scale Chinese seismic benchmark dataset for artificial intelligence in seismology

In recent years,artificial intelligence technology has exhibited great potential in seismic signal recognition,setting off a new wave of research.Vast amounts of high-quality labeled data are required to develop and apply artificial intelligence in seismology research.In this study,based on the 2013–2020 seismic cataloging reports of the China Earthquake Networks Center,we constructed an artificial intelligence seismological training dataset(“DiTing”)with the largest known total time length.Data were recorded using broadband and short-period seismometers.The obtained dataset included 2,734,748 threecomponent waveform traces from 787,010 regional seismic events,the corresponding P-and S-phase arrival time labels,and 641,025 P-wave first-motion polarity labels.All waveforms were sampled at 50 Hz and cut to a time length of 180 s starting from a random number of seconds before the occurrence of an earthquake.Each three-component waveform contained a considerable amount of descriptive information,such as the epicentral distance,back azimuth,and signal-to-noise ratios.The magnitudes of seismic events,epicentral distance,signal-to-noise ratio of P-wave data,and signal-to-noise ratio of S-wave data ranged from 0 to 7.7,0 to 330 km,–0.05 to 5.31 dB,and–0.05 to 4.73 dB,respectively.The dataset compiled in this study can serve as a high-quality benchmark for machine learning model development and data-driven seismological research on earthquake detection,seismic phase picking,first-motion polarity determination,earthquake magnitude prediction,early warning systems,and strong ground-motion prediction.Such research will further promote the development and application of artificial intelligence in seismology.

Blockly earthquake transformer:A deep learning platform for custom phase picking

Deep-learning(DL)algorithms are increasingly used for routine seismic data processing tasks,including seismic event detection and phase arrival picking.Despite many examples of the remarkable performance of existing(i.e.,pre-trained)deep-learning detector/picker models,there are still some cases where the direct applications of such models do not generalize well.In such cases,substantial effort is required to improve the performance by either developing a new model or fine-tuning an existing one.To address this challenge,we present Blockly Earthquake Transformer(BET),a deep-learning platform for efficient customization of deep-learning phase pickers.BET implements Earthquake Transformer as its baseline model,and offers transfer learning and fine-tuning extensions.BET provides an interactive dashboard to customize a model based on a particular dataset.Once the parameters are specified,BET executes the corresponding phase-picking task without direct user interaction with the base code.Within the transfer-learning module,BET extends the application of a deep-learning P and S phase picker to more specific phases(e.g.,Pn,Pg,Sn and Sg phases).In the fine-tuning module,the model performance is enhanced by customizing the model architecture.This no-code platform is designed to quickly deploy reusable workflows,build customized models,visualize training processes,and produce publishable figures in a lightweight,interactive,and open-source Python toolbox.

A study on small magnitude seismic phase identification using 1D deep residual neural network

Reliable seismic phase identification is often challenging especially in the circumstances of low-magnitude events or poor signal-to-noise ratio.With improved seismometers and better global coverage,a sharp increase in the volume of recorded seismic data has been achieved.This makes handling seismic data rather daunting by using traditional approaches and therefore fuels the need for more robust and reliable methods.In this study,we develop 1D deep Residual Neural Network(ResNet),for tackling the problem of seismic signal detection and phase identification.This method is trained and tested on the dataset recorded by the Southern California Seismic Network.Results demonstrate that the proposed method can achieve robust performance for the detection of seismic signals and the identification of seismic phases.Compared to previously proposed deep learning methods,the introduced framework achieves around 4%improvement in earthquake detection and a slightly better performance in seismic phase identification on the dataset recorded by Southern California Earthquake Data Center.The model generalizability is also tested further on the STanford EArthquake Dataset.In addition,the experimental result on the same subset of the STanford EArthquake Dataset,when masked by different noise levels,demonstrates the model’s robustness in identifying the seismic phases of small magnitude.

Research on model testing of a new ultrasonic multi-wav detecting method

The traditional crack exploring method with echo （reflected wave） in metals is called the ＂single-wave detecting method＂ that uses a probe of single weight. This method is not able to detect directly the size and shape of the crack and the result can only be obtained by relative comparison, that is to compare the echo amplitudes of the unknown quantity （crack） with the known quantity （regular artificial crack） to determine the equivalent size and shape of a certain crack.

A Study on the Early Warning Time of Strong Earthquakes in Taiwan

The paper collects the records by the Fujian Digital Seismic Network of 40 shallow earthquakes in Taiwan with M_S≥5.0 from 1999 to 2013,analyzes the seismic phase(Pn,Sn phase)characteristics and travel-time rules,determines travel-time models and develops a seismic phase travel-time equation based on the two-step fitting algorithm.With the deduction of processing time and network delay time,this method can provide an accurate estimation of early warning time of Taiwan earthquakes for the Fujian region,and has been officially employed in the earthquake early warning system of Fujian Province.

Seismic response analysis of organic reef in gentle slope of rift basin

Analyzing the formation and sediment characteristics of gentle slope, the authors elaborate formation mechanism of organic reef and characteristics of reservoir in gentle slope of rift basin. Using the forward model of seismic exploration, the study provides the objective judgment for the exploration of organic reef reservoir in gentle slope of rift basin.

Determination Method of Focal Depth of Local Earthquake Using the Travel-time Difference between Pn and sPn

In this paper,we derived the relationships between the travel time difference of sPn and Pn and the local earthquake focal depth.In these equations,the travel time difference of sPn and Pn is not related to the epicentral distance,but depends only on the regional crustal mode and the focal depth.According to the equations,we provided a simple and accurate method to determine local earthquake focal depth by using the travel time difference between phase sPn and Pn.This method has been used to determine the focal depths of two earthquake of MS6.1 and MS5.6 which occurred at the junction of Panzhihua and Huili,Sichuan on August 30 and 31,2008.The results were compared to those from other sources such as the China Earthquake Networks Center,and the comparison shows that the results are accurate and reliable.