Earthquake Database Research Articles

Assessment of Seismic Hazard in Yemen

The seismic hazard analyses for Yemen has attracted the attention of researchers during the last two decades. However, the studies are limited and mainly employ deterministic or approximate conventional probabilistic approaches. Furthermore, the conclusions drawn from these researches do not fit with current seismic design codes. This article describes the method and findings of a probabilistic seismic hazard assessment for Yemen, in accordance with current seismic design building regulations. The study utilized all the data sources, available nationally and internationally, in compiling earthquake database by covering the recent records and the seismic activity maps of the study region and regionalized the study area to eleven seismotectonic area sources based on the seismicity maps and available tectonic maps. On the analytical side, the study evaluated the earthquake recurrence analysis for each source, employed the logic tree concept to model the seismic sources, that may have significant effect on seismic hazard evaluation of Yemen, as a combination of area and line sources. Moreover, the study formulated a probabilistic forecasting model, used appropriate ground motion attenuation relationships, and developed seismic hazard contour maps for the entire Yemen area. The maps present dense contours of both peak ground accelerations, and short as well as long period spectral accelerations for different return periods. The highest predicted seismic hazard is found in Dhamar city. This study furnishes basic and essential requirements that will be of great help in developing advanced seismic design criteria for Yemen.

The missing craton edge: Crustal structure of the East European Craton beneath the Carpathian Orogen revealed by double-difference tomography

The Trans-European Suture Zone (TESZ) is the most important and extensive continental suture in Europe, marking the edge of the East European Craton (EEC), from the North Sea to the Black Sea. It corresponds to significant changes in surface geology and deep crustal structure, evident in seismic, gravitational and magnetic studies. However, the TESZ disappears beneath the Eastern Carpathians accretionary nappes and Neotethys ophiolites, thrusted over the subducted EEC passive margin in Romania that may have experienced progressive southward break-off generating post-collisional volcanism and anomalous seismicity in the Vrancea region of the South-East Carpathians. To illuminate the missing TESZ section and investigate the change in crustal properties from the Precambrian EEC across the collisional orogen and the impact of related volcanism, we determined a 3D seismic model of P and S wave velocities of the Eastern Carpathians in Romania. With the advent of new permanent broadband and short-period seismic stations of the Romanian Seismic Network, we were able to lower the earthquake magnitude detection threshold to 0.5 ML, largely expanding the earthquake database and create seismic images of the crust across the Carpathians. Using double-difference tomography and waveform cross-correlation differential times, we relocated local crustal earthquakes between 2010 and 2017 and jointly inverted for the 3D P and S -wave velocity structure down to the Moho discontinuity. Our study provides the highest resolution 3D crustal seismic model of this area to date and emphasizes the manifestation of surface tectonic boundaries at lower crustal depths. The TESZ is highlighted at mid-to lower-crustal depths beneath the Carpathian nappes, east of the post-collisional volcanic belt, as a gently-dipping transition from high Vp in the eastern footwall, the dipping Precambrian basement, to low Vp beneath the Carpathian Orogen to the west and a downward decrease in Vp/Vs. In parts of the volcanic region, Vp/Vs ratios increase suggesting the presence of fluids, mafic material and partial melts that may have altered the seismic structure of the TESZ. Relocated hypocenters are vertically distributed in the velocity transition zone from low to high especially beneath the youngest volcanoes in the south, consistent with a hypothetically active magmatic plumbing system. In the northern segment, where volcanoes are older, relatively high velocities are estimated and seismicity tends to cluster in the top 10 km, likely connected to a recently reactivated fault system. Seismicity in the foreland of the East Carpathians aligns along the NW-SE trending major crustal faults and continues beneath the thrust belt, suggesting the East European Craton basement extends beneath the orogen as far as the volcanic belt.

