Figure - available from: Journal of Seismology
This content is subject to copyright. Terms and conditions apply.
1999 Chi-Chi Taiwan Earthquake (Mw = 7.6), TCU051 Station (rep = 38.53 km) velocity waveform. Red line and blue lines represent the width (Tp) and borders (ts and te) of the pulse region around PGV, respectively. The green line and cyan lines represent the width (Tp,emax) and borders (teemax and tsemax) of the area where the maximum energy is concentrated, respectively. Background image is Ricker wavelet power spectrum of the signal with the same color content of Fig. 1
Source publication
Near fault ground motions may contain impulse behavior on velocity records. Such signals have a particular indicator which makes it possible to distinguish them from non-impulsive signals. These signals have significant effects on structures; therefore, they have been investigated for more than 20 years. In this study, we used Ricker and Morlet wav...
Citations
... In this study, we analyze strong motion records (see Sect. 2) from the Kahramanmaraş earthquakes to detect and analyze the impulsive ground motions. We used the methods of Shahi and Baker (2014), Ertuncay and Costa (2019) as detection algorithms. Shahi and Baker (2014) use the NGA-West 2 database (Ancheta et al. 2014) to define what impulsive motion is and provides the detection algorithm that is explained below. ...
... In this study, we use the algorithms of Shahi and Baker (2014) and Ertuncay and Costa (2019) to detect impulsive motions in Kahramanmaraş earthquakes. Pulse detection algorithms are explained in Sect. ...
... To detect impulsive motions methods of Shahi and Baker (2014) and Ertuncay and Costa (2019) are selected. The methods are neither modified nor updated by using the data collected for this study. ...
On the 6th of February 2023, a large magnitude earthquake (Pazarcık earthquake), $$M_{w}=$$ M w = 7.7, occurred in southeast Türkiye, which caused significant destruction in Türkiye and Syria. Relatively large magnitude aftershocks followed the main shock, and after 9 hours of the main event, another large magnitude earthquake (Elbistan earthquake) occurred, $$M_{w}=$$ M w = 7.6, on a nearby fault. This study analyzes the near-fault seismic signals from earthquakes larger than 5.5 recorded between the main shock and the 31st of March 2023. More than 60 impulsive motions are detected in 3 earthquakes, mostly concentrated in the Pazarcık and Elbistan earthquakes. In the Pazarcık earthquake, many impulsive motions are recorded in near-fault stations with periods of up to 14 s. In contrast, in the Elbistan earthquake, impulsive motions are spatially distributed, with pulse periods of up to 11 s and at distances greater than 150 km. Pulse periods mostly correlate with the magnitude of the earthquake, but pulse probability models do not predict impulsive motions over long distances. The presence of strong impulsive motions in vertical components is also observed. For both earthquakes, peak ground velocities (PGVs) are larger than predicted by ground motion prediction equations. The observation of long-period, large amplitude signals may indicate the presence of a directivity effect for both earthquakes. In some stations, spectral periods exceed the 2018 Turkish building design codes for long periods ( $$\ge$$ ≥ 1 s).
... Numerous studies have been performed to examine the correlation between different parameters of ground motion and to develop predictive equations for the classification of ground motion such as Baker (2007), Ertuncay and Costa (2019), Kardoutsou et al. (2017), Panella et al. (2017), and Zhai et al. (2013). Baker employed the Daubechies wavelet of order 4 as the mother wavelet to extract the largest velocity pulse. ...
... The box plot in Fig. 3 explains that (M w , R jb , IP, and PGV ratio) are strongly sensitive to the classification of ground motion. However, it should be emphasized that prior research has indicated Graphical representation of five-fold cross-validation accuracy score of different machine learning models J Seismol a substantial correlation between PGV ratio, energy ratio, development length, and PGV with PL ground motion (Baker 2007;Zhai et al. 2013;Kardoutsou et al. 2017;Ertuncay and Costa 2019;Habib et al. 2022;Mavroeidis andPapageorgiou 2002, 2003;Panella et al. 2017). Additionally, from this study, we observed the relationships between the source parameter and earthquake magnitude with PL ground motions. ...
Machine learning approaches are increasingly being employed to forecast the key characteristics of strong ground motions, including the challenging classification of Pulse-Like (PL) ground motions. The PL ground motions are characterized by their impulsive nature and have the potential to cause significant damage to structures. The classification of PL ground motions continues to be a significant challenge due to the absence of consensus on their definition and categorization. This paper investigates the potential benefits of several Machine Learning Classifiers (MLCs) algorithms such as decision tree, random forest, logistic regression, naive Bayes, support vector machine, K-nearest neighbor, ensemble model, and artificial neural network models for predicting PL and Non-Pulse-Like (NPL) ground motions. In this regard, a dataset comprising 200 near-fault ground motions records compiled from active tectonic regions like Taiwan, Turkey, Iran, and Japan, was divided into 2 portions, with 75% used to train the model and the remaining 25% used for testing. Plots of performance metrics, confusion matrix, and receiver operating curve indicate that the ensemble classifier outperforms the other classifier with 86.2% accuracy and the lowest misclassification of PL and NPL ground motions. Additionally, the trained MLC has been compared with the existing ground motion classification models to further assess the accuracy of the different classifiers in the present study.
... Therefore, such ground motions have different characteristics than pulse-like ground motions with directivity and fling step effects. To explore impulsiveness, Ertuncay and Costa (2019) used Ricker and Morlet wavelets to assess the wavelet power spectrum of strong ground motion data. Recently, Pavel (2021) used wavelet method to extract velocity pulses of pulse-like ground motion recordings from Vrancea intermediate-depth earthquakes. ...
Pulse-like ground motions may have only a distinct strong pulse or multiple pulses within the velocity time-history. These intrinsic pulses are hidden in low-frequency components that can impose extreme seismic demands on structures. This study presents a simple approach based on the empirical Fourier decomposition (EFD) to extract the intrinsic pulses of pulse-like ground motions. Based on the proposed approach, first, the original ground velocity is decomposed into several Fourier spectrum components (FSCs) via the EFD method. Among these FSCs, the significant low-frequency components are identified based on a proposed relative energy indicator (Ere). Ere is defined as the ratio of the Fourier square amplitude of the FSC to that of the original ground velocity. Next, pulse component is obtained by superimposing the minimum number of significant low-frequency components so that their total energy is above than 70% of the original ground velocity energy. Finally, the strong pulse is extracted from the pulse component by the peak point method. Results obtained by the EFD decomposed method are compared to those obtained from wavelet method for 91 pulse-like ground motions. The results show that the proposed method can extract the intrinsic pulses of pulse-like ground motions with reasonable accuracy. The proposed approach is further applied for classification of near-fault pulse-like ground motions in a dataset of ground motion records. According to the classification results, ground motions with a relative energy value greater than 0.30 can be characterized as pulse-like.
Highlights
• The intrinsic pulses of pulse like ground motions are extracted based on the empirical Fourier decomposition (EFD) technique.
• The Fourier component with the high relative energy can be considered as significant low-frequency component.
• Ground motions with Ep values above 0.30 are considered as pulse-like ground motions.
... The ranges of significant recorded values for the above-mentioned parameters are reported in Table 2, while the details of the GMPEs calculation are reported in a paper that is under review in an international journal. Ertuncay and Costa [47] used the RAN database along with other databases around the globe to develop an algorithm to detect impulsive motions in seismic records. Impulsive motions may occur due to directivity [48] and fling step [49] effects along with local soil conditions [50]. ...
A strong motion monitoring network records data that provide an excellent way to study how source, path, and site effects influence the ground motion, specifically in the near-source area. Such data are essential for updating seismic hazard maps and consequently building codes and earthquake-resistant design. This paper aims to present the Italian Strong Motion Network (RAN), describing its current status, employment, and further developments. It has 648 stations and is the result of a fruitful co-operation between the Italian government, regions, and local authorities. In fact, the network can be divided into three sub-networks: the Friuli Venezia Giulia Accelerometric Network, the Irpinia Seismic Network, and all the other stations. The Antelope software automatically collects, processes, and archives data in the data acquisition centre in Rome (Italy). The efficiency of the network on a daily basis is today more than 97%. The automatic and fast procedures that run in Antelope for the real-time strong motion data analysis are continuously improved at the University of Trieste: a large set of strong motion parameters and correspondent Ground Motion Prediction Equations allow ground shaking intensity maps to be provided for moderate to strong earthquakes occurring within the Italian territory. These maps and strong motion parameters are included in automatic reports generated for civil protection purposes.
... The use of the proposed model to generate pulse-like near-fault ground motion is illustrated by numerical examples. The model is validated by comparing the pseudo-spectral investigated in Mavroeidis and Papageorgiou (2003), Baker (2007), He and Agrawal (2008), Dickinson and Gavin (2010), Yaghmaei-Sabegh (2010), Zhai et al. (2013), Shahi and Baker (2014), Cork et al. (2016), Kardoutsou et al. (2017), Chang et al. (2019), Ertuncay and Costa (2019), and Whitney (2019). For quantifying and identifying the pulse-like near-fault ground motions, a ground motion record is usually decomposed based on the time-scale (or time-frequency) analysis using the wavelet transforms, and the energy associated with the decomposed signal is ranked. ...
