Arrival Times Of Seismic Waves Research Articles

Stochastic simulation of earthquake ground motions based on improved finite-fault model

In the seismic reliability analysis of engineering structures, it is often necessary to adopt stochastic ground motions as inputs. However, the current methods used to generate stochastic samples, such as the spectral representation method, the spectral transfer method, the dynamic source method, the physical source method, and the comprehensive method, etc., are not very convincing since they do not contain the complete information of seismic occurrence and propagation. In this study, an improved finite-fault model for stochastic simulation of ground motions is proposed. Compared to existing stochastic finite-fault model, the improved model considers the propagation law of phase spectrum, local site effect, and the impact of wave velocity on the arrival time of seismic waves. Besides, identification and statistics of model parameters are carried out based on ground motion records. Reasonable samples of stochastic ground motions are then obtained. In addition, parameter sensitivity analysis is conducted to determine critical random variables, which reduces the complexity of stochastic simulation. Simulation results are compared with recorded accelerograms from the 1999 Chi-Chi earthquake. It is shown that the improved stochastic finite-fault model is reliable, and the acceleration response spectra of simulated samples are consistent with those of station records collected at various site classes and the design spectra.

Automated Detection of Microseismic Arrival Based on Convolutional Neural Networks

It is difficult to accurately and efficiently detect seismic wave signals at the time of arrival for automatic positioning from microseismic waves. A U-net model to detect the arrival time of seismic waves is constructed based on the convolutional neural network (CNN) theory. The original data for 1555 segments and synthetic data of 7764 segments were detected using Akaike’s information criterion (AIC) algorithm, the time window energy eigenvalue algorithm, and the U-net model. During uniaxial compression of the test block, acoustic emission equipment is used to collect the vibration wave generated by the rupture of the test block. Source imaging images are drawn using the Origin software, the arrival time error is counted, and the advantages and disadvantages of the three arrival time methods are discussed. Similarities between the source image and the actual fracture image are observed. There is a high similarity between the source imaging map and the physical trajectory map when the U-net model is used. Thus, it is feasible to use the U-net model to detect the arrival time of seismic waves. Its accuracy is greater than that of the time window energy eigenvalue algorithm but lower than that of the AIC algorithm for high signal-to-noise ratios. After reducing the signal-to-noise ratio, the stability and accuracy of the U-net model to detect the arrival time have improved over the other two algorithms.

Traveltime-based microseismic event location using artificial neural network

Location of earthquakes is a primary task in seismology and microseismic monitoring, essential for almost any further analysis. Earthquake hypocenters can be determined by the inversion of arrival times of seismic waves observed at seismic stations, which is a non-linear inverse problem. Growing amounts of seismic data and real-time processing requirements imply the use of robust machine learning applications for characterization of seismicity. Convolutional neural networks have been proposed for hypocenter determination assuming training on previously processed seismic event catalogs. We propose an alternative machine learning approach, which does not require any pre-existing observations, except a velocity model. This is particularly important for microseismic monitoring when labeled seismic events are not available due to lack of seismicity before monitoring commenced (e.g., induced seismicity). The proposed algorithm is based on a feed-forward neural network trained on synthetic arrival times. Once trained, the neural network can be deployed for fast location of seismic events using observed P-wave (or S-wave) arrival times. We benchmark the neural network method against the conventional location technique and show that the new approach provides the same or better location accuracy. We study the sensitivity of the proposed method to the training dataset, noise in the arrival times of the detected events, and the size of the monitoring network. Finally, we apply the method to real microseismic monitoring data and show that it is able to deal with missing arrival times in efficient way with the help of fine tuning and early stopping. This is achieved by re-training the neural network for each individual set of picked arrivals. To reduce the training time we used previously determined weights and fine tune them. This allows us to obtain hypocenter locations in near real-time.