A Computational Tool for Ground-Motion Simulations Incorporating Regional Crustal Conditions

AbstractThis article introduces a computational tool, namely ground-motion simulation system (GMSS), for generating synthetic accelerograms based on stochastic simulations. The distinctive feature of GMSS is that it has two independently developed upper-crustal models (expressed in the form of shear-wave velocity profiles), which have been built into the program for deriving the frequency-dependent crustal factors, and one of these models was originally developed by the authors. GMSS also has provisions to allow the user to specify their own preferred crustal profile. Sufficient details of both crustal models (forming part of the seismological model) and the accelerogram simulation methodology are presented herein in one article, to allow any person who has programming skills (on a user-friendly platform such as MATLAB, see Data and Resources), to develop their computational tools to implement any further innovations in crustal modeling for direct engineering applications. Crustal properties deep into bedrock can only be accounted for implicitly by conventional ground-motion prediction equation (GMPE) as much depends on the region where the ground motion was recorded. This limitation of existing GMPEs poses a challenge to engineering in regions that are not well represented by any strong-motion database. Toward the end of this article, readers are enlightened with the potential transdisciplinary utility of using GMSS, to facilitate the retrieval and scaling of accelerograms sourced from a database of real earthquake records through the construction of a conditional mean spectrum.

Hybrid Low-Viscous-Damping Inelastic Displacement Ratios for Seismic Design and Assessment of Structures

ABSTRACT Until recently, the viscous damping ratio pervasively assumed has been . However, the dynamic identification of the experimental response recorded in recent large-scale shake-table tests shows that lower levels of viscous damping are more appropriate when material nonlinearities are explicitly accounted for. This paper investigates a hybrid constant-ductility inelastic displacement ratio, , which identifies the inelastic demand for a viscous damping ratio when the elastic demand for is known. A comprehensive statistical analysis is developed, and analytical estimates are provided, for inelastic single-degree-of-freedom systems with a wide variation of the constitutive parameters and a large earthquake database.

Identification of Electromagnetic Pre-Earthquake Perturbations from the DEMETER Data by Machine Learning

The low-altitude satellite DEMETER recorded many cases of ionospheric perturbations observed on occasion of large seismic events. In this paper, we explore 16 spot-checking classification algorithms, among which, the top classifier with low-frequency power spectra of electric and magnetic fields was used for ionospheric perturbation analysis. This study included the analysis of satellite data spanning over six years, during which about 8760 earthquakes with magnitude greater than or equal to 5.0 occurred in the world. We discover that among these methods, a gradient boosting-based method called LightGBM outperforms others and achieves superior performance in a five-fold cross-validation test on the benchmarking datasets, which shows a strong capability in discriminating electromagnetic pre-earthquake perturbations. The results show that the electromagnetic pre-earthquake data within a circular region with its center at the epicenter and its radius given by the Dobrovolsky’s formula and the time window of about a few hours before shocks are much better at discriminating electromagnetic pre-earthquake perturbations. Moreover, by investigating different earthquake databases, we confirm that some low-frequency electric and magnetic fields’ frequency bands are the dominant features for electromagnetic pre-earthquake perturbations identification. We have also found that the choice of the geographical region used to simulate the training set of non-seismic data influences, to a certain extent, the performance of the LightGBM model, by reducing its capability in discriminating electromagnetic pre-earthquake perturbations.

Probabilistic effectiveness of visco-elastic dampers considering earthquake excitation uncertainty and ambient temperature fluctuation