... Baker (2007) proposed a pulse indicator based on the consideration that for a given record component, the pulse arrives earlier, and the amplitude of the velocity pulse is large. The use of wavelets to identify the pulse was also considered in Shahi and Baker (2014), Yaghmaei-Sabegh (2010), Ertuncay and Costa (2019), and Whitney (2019). A modification to the indicator proposed by Baker (2007) was presented in Kardoutsou et al. (2017). ...
Seismic near-fault ground motions differ from those in the far-field. They could contain large amplitude and long period velocity pulses and cause damage to structures. In the present study, we propose a new stochastic model for the pulse-like near-fault ground motions by considering the earthquake magnitude, site-to-source distance, and soil condition. The proposed model consists of the pulse-like (PL) and non-pulse-like (NPL) components. The PL component is represented by using parametric models from literature; the NPL component is characterized using the S-transform. A time- and frequency-dependent power spectral density model for the NPL component is proposed, where the frequency modulation and the amplitude modulation are implicitly considered. Probabilistic predicting equations for evaluating the parameters of the proposed stochastic model are developed by considering the selected ground motions records with pulse-like motions extracted from Next Generation Attenuation–West 2 (https://ngawest2.berkeley.edu/) database. The use of the proposed model to generate pulse-like near-fault ground motion is illustrated by numerical examples. The model is validated by comparing the pseudo-spectral acceleration of the simulated records and actual records.
... Any scaling function is determined by a filter. erefore, the design of the filter is one of the key points of wavelet analysis, and its frequency characteristics must meet the condition of accurate reconstruction [22]. ...
In this study, a sports parameter acquisition model based on the internet of things and wavelet analysis is studied to improve the accuracy and timeliness of human sports parameter acquisition. A motion parameter acquisition model including a sensing layer, transmission layer, and application layer is designed. The acceleration sensor and temperature sensor in the information acquisition node in the sensing layer are used to collect the motion parameter data, which are uploaded to the application layer by the network in the transmission layer. The received data are denoised by the wavelet analysis method through the data processing unit in this layer and then sent to the ZigBee coordinator for coordination. The results show that the model can achieve the effective acquisition of different sports parameters of different moving objects and analyze the actual movement of moving objects according to the acquisition results. In the acquisition process, the signal burr can be effectively removed, the signal noise can be reduced, the high signal-to-noise ratio signal can be output, and the accuracy of acquisition is improved. It has high timeliness, stable performance, and strong practical application, which can provide an effective guarantee for users to monitor sports parameter data in real time.
... Baker utilized a wavelet-based method to extract the largest velocity pulse signal [18], and the amplitude and energy of the pulse signal are key parameters for the classification of pulse-like ground motions. This method has a high degree of quantization, but it is greatly affected by the type of mother wavelet [19][20][21]. Due to the complexity of ground motion records, it is impossible to extract all near-fault pulses effectively with one mother wavelet. The waveform and amplitude of the pulse signal based on different mother wavelets are quite different, which will affect the classification of near-fault ground motion and the estimation of the pulse period. ...
Near-fault ground motions have distinctive features different from typical far-field ground motions. The approach detects near-fault pulse-like ground motion based on the variational mode decomposition (VMD) technique. First, the original ground motion can be decomposed into a series of intrinsic mode functions (IMFs) with the VMD technique. The low-frequency signal is obtained by synthesizing the IMFs with the center frequency lower than an adaptive frequency threshold proposed in this paper. Second, the pulse signal and the pulse period are obtained from the best-fitting low-frequency signal by the peak-point method. Finally, 596 records are utilized to calibrate the classification standard. It is concluded that these ground motions with relative energy indicator greater than 0.27 can be classified as pulse-like. The approach is further used to identify the pulse-like ground motion with a significant velocity pulse at the beginning, which may be caused by the forward-directivity effects.
... Chang et al. [22] used energy-based classification using the energy ratio around the peak ground velocity (PGV) location and the total energy of the signal. Ertuncay and Costa [23] used Ricker and Morlet wavelets to analyse and determine impulsive signals. An energy-based classification is implemented in both the velocity time history and wavelet power spectrum. ...
... Signals are analysed with three different algorithms for the determination of impulsive signals. The number of waveforms labelled as impulsive by Shahi and Baker [17], Chang et al. [22], Ertuncay and Costa [23] are 405, 454, and 438, respectively. Shahi and Baker [17] uses two multi-component ground motion data, whereas other algorithms use a single component. ...
... The idea behind the manual labelling is to overcome different criteria between previous studies. For instance, Chang et al. [22], Ertuncay and Costa [23] use the threshold of 30 cm s −1 for PGV, which is implemented by Baker [21], which is defined for the potential damaging effect of such amplitudes on structures. Shahi and Baker [17] uses a pulse indicator (PI) to detect impulsive signals that have the impulsive part at the signal's beginning. ...
Ground motions recorded in near-fault regions may contain pulse-like traces in the velocity domain. Their long periodicity can identify such signals with large amplitudes. Impulsive signals can be hazardous for buildings, creating large demands due to their long periods. In this study, a dataset was collected from various data centres. Initially, all the impulsive signals, which are in reality rare, are manually identified. Furthermore, then, synthetic velocity waveforms are created to increase the number of impulsive signals by using the model developed by Mavroeidis and Papageorgiou, and k−2 kinematic modelling. In accordance, a convolutional neural network (CNN) was trained to detect impulsive signals by using these synthetic impulsive signals and ordinary signals. Furthermore, manually labelled impulsive signals are used to detect the initiation and the termination positions of impulsive signals. To do so, the velocity waveform and position and amplitude information of the maximum and minimum points are used. Once the model detects the positions, the period of the pulse is calculated by analysing spectral periods. Although our detection algorithm works relatively worse than three robust algorithms used for benchmarks, it works significantly better in the determination of initiation and termination positions. At this moment, our models understand the features of the impulsive signals and detect their location without using any thresholds or any formulations that are heavily used in previous studies.
... In order to identify the impulsive signals, various methods are developed (Baker 2007;Chang et al. 2016;Ertuncay and Costa 2019;Kardoutsou et al. 2017;Mena and Mai 2011;Shahi and Baker 2014;Zhai et al. 2018). These methods are focused to analyze seismic signals by using the indicators mentioned above. ...
... Shahi and Baker (2014) uses the horizontal components to detect the impulsive behavior of a given station. We use the method of Ertuncay and Costa (2019) to detect the impulsive signals since the method can analyze the components individually. ...
... Shahi and Baker (2014). Green, yellow and purple triangles indicate the impulsive signals detected by Shahi and Baker (2014), Ertuncay and Costa (2019) and both of the studies, respectively. Black triangles indicates nonimpulsive signals. ...
Near-fault ground motions may contain impulse behavior on velocity records. To calculate the probability of occurrence of the impulsive signals, a large dataset is collected from various national data providers and strong motion databases. The dataset has a large number of parameters which carry information on the earthquake physics, ruptured faults, ground motion parameters, distance between the station and several parts of the ruptured fault. Relation between the parameters and impulsive signals is calculated. It is found that fault type, moment magnitude, distance and azimuth between a site of interest and the surface projection of the ruptured fault are correlated with the impulsiveness of the signals. Separate models are created for strike-slip faults and non-strike-slip faults by using multivariate naïve Bayes classifier method. Naïve Bayes classifier allows us to have the probability of observing impulsive signals. The models have comparable accuracy rates, and they are more consistent on different fault types with respect to previous studies.
... Chang et al. [52] used the energy ratio in the area where PGV is measured in the waveform. Ertuncay and Costa [53] used a mixed decision-making algorithm by using the energy ration in both time and frequency domain. Baker [50] and Shahi and Baker [51] analyzed only the horizontal motions, whereas Chang et al. [52] and Ertuncay and Costa [53] also analyzed the vertical motion. ...
... Ertuncay and Costa [53] used a mixed decision-making algorithm by using the energy ration in both time and frequency domain. Baker [50] and Shahi and Baker [51] analyzed only the horizontal motions, whereas Chang et al. [52] and Ertuncay and Costa [53] also analyzed the vertical motion. The latter is important since it could create large demand on some typologies of buildings [11], and it was identified in many stations in the Amatrice earthquakes (see Results section). ...
... To determine the impulsive signals, we used the algorithms of Shahi and Baker [51], Chang et al. [52], Ertuncay and Costa [53]. These methods are selected due to their unique capabilities to capture impulsive signals by using different approaches and their effectiveness in capturing the impulsive signals on different parts of the waveforms. ...
Near fault seismic records may contain impulsive motions in velocity-time history. The seismic records can be identified as impulsive and non-impulsive depending on the features that their waveforms have. These motions can be an indicator of directivity or fling step effect, and they may cause dangerous effects on structures; for this reason, there is increasing attention on this subject in the last years. In this study, we collect the major earthquakes in Italy, with a magnitude large or equal to Mw 5.0, and identify the impulsive motions recorded by seismic stations. We correlate impulsive motions with directivity and fling step effects. We find that most earthquakes produced impulsive signals due to the directivity effect, though those at close stations to the 30 October 2016 Amatrice earthquake might be generated by the fling step effect. Starting from the analyzed impulses, we discuss on the potential influence of site effects on impulsive signals and suggest a characterization based on the main displacement directions of the impulsive horizontal displacements. Finally, we discuss on the damage of three churches in Emilia, which were subject to impulsive ground motion, underlying in a qualitative way, how the characteristics of the pulses may have had influences the structural response of the façades.