A new solution to estimate the time delay on the topographic site using time domain 3D boundary element method

This study focuses on investigating spatial variation of ground motion that has great influence on the dynamic behavior of the large structures located on the surface topography. One of the most effective parameters on the spatial variation of ground motion is the difference between the arrival time of seismic waves in different points located on the abutments. In this research, a three-dimensional model of the Pacoima Dam site is prepared. The time domain 3D boundary element method is used to apply non-uniform excitation at the dam supports. This model is subjected to vertically propagating incident SH and P waves. The time delay can be characterized by calculating the value of the time delay for which the cross-correlation between two records is maximized. Finally, to obtain the time delay in a topographic site, a function considering effective parameters such as the height from the canyon base, wave velocity and predominant frequency, is presented. Furthermore, a code was developed for generating the spatially variation of seismic ground motions. The results show that the proposed functions have an acceptable accuracy in estimating the time delay to generate non-uniform ground motion.

Arrival-time picking uncertainty: Theoretical estimations and their application to microseismic data

Traveltime-based methods depend on the accurate determination of the arrival times of seismic waves. They further benefit from information on the uncertainty with which the arrival times are determined. Among other applications, arrival-time uncertainties are used to weight data in inversion algorithms and to define the resolution of reconstructed velocity models. The most physically meaningful approaches for the estimation of arrival-time uncertainties are based on probabilistic formulations. The two approaches for the assessment of the lower bound of arrival-time uncertainties, the Cramér–Rao Bound (CRB) and the Ziv–Zakai Bound (ZZB), have been reviewed. The CRB determines the minimum-achievable estimation error under the assumption of a high signal-to-noise ratio (S/N) but underestimates said error for small S/N. The ZZB provides a better result for noisy data because it utilizes a priori information. The CRB and ZZB require knowledge of the spectral variance of the signal, which often is hard to determine in seismic experiments. Furthermore, both bounds assume additive white Gaussian noise (AWGN), which does not hold for seismic data. To overcome these problems, alternative expressions have been proposed, which yield comparable estimates as CRB and ZZB but are solely based on the S/N and the dominant period in the data. Moreover, a recipe to correct the S/N and account for the difference between the seismic noise and AWGN has been provided. For a case study of downhole microseismic monitoring, it is determined that the new expressions provide station-dependent arrival-time uncertainties, which are used as weights to improve source location uncertainties.

DETERMINATION OF EARTHQUAKE FOCUS COORDINATES USING THE CASSINI OVAL METHOD WITH SECOND- AND FOURTH-ORDER HYPERBOLA FIGURES

Objectives. To study the dependencies obtained when determining the coordinates of an earthquake hypocentre using the figures of fourth and second orders.Method. A comparative analysis of determining the coordinates of the earthquake focus using the Cassini oval method, both taking errors in the readings of seismic sensors into account the and ignoring them, is presented.Result. A new method is proposed for determining the coordinates of the earthquake hypocentre, which uses the fourth-order figure, the Cassini oval, in the calculations. A graph is obtained for the distribution of errors in determining the coordinates of the earthquake focus (using the Cassini oval) depending on the relative position of two seismic sensors with different values of their errors in determining the difference in travel times of seismic waves.Conclusion. Since the calculation results are independent of the error sign in determining the difference in the arrival times of seismic waves, the method is suitable for the initial determination of the coordinates of the earthquake hypocentre as well as for comparison with the results of other methods for identifying the error sign.

First Arrival Time Identification Using Transfer Learning With Continuous Wavelet Transform Feature Images

In our work, the deep learning technique has been used to develop an automatic method for identifying the first arrival times of seismic waves. This method introduces transfer learning to train a deep neural network, given a limited number of continuous wavelet transform (CWT) feature images as input. The application of the CWT for feature extraction, aimed at detecting abrupt changes in the amplitude, phase, and frequency produced by first arrivals as a whole rather than any single characteristic, provides the most informative images. First, we apply the CWT to each seismic trace to obtain the CWT feature images and split them into a set of subimages. Then, a pretrained convolutional neural network (CNN) is fine-tuned with limited labeled subimages. The resulting model can be used to predict probability distributions of noise, first-break, and post first-break. Finally, the first arrival times are extracted from the peaks of the probability distributions. We have tested the performance of the method using vibroseis, dynamite, and air gun shot records, which include various types of seismic waves and noise. More accurate and robust results can be obtained with the proposed method compared with the short-time and long-time average (STA/LTA) algorithm and the adaptive multiband picking algorithm (AMPA).

