http://www.sciencedirect.com/science/article/pii/S1674987114001376

A new model is derived to predict the peak ground acceleration(PGA) utilizing a hybrid method coupling artificial neural network(ANN) and simulated annealing(SA), called SA-ANN. The proposed model relates PGA to earthquake source to site distance, earthquake magnitude, average shear-wave velocity,faulting mechanisms, and focal depth. A database of strong ground-motion recordings of 36 earthquakes,which happened in Iran's tectonic regions, is used to establish the model. For more validity verification,the SA-ANN model is employed to predict the PGA of a part of the database beyond the training data domain. The proposed SA-ANN model is compared with the simple ANN in addition to 10 well-known models proposed in the literature. The proposed model performance is superior to the single ANN and other existing attenuation models. The SA-ANN model is highly correlated to the actual records(R=0.835 and r =0.0908) and it is subsequently converted into a tractable design equation.     

Probability-based damage assessment for reinforced concrete bridge columns considering the corrosive and seismic hazards in Taiwan

In this paper, we consider the time at which earthquake events occur when analyzing seismic structural damage to a deteriorating RC bridge within a specified period. Because the uncertainty exists in the occurrence time of earthquake events, Monte Carlo simulation is applied. The proposed procedure for evaluating the exceedance probability, which corresponds to a specified limit state, is then applied to a case study of RC bridges in Taiwan to demonstrate its applicability. This study selects three typical RC bridges located in the Taipei Basin, Taiwan, to analyze exceedance probabilities of specified damage states during various specified periods and then discusses the cumulative damage effect on the exceedance probabilities of specified damage states. Additionally, for the chloride-induced deteriorating bridges at various distances to the sea in Suao, Taiwan, the effects of the deterioration and seismic structural damage on the exceedance probabilities of specified damage states are demonstrated and discussed. The proposed assessment procedure can help engineers understand whether the deterioration would accelerate the declining seismic performance of bridges and shorten their serviceability-related and safety-related service lives, as well as provide reference for repairing RC bridges and retrofitting their seismic performance.     

Peculiarities of the within-year distribution of earthquakes in the Kuril region

Recent investigations have shown that the probability of the occurrence of earthquakes in a specified region depends on several factors, such as the latitude of the study region, as well as the lunar and solar tidal forces, which are governed by the mutual arrangement of bodies in the Sun-Earth-Moon system. The objective of our work is to prove that the irregularity of the within-year distribution of seismic events in the Pacific regions (the Kuril Islands and Hokkaido Island) is a statistically significant process, which is manifested in different ways for earthquakes with various source depths and energy levels. The hypothesis about the uniform within-year distribution of earthquakes is refuted for shallow events. However, it is shown that deep earthquakes are distributed uniformly. This work attempts for the first time to determine the stability degree of the within-year irregularity of seismic events with respect to the observation time interval (from 28 to 5 years). Two peaks are noted in the annual distribution of the earthquakes. The relation of the peaks of the within-year seismic activity to the Earth’s position on the ecliptic plane, as well as the relationships between the peaks, the magnitude range of the events, and the position of the specified subregion, is considered. The principal maximum of the seismic activity falls in the November-March period, which matches the minimum Earth-Sun distance.     

Earthquake magnitude prediction by adaptive neuro-fuzzy inference system (ANFIS) based on fuzzy C-means algorithm

This paper investigates the prediction of future earthquakes that would occur with magnitude 5.5 or greater using adaptive neuro-fuzzy inference system (ANFIS). For this purpose, the earthquake data between 1950 and 2013 that had been recorded in the region with 2°E longitude and 4°N latitude in Iran has been used. Thereupon, three algorithms including grid partition (GP), subtractive clustering (SC) and fuzzy C-means (FCM) were used to develop models with the structure of ANFIS. Since the earthquake data for the specified region had been reported on different magnitude scales, suitable relationships were determined to convert the magnitude scales into moment magnitude and all records uniformed based on the relationships. The uniform data were used to calculate seismicity indicators, and ANFIS was developed based on considered algorithms. The results showed that ANFIS-FCM with a high accuracy was able to predict earthquake magnitude.     

Three-parameter generalized exponential distribution in earthquake recurrence interval estimation

The purpose of this article is to study the three-parameter (scale, shape, and location) generalized exponential (GE) distribution and examine its suitability in probabilistic earthquake recurrence modeling. The GE distribution shares many physical properties of the gamma and Weibull distributions. This distribution, unlike the exponential distribution, overcomes the burden of memoryless property. For shape parameter  β> 1, the GE distribution offers increasing hazard function, which is in accordance with the elastic rebound theory of earthquake generation. In the present study, we consider a real, complete, and homogeneous earthquake catalog of 20 events with magnitude above 7.0 (Yadav et al. in Pure Appl Geophys 167:1331–1342, 2010) from northeast India and its adjacent regions (20°–32°N and 87°–100°E) to analyze earthquake inter-occurrence time from the GE distribution. We apply the modified maximum likelihood estimation method to estimate model parameters. We then perform a number of goodness-of-fit tests to evaluate the suitability of the GE model to other competitive models, such as the gamma and Weibull models. It is observed that for the present data set, the GE distribution has a better and more economical representation than the gamma and Weibull distributions. Finally, a few conditional probability curves (hazard curves) are presented to demonstrate the significance of the GE distribution in probabilistic assessment of earthquake hazards.     