The mechanical property of the visco-elastic damper (VED) is dependent on the loading frequency and the ambient temperature. Due to the frequency-dependency, the effectiveness of VED varies as the earthquake excitation alters. However, such an excitation-sensitivity is usually ignored in the design and evaluation of VED. To rationally quantify the VED effectiveness considering the excitation uncertainty, a probabilistic analysis scheme is proposed. In the scheme, multiple response history analyses (RHA) are conducted to reveal the probabilistic properties of the VED effectiveness. The required number of RHA is studied based on three earthquake databases and three VED-structure models with different natural periods. It is shown that at least 30 RHA analyses are needed to get a good result. Due to the temperature-dependency of VED, the probabilistic results are notably influenced by the ambient temperature. Unfortunately, by far, there is no approach to account for the effect of temperature fluctuation on the probabilistic effectiveness of VED. To address this problem, the Fourier series are adopted to model the building thermal environment. By doing so, the seasonal fluctuation of temperature is well incorporated, and the probabilistic property of the working temperature of VED is obtained. Based on the law of total probability, a probabilistic framework for evaluating the effectiveness of VED is established by integrating the effects of excitation uncertainty and temperature fluctuation. The framework consists of two major steps. In step one, the probabilistic effectiveness of VED is computed under a series of pre-defined temperatures. In step two, the probabilistic results under the specific temperatures are integrated over the fluctuating range of the temperature. An example is provided to illustrate the application of the proposed framework. From the example, the temperature fluctuation has little influence on the expectation of the VED effectiveness, but it notably affects the dispersion of the probabilistic effectiveness of VED. In this study, two types of VED (Acrylic and Diene) are considered. The conclusions drawn from the two types of VED are alike.

Probability of magnitude detection for the seismological network of Oman

The current work aims to evaluate the probability of magnitude detection in the Sultanate of Oman, adopting the probability-based magnitude of completeness method. The spatial configuration of the seismic network in addition to the recorded phases and the applied magnitude equation are required for implementation of the method. Therefore, the history of the Oman Seismological Network and the up-to-date configuration of seismic stations were documented. The analysis is done based on a unified magnitude earthquake database for all recorded events since November 2013 until April 2018. The probability of detection is computed for each seismic station, and then synthesized maps of the detection probability of an event with specific magnitude were developed. These maps show that Oman Seismological Network is able to record small earthquakes occurring close to inhabitant areas. The probability of completeness magnitude map is presented and indicates that most earthquakes generated within the territory of the Sultanate with magnitude equal to or greater than ML 2.5 are completely recorded since 2013. Moreover, all earthquakes with magnitude ML 3.5 initiated from the potential seismic sources in and around Oman are efficiently recorded by the Oman Seismological Network. Nevertheless, the results showed that additional seismic stations should be installed to enhance the detectability of man-made and small natural earthquakes at specific areas in the Sultanate of Oman.

Attention-Based Convolutional Neural Network for Earthquake Event Classification

This letter presents a deep convolutional neural network (CNN) with attention module that improves the performance of the classification of various earthquake events. Addressing all possible earthquake events, including not only microearthquakes and artificial-earthquakes but also large-earthquakes, requires both suitable feature expression and a classifier that can effectively discriminate seismic waveforms under adverse conditions. To robustly classify earthquake events, a deep CNN with an attention module was proposed in raw seismic waveforms. Representative experimental results show that the proposed method provides an effective structure for earthquake events classification and, with the Korean peninsula earthquake database from 2016 to 2018, outperforms previous state-of-the-art methods.

Applications and Evaluation of the IRIS Earthquake Browser: A Web-Based Tool That Enables Multidimensional Earthquake Visualization

Abstract The Incorporated Research Institutions for Seismology Earthquake Browser (IEB; see Data and Resources) is a web-based tool that enables anyone to query an earthquake database composed of over five million events recorded over the past 50 yr. The IEB visitor can query on earthquake magnitude, depth, timing, and location and can visually display the results in 2D map view or as an interactive pseudo 3D view. The user can toggle features such as plate tectonic boundaries, terrain or satellite mapping, and zoom to place the results in a geologic or geopolitical context and add visual appeal. To better understand who visits the IEB and why, to include information on demographics and how users perceive the IEB functionality and ability to meet their needs, we conducted a pop-up user survey on the IEB from 25 January to 21 February 2018. We received 495 useable responses from 58 countries, with 40% of the total respondents from the United States. The largest demographic consists of interested citizens who are 55 yr of age or older and have a high school education. We also find that visitors come to the IEB to learn about earthquakes for two main reasons: for their own personal knowledge or because expanding their knowledge is important to their research or work in a professional context. We also find a dramatic increase in survey respondent activity following the 16 February 2018 M 7.2 Oaxaca, Mexico, earthquake, with many respondents interested in finding more information about recent earthquake events that affect them or their family. Our observations indicate that users are successful and satisfied with the ease of use and amount of time spent on the IEB to find answers to their questions about earthquakes. The most beneficial feature of the IEB as identified by survey respondents is the spatiotemporal visual display of real earthquake data.

Toward Homogeneous Estimation of Long-Term Seismicity from Historical Materials: Number of Felt Earthquakes in Tokyo since 1668

Abstract Long-term seismicity in a certain region can be estimated from the number of felt earthquakes obtained from seismic intensity measurements or historical documents. To obtain a homogeneous estimate, continuous records with a uniform detection threshold are required. Seismic intensity data in Tokyo, which are measured by professional observers and archived by the Japan Meteorological Agency (JMA) since 1885, indicate that the mean annual number of felt earthquakes with intensities ≥1 and ≥2 on the JMA scale is 45 and 14, respectively. The database of felt earthquakes can be extended back to more than 200 yr using historical daily records. During the Edo period from 1603 to 1867, numerous historical documents in Edo (former Tokyo) recorded information about damaging earthquakes, such as the 1703 Kanto earthquake and the 1855 Edo earthquake. Several feudal clans documented daily weather and felt earthquakes in Edo. Among them, daily records documented by the Tsugaru and Sakakibara clans cover a period of ∼200 yr. Although these records do not cover the entire period and weather reports for several years are missing, the Tsugaru and Sakakibara clans documented ∼1600 and ∼700 earthquakes, respectively. Using these data, we determined that a total of ∼2000 felt earthquakes occurred between 1668 and 1867, thereby yielding a mean rate of ∼10 earthquakes per year. When we exclude the missing period of daily records, the average rate increased to 11 per year, which is similar to the annual number of earthquakes with seismic intensities ≥2 in the modern JMA record. These data of felt earthquakes in Tokyo are considered to be more homogeneous than the previous estimates, although the total number is smaller, and can be used to study the long-term seismicity rate change.

Earthquake Event Classification Using Multitasking Deep Learning

This letter proposes an attention-based convolutional neural network architecture for multitasking learning to accurately classify not only the presence of an earthquake but also the event type of the earthquake. In particular, to improve the performance in earthquake-type classification, we develop an attention-based feature aggregation framework embedded in multitask learning architecture. Representative experimental results show that the proposed method provides an effective structure for an earthquake detection and event classification with an earthquake database of the Korean peninsula and the Circum-Pacific belt.

A coupled stick-slip-rotation model for earthquake-induced sliding displacement of slopes in Iran

This paper presents a semi-empirical model to predict earthquake-induced displacement of slopes, based on a coupled stick-slip-rotation (CSSR) analysis recently developed by the authors in a preceding study. The earthquake database selected for the analyses was constituted of 3319 records (including two horizontal components) of ground accelerations associated with 461 earthquakes that occurred in Iran. The presented predictive model was developed by the multiple regression analyses and represents sliding displacement as a function of the maximum velocity coefficient-time history (kv-max), yield acceleration coefficient (ky), slip length (L), and period ratio (Ts/Tm). The conditions of efficiency, sufficiency, and predictability of the seismic parameters have indicated that kv-max is the best choice satisfying these conditions to be used in the predictive model compared with the other intensity measures. A semi-empirical equation has been proposed to assure the predictability condition of kv-max. Performance of the presented semi-empirical model is evaluated by comparing its predictions with a few selected regional and global models recently proposed for sliding displacement of slopes.

Multifeature Fusion-Based Earthquake Event Classification Using Transfer Learning

This letter proposes a multifeature fusion model using deep convolution neural networks and transfer learning approach for earthquake event classification. There are several feature representations for seismic analysis, such as the time domain, the frequency domain, and the time–frequency domain. To successfully classify various earthquake events, we propose a novel model that combines these features hierarchically. In addition, we apply a transfer learning to mitigate overfitting problem of deep learning model while achieving high classification performance. To evaluate our approach, we conduct experiments with the Korean peninsula earthquake database from 2016 to 2018 and a large earthquake database on the Circum-Pacific belt in 2019. The experimental results show that the proposed method outperforms over the compared state-of-the-art methods.

Global adjoint tomography—model GLAD-M25

SUMMARYBuilding on global adjoint tomography model GLAD-M15, we present transversely isotropic global model GLAD-M25, which is the result of 10 quasi-Newton tomographic iterations with an earthquake database consisting of 1480 events in the magnitude range 5.5 ≤ Mw ≤ 7.2, an almost sixfold increase over the first-generation model. We calculated fully 3-D synthetic seismograms with a shortest period of 17 s based on a GPU-accelerated spectral-element wave propagation solver which accommodates effects due to 3-D anelastic crust and mantle structure, topography and bathymetry, the ocean load, ellipticity, rotation and self-gravitation. We used an adjoint-state method to calculate Fréchet derivatives in 3-D anelastic Earth models facilitated by a parsimonious storage algorithm. The simulations were performed on the Cray XK7 ‘Titan’ and the IBM Power 9 ‘Summit’ at the Oak Ridge Leadership Computing Facility. We quantitatively evaluated GLAD-M25 by assessing misfit reductions and traveltime anomaly histograms in 12 measurement categories. We performed similar assessments for a held-out data set consisting of 360 earthquakes, with results comparable to the actual inversion. We highlight the new model for a variety of plumes and subduction zones.

An empirical model for fatality estimation of earthquakes in Iran

In order to estimate the human loss after an earthquake to address risk mitigation and response measures, appropriate models should be developed based on local conditions. In this paper, an empirical model for estimating the mortality rate based on shaking related parameter (PGA) is presented for Iran. For this purpose, a reliable fatality database of past earthquakes occurred in the country (between 1962 and 2017) along with corresponding ground motion shaking maps were compiled. It includes information of 88 fatal earthquakes in different cities and villages, compiled from reliable resources. Three distinct functional forms including log-linear, exponential and lognormal cumulative distribution were applied to be fitted to data. To evaluate the appropriateness of different functional forms a residual analysis was performed. The results indicate that the log-linear model shows the best performance. Additionally, a sensitivity analysis was performed to evaluate the impact of events with highest contributions in database on fatality function. The results depicted that excluding data of Bam (2003), Iran Earthquake may reduce fatality ratio to about 5%. This can be related to the paucity of data in high acceleration ranges (near 800 cm/s2) in the database. Finally, two separate curves have been developed for day and night. As expected, the result depicted that fatality ratio in day time is much lower than the night hours. The proposed model can be used for rapid loss assessment in Iran and other countries with similar construction types to provide an initial estimation of deaths after earthquakes or determining the priorities for risk reduction.

Prediction the Locations of Future Earthquakes in Eastern Part of Iraq Using GIS Techniques

An earthquake is a natural phenomenon and natural catastrophe. It can cause massive damage and losses of life. Future prediction of natural catastrophes such as earthquakes can minimize their risks and save people lives and properties. GIS can be very essential for predicting and estimating a future earthquake. Thus, this study is dedicated to predict the locations of future earthquakes in the eastern part of Iraq using GIS techniques. The study involves establishing a GIS database (DB) of the previous earthquake in the period (1985 - 2015), identifying the parameters that a map layers of all parameters (factors) which are responsible on occurring earthquakes, and using GIS techniques to identify and predict the locations of the future earthquakes.This research is accomplished through a number of essential stages. These are data gathering, data processing and data analyzing. The gathered data among the period of the study (from 1985 to 2015) was classified into three stages ((1985-1995), (1996-2005), and (2006 to 2015) to distinguish a fine details of past earthquakes easily and precisely. The results show that GIS is a powerful tool that can provide an accurate prediction of upcoming natural phenomena such as earthquakes. This can evaluate earthquake hazards and raise the awareness of the government to save people’s lives, properties and infrastructures against the possible future earthquakes and the accuracy assessment processes reveals that about 80% of coming quakes happened where they expected (very close to the prediction seismic line).

Correlations of spectral accelerations in the Chilean subduction zone

The correlation between spectral accelerations is key in the construction of conditional mean spectra, the computation of vector-valued seismic hazard, and the assessment of seismic risk of spatially distributed systems, among other applications. Spectral correlations are highly dependent on the earthquake database used, and thus, region-specific correlation models have been developed mainly for earthquakes in western United States, Europe, Middle East, and Japan. Correlation models based on global data sets for crustal and subduction zones have also become available, but there is no consensus about their applicability on a specific region. This study proposes a new correlation model for 5% damped spectral accelerations and peak ground velocity in the Chilean subduction zone. The correlations obtained were generally higher than those observed from shallow crustal earthquakes and subduction zones such as Japan and Taiwan. The study provides two illustrative applications of the correlation model: (1) computation of conditional spectra for a firm soil site located in Santiago, Chile and (2) computation of bivariate hazard for spectral accelerations at two structural periods.

Spectral Shape in Accordance with the Magnitude and Distance of Earthquakes and Its Effect on Multi-DOF Structures

In this study earthquake records were collected for rock conditions that do not reflect seismic amplification by soil from global earthquake databases such as PEER, USGS, and ESMD. The collected earthquake records were classified and analyzed based on the magnitude and distance of earthquakes. Based on the analyzed earthquakes, the design response spectrum shape, effective ground acceleration, and amplification ratios for each period band are presented. In addition, based on the analyzed data, the story shear force for 5F, 10F, 15F, and 20F were derived through an analysis of the elastic time history for multi-DOF structures. The results from analyzing the rock earthquake record show that the seismic load tends to be amplified greatly in the short period region, which is similar to results observed from the Gyeongju and Pohang earthquakes. In addition, the results of the multi-DOF structure analysis show that existing seismic design criteria can be underestimated and designed in the high-order mode of short- and medium-long cycle structures.

A database of damaging small-to-medium magnitude earthquakes

Interest in small-to-medium magnitude earthquakes and their potential consequences has increased significantly in recent years, mostly due to the occurrence of some unusually damaging small events, the development of seismic risk assessment methodologies for existing building stock, and the recognition of the potential risk of induced seismicity. As part of a clear ongoing effort of the earthquake engineering community to develop knowledge on the risk posed by smaller events, a global database of earthquakes with moment magnitudes in the range from 4.0 to 5.5 for which damage and/or casualties have been reported has been compiled and is made publicly available. The two main purposes were to facilitate studies on the potential for earthquakes in this magnitude range to cause material damage and to carry out a statistical study to characterise the frequency with which earthquakes of this size cause damage and/or casualties (published separately). The present paper describes the data sources and process followed for the compilation of the database, while providing critical discussions on the challenges encountered and decisions made, which are of relevance for its interpretation and use. The geographic, temporal, and magnitude distributions of the 1958 earthquakes that make up the database are presented alongside the general statistics on damage and casualties, noting that these stem from a variety of sources of differing reliability. Despite its inherent limitations, we believe it is an important contribution to the understanding of the extent of the consequences that may arise from earthquakes in the magnitude range of study.

New Model for Simulating Random Synthetic Stochastic Earthquake Scenarios

ABSTRACT This research attempts to propose a new simulating model capable of generating synthetic stochastic earthquake scenario including both seismological, and acceleration signal properties. The Iranian plateau is selected as a mainland for developing the model due to the existence of high-quality earthquake databases. This model is composed of two sub-models. The first one generates random seismic event scenarios including mainshocks and aftershocks. The second one procreates 3-dimensional synthetic stochastic accelerograms for each of the previously simulated events. This scenario-based predictive model offers a powerful tool to the risk modelers interested in lifetime seismic risk assessment through more realistic earthquake scenarios.