About seismic observations on Sakhalin with the use of molecular-electronic seismic sensors of new type

To conduct complex geophysical observations in the area of the Central Sakhalin Fault, an experimental geophysical test site was created, including a broadband molecular electronic type station and a new development of the Molecular Electronics Center - molecular electronic hydrophone. A module for placing a molecular-electronic seismometer at the test site has been developed and arranged. The noise level of the seismometer was analyzed and compared with the reference data. The results of hydroacoustic observations of weak earthquakes in the area of the Central-Sakhalin fault using a molecular-electron hydrophone are presented. The characteristics of the recorded acoustic emission signals from earthquakes perceived by the hydrophone station (frequency spectrum, arrival times of seismic waves) are investigated. The issues of operation and maintenance of molecular-electronic sensors with seismic signal data logger at various stages of seismological observations (setting up equipment, acquisition and storing data) were considered. The analysis of recording capabilities of seismic equipment was performed. Conclusions about the advantages and disadvantages of molecular-electronic sensors and their use on Sakhalin are made.

Spatial Distribution of Crustal Earthquakes in the Caucasus

Previous results on the location of crustal earthquake foci in the Caucasus clearly show that the methods used in seismic service practice to determine the hypocenters of earthquakes do not allow accurate localization of events and can hardly be used to forecast earthquakes or study the structure of the Earth’s crust and upper mantle in detail. The arrival times of seismic waves often have large errors that prevent qualitative processing of observations and, therefore, is is difficult to obtain reliable results. In addition, the use of only one velocity curve or one set of travel-time curves for all stations of the network hinders accurate determination of the location of earthquake hypocenters. In order to localize earthquake foci with high precision, it is also necessary to have a sufficiently detailed representation of the structure of the Earth’s crust and upper mantle in the studied region. It is also desirable to take information about the structure of the region not from earthquake data, but based on other observations, e.g., deep seismic sounding (DSS) data. Using seismological bulletins for 1970–2015, we have redetermined the hypocentral coordinates of crustal earthquakes in the Caucasus. The new distribution of crustal earthquake hypocenters in the Caucasus differs significantly from the old distribution obtained from the catalog. This may become the basis for refining ideas about the regional geodynamics.

APASVO: A free software tool for automatic P-phase picking and event detection in seismic traces

The accurate estimation of the arrival time of seismic waves or picking is a problem of major interest in seismic research given its relevance in many seismological applications, such as earthquake source location and active seismic tomography. In the last decades, several automatic picking methods have been proposed with the ultimate goal of implementing picking algorithms whose results are comparable to those obtained by manual picking. In order to facilitate the use of these automated methods in the analysis of seismic traces, this paper presents a new free, open source, software graphical tool, named APASVO, which allows picking tasks in an easy and user-friendly way. The tool also provides event detection functionality, where a relatively imprecise estimation of the onset time is sufficient. The application implements the STA-LTA detection algorithm and the AMPA picking algorithm. An autoregressive AIC-based picking method can also be applied. Besides, this graphical tool is complemented with two additional command line tools, an event picking tool and a synthetic earthquake generator. APASVO is a multiplatform tool that works on Windows, Linux and OS X. The application can process data in a large variety of file formats. It is implemented in Python and relies on well-known scientific computing packages such as ObsPy, NumPy, SciPy and Matplotlib.

Shear Velocity Structure of Abyssal Plain Sediments in Cascadia

The sediments that form the upper layer of the oceanic crust vary significantly in thickness. As the crust ages, it accumulates more and more sediment, and when it is near a continental shelf, it can receive large deposits of continental sediment deposited by turbidity currents forming abyssal plains. Because of their soft, unconsolidated nature, the uppermost sediments tend to have very low shear velocities. These low velocities create significant delays in the arrival times of seismic waves moving through them, and these delay times can provide a significant source of error for body‐wave tomography. For surface‐wave tomography, these soft sediments can also provide a source of error by disproportionately slowing short‐period waves in a layer too thin to independently resolve. Correcting for sediments can significantly improve both body‐ and surface‐wave tomography. A deployment of ocean‐bottom seismometer (OBS) stations around the Juan de Fuca plate, the Cascadia Initiative is being deployed in four phases (Fig. 1; Toomey et al. , 2014). Year one focused on the northern portion of Juan de Fuca plate, and year two focused on the southern portion of the Juan de Fuca plate, including Gorda. Year two was coupled with the Blanco experiment, an independent National Science Foundation‐funded experiment, five stations of which have generously been made available to us by principal investigator John Nabelek. Year‐three stations largely reoccupied year‐one sites. Year‐four stations are currently deployed and their data are not yet available for analysis. Figure 1. A contour map of delay times. For each station identified as sitting on topography rising above the abyssal plains, the delay time was set at zero, and tie points with zero delay times were placed along the ridges. The triangles represent stations where we measured a sediment thickness, squares represent nonabyssal plain stations where we could not measure a sediment thickness, and …

Station Travel-Time Calibration Method Improves Location Accuracy on a Global Scale

Online Material: Earthquake catalog and tables of station and data parameters. Earthquake location has been an outstanding issue for seismologists for over more than a century, and it is still an important subject of study. In fact, accurate locations of earthquakes are a requisite for investigations on the dynamics of the Earth and seismogenic processes. Another possible use of seismic locations is, for example, monitoring underground explosions in the context of the International Monitoring System (IMS) for verifying compliance to the Comprehensive Test Ban Treaty (CTBT). In fact, the CTBT makes provisions for On‐Site Inspections (OSI) requested by a member State that detects a suspicious event in the territory of another member State. As the CTBT states that the OSI area should not exceed a size of 1000 km2, it is crucial that the seismic event triggering the OSI be known with an uncertainty not exceeding a few km. Earthquake location accuracy is related to the degree of knowledge about the 3D structure of seismic‐wave velocity in the Earth. In fact, the location process is essentially the solution of an inverse problem in which observed arrival times of seismic waves are compared with theoretical arrival times computed from a given velocity model. Assuming such a velocity model is known, the unknown parameters of this problem are the origin time and the three spatial coordinates of the earthquake source. It is generally recognized that, and independently of the location algorithm, uncertainties in earthquake locations are dominated by two main factors (Husen and Hardebeck, 2010): 1. Measurement errors of seismic arrival times, and 2. Modeling errors of calculated travel times. Among numerous location methods developed by seismologists, the group of algorithms is very popular that is based on linearization of the inherently non‐linear inverse problem, and solution of the over‐determined problem by a …

Analysis of signals of a borehole strainmeter in the western rift of Corinth, Greece

This paper presents the orst analysis of the records of an elliptical 3-component Sacks-Evertson borehole strainmeter. This highresolution prototype by the Carnegie Institution of Washington, is installed since 2006 in the western rift of Corinth, Greece. We orst present the calibration and the correction from external inouences, in order to quantify the detection level of the instrument. We show evidence for pore pressure diffusion from the sea, mostly affecting one component. Neglecting this effect, a orst order correction reduces the signal by 90% at tidal periods for 2 components and about 70% for the third one. The residual noise vary from 1 nstrain at 1-hour period to 10 nstrain at 1-day period. It allows to detect slow earthquakes lasting 1 day down to magnitude 4 at an hypocentral distance of 8 kilometers. The uncorrected records at periods smaller than semidiurnal does not reveal any slow strain transient with strong amplitude. During the closest seismic swarm to the site in 2011, the analysis of the records reveals strain steps occuring at the arrival times of seismic waves radiated by the local earthquakes, uncorrelated with the amplitudes and mostly related to dynamic pore pressure instabilities.

Hydroacoustic observations of weak earthquakes in shallow waters of the Southern Kuril Islands

Results of hydroacoustic observations of signals from weak earthquakes in natural conditions in the region of the Southern Kuril Islands are presented. Some earthquakes were registered by the the Yuzhno-Kurilsk Seismic Station, other were only recorded by hydrophone stations. The observations were specific as seismic signals were recorded in shallow waters, i.e. in high noise level conditions. Hydrophones were installed in Lake Lagunnoe (Kunashir) and Khromovaya Bay (Shikotan). Our analysis of hydroacoustic records received from the hydrophone stations revealed no evident precursory response of the geological medium to weak distant events. This means that neither before the period of earthquake preparation nor during the earthquake preparation period, any geoacoustic emission was not detected. It is shown that despite the unfavourable noise level conditions, even distant weak earthquakes can be confidently registered by hydrophone stations, and pending application of proper signal processing techniques, it can be possible to determine arrival times of seismic waves and to measure parameters of seismic waves. It is also established that the frequency spectrum of acoustic signals from the weak earthquakes recordable by the hydrophone stations is continuous and of noise type in the frequency range up to 90–100 Hz. It is revealed that in some cases, weak earthquakes and microearthquakes may be forerun by low frequency signals.

Estimation of arrival times from seismic waves: a manifold-based approach*

We propose a new method to analyse seismic time-series and estimate the arrival times of seismic waves. Our approach combines two ingredients: the time-series are first lifted into a high-dimensional space using time-delay embedding; the resulting phase space is then parametrized using a non-linear method based on the eigenvectors of the graph Laplacian. We validate our approach using a data set of seismic events that occurred in Idaho, Montana, Wyoming and Utah between 2005 and 2006. Our approach outperforms methods based on singular-spectrum analysis, wavelet analysis and short-term average/long-term average (STA/LTA).

Automatic seismic wave arrival detection and picking with stationary analysis: Application of the KM2O-Langevin equations

AbstractAn automatic detection and a precise picking of the arrival times of seismic waves using digital seismograms are important for earthquake early detection systems. Here we suggest a new method for detecting and pickingP-andS-wave signals automatically. Compared to methods currently in use, our method requires fewer assumption with properties of the data time series. We divide a record into intervals of equal lengths and check the “local and weak stationarity” of each interval using the theory of the KM2O-Langevin equations. The intervals are stationary when these include only background noise, but the stationarity breaks abruptly when a seismic signal arrives and the intervals include both the background noise and theP-wave. This break of stationarity makes us possible to detectP-wave arrival. We expand the method for picking ofS-waves. We applied our method to earthquake data from Hi-net Japan, and 90% ofP-wave auto-picks were found to be within 0.1 s of the corresponding manual picks, and 70% ofS-wave picks were within 0.1 s of the manual picks. This means that our method is accurate enough to use as a part of the seismic early detection system.

Seismic velocity reductions caused by the 2003 Tokachi‐Oki earthquake

We use four repeating earthquake sequences located near Hokkaido to identify velocity changes caused by the Mw 8.0 2003 Tokachi‐Oki earthquake. Using a moving window cross‐correlation technique, we identify delays in the arrival time of seismic waves that accumulate linearly with time into the seismogram. This behavior is indicative of multiple scattering within a medium where the seismic velocity has been reduced. For all of our earthquake‐receiver geometries, we find evidence of significant velocity reductions close to the receiver. The correlation of the size of the velocity reductions with both strong shaking and site characteristics suggests that these velocity reductions are caused by damage to near‐surface materials created by nonlinear strong ground motion. For earthquake/receiver geometries where the seismic waves cross the Tokachi‐Oki rupture zone, we identify particularly large reductions in velocity as a result of the earthquake. For these geometries, we believe that the rupture zone of the Tokachi‐Oki earthquake or the shallow crust above it represents a second region where seismic velocities are reduced as a result of the main shock.

Local earthquake tomography of the Altiplano and the Eastern Cordillera of northern Bolivia

The velocity structure in the crust beneath the central Andes of northern Bolivia is obtained by the simultaneous inversion of local earthquake arrival-time data for velocity and hypocentral parameters. The data were recorded between October 1990 and February 1991 by the French “Lithoscope” network; the 34 vertical, short-period seismic stations were installed along a 320-km-long profile crossing two main morphological units of the Andean chain; the Altiplano and the Eastern Cordillera. The arrival times of seismic waves generated by earthquakes located in the subduction zone of the Nazca plate underneath the South American plate were inverted using a modified version of Thurber's 3-D iterative simultaneous inversion code. The results are consistent with the inversion of teleseismic travel-time residuals from the same network and closely correlate with the mapped surface geological features. The Altiplano Basin is characterized by a low-velocity zone down to 20 km, enclosed by the San Andres and Laurani faults. At the surface, the Cordillera Real fault zone (CRFZ) bounds to the west the high P-wave velocity anomaly lying under the axial zone of the Eastern Cordillera. The velocity discontinuity dips steeply to the southwest, to the bottom of the crust. From the velocity model, the depth of the Moho is estimated to be 66 km under the Altiplano. Moreover, the different behaviour of P- and S-waves found below the Western Cordillera and above the slab is consistent with the presence of partially melted asthenosphere that may be responsible for the magmatic activity. Finally, the seismicity defines, between 90 and 225 km, a part of the slab dipping to the north-northeast with a low angle of about 30°.

Internal structure of Piton de la Fournaise volcano from seismic wave propagation and earthquake distribution

Arrival times of seismic waves from local earthquakes are inverted for both locating the source and defining the 3-D velocity heterogeneity of Piton de la Fournaise. The lateral heterogeneity of the 2632 m high edifice is resolved as a high-velocity plug, 1.5 km in diameter, surrounded by a low-velocity ring, which may be interpreted as due to the construction of Fournaise on the flank of the older volcano Piton des Neiges. Wave mode conversion detected on three-component seismograms provides evidence for boundaries of contrasted velocities. Pre-eruptive swarm earthquakes cluster in the high-velocity zone, under the Dolomieu summit crater. Low strength and cohesion of the surrounding material account for the lack of seismicity for the final 1–3 km radial flow of magma to the vents in Enclos Fouqué. Beneath the high-velocity plug the existence of a body with low velocity for P, and even for S, waves is well constrained. However, the walls and base are poorly defined because of the lack of deep earthquakes for sampling. The few earthquakes that are located in this depth region usually occur at a depth of around 1.5 km below sea level in the region of the cone. This can be considered providing the upper constraint on the lower limit of the aseismic part of the low-velocity body. The coincidence in time of their occurrence with the swarms above sea level and the eruptions suggests magmatic activation of the low-velocity aseismic volume 1.5 km below sea level under the high-velocity plug of the cone. Further down, the concentration of seismicity in two swarms, between 2 and 4 km, under the eastern flank does not allow the structure to be sampled effectively.

Effects of seismic wave propagation on a long and narrow building: Body wave and surface wave propagation

In the case of the dynamic analysis of the structures using the recorded earthquake ground motions, it is usually assumed that the ground motion consists of body waves propagating vertically. However, the response of a long and narrow structure may be influenced by the oblique propagation of body waves and the dispersion of surface waves. In this paper, the effects of the seismic wave propagation on the response of this kind of structure are investigated. The characteristics of the wave propagation were verified using the recorded motions and soil information at the building site. The ground motion at every input point of the building was evaluated using the difference of arrival time of seismic waves calculated by assuming the velocity and the direction of the wave propagation. Using these ground motions, response analyses of the lumped mass model for the structure were performed. By considering the characteristics of the seismic wave propagation, the average response decreased but the local response increased around the end of the roof. Further studies of the structure were also performed in order to restrain the response around the end of the roof.