Global strain rates in western to central Himalayas and their implications in seismic hazard assessment

The Himalayas has experienced varying rates of earthquake occurrence in the past in its seismo-tectonically distinguished segments which may be attributed to different physical processes of accumulation of stress and its release, and due diligence is required for its inclusion for working out the seismic hazard. The present paper intends to revisit the various earthquake occurrence models applied to Himalayas and examines it in the light of recent damaging earthquakes in Himalayan belt. Due to discordant seismicity of Himalayas, three types of regions have been considered to estimate larger return period events. The regions selected are (1) the North-West Himalayan Fold and Thrust Belt which is seismically very active, (2) the Garhwal Himalaya which has never experienced large earthquake although sufficient stress exists and (3) the Nepal region which is very seismically active region due to unlocked rupture and frequently experienced large earthquake events. The seismicity parameters have been revisited using two earthquake recurrence models namely constant seismicity and constant moment release. For constant moment release model, the strain rates have been derived from global strain rate model and are converted into seismic moment of earthquake events considering the geometry of the finite source and the rates being consumed fully by the contemporary seismicity. Probability of earthquake occurrence with time has been estimated for each region using both models and compared assuming Poissonian distribution. The results show that seismicity for North-West region is observed to be relatively less when estimated using constant seismicity model which implies that either the occupied accumulated stress is not being unconfined in the form of earthquakes or the compiled earthquake catalogue is insufficient. Similar trend has been observed for seismic gap area but with lesser difference reported from both methods. However, for the Nepal region, the estimated seismicity by the two methods has been found to be relatively less when estimated using constant moment release model which implies that in the Nepal region, accumulated strain is releasing in the form of large earthquake occurrence event. The partial release in second event of May 2015 of similar size shows that the physical process is trying to release the energy with large earthquake event. If it would have been in other regions like that of seismic gap region, the fault may not have released the energy and may be inviting even bigger event in future. It is, therefore, necessary to look into the seismicity from strain rates also for its due interpretation in terms of predicting the seismic hazard in various segments of Himalayas.     

Integrated model for earthquake risk assessment using neural network and analytic hierarchy process:Aceh province,Indonesia

Catastrophic natural hazards,such as earthquake,pose serious threats to properties and human lives in urban areas.Therefore,earthquake risk assessment(ERA)is indispensable in disaster management.ERA is an integration of the extent of probability and vulnerability of assets.This study develops an integrated model by using the artificial neural network–analytic hierarchy process(ANN–AHP)model for constructing the ERA map.The aim of the study is to quantify urban population risk that may be caused by impending earthquakes.The model is applied to the city of Banda Aceh in Indonesia,a seismically active zone of Aceh province frequently affected by devastating earthquakes.ANN is used for probability mapping,whereas AHP is used to assess urban vulnerability after the hazard map is created with the aid of earthquake intensity variation thematic layering.The risk map is subsequently created by combining the probability,hazard,and vulnerability maps.Then,the risk levels of various zones are obtained.The validation process reveals that the proposed model can map the earthquake probability based on historical events with an accuracy of 84%.Furthermore,results show that the central and southeastern regions of the city have moderate to very high risk classifications,whereas the other parts of the city fall under low to very low earthquake risk classifications.The findings of this research are useful for government agencies and decision makers,particularly in estimating risk dimensions in urban areas and for the future studies to project the preparedness strategies for Banda Aceh.     

A new model for the prediction of earthquake ground-motion duration in Iran

The paper proposes a new empirical model to estimate earthquake ground-motion duration, which significantly influences the damage potential of an earthquake. The paper is concerned with significant duration parameters that are defined as the time intervals between which specified values of Arias intensity are reached. In the proposed model, significant duration parameters have been expressed as a function of moment magnitude, closest site-source distance, and site condition. The predictive model has been developed based on a database of earthquake ground-motion records in Iran, containing 286 records up to the year 2007, and a random-effect regression procedure. The result of the proposed model has been compared with that of other published models. It has been found that the proposed model can predict earthquake ground-motion duration in Iran with adequate accuracy.     

The distribution of annual maximum earthquake magnitude around Taiwan and its application in the estimation of catastrophic earthquake recurrence probability

The annual maximum earthquake magnitudes around Taiwan from 1900 to 2009 are presented in this paper. Using the distribution of the AMEM, a probabilistic framework to estimate the recurrence probability of a large-size earthquake is also proposed and an illustration was made in this paper. The mean value of the 110-AMEM is 6.433, and the coefficient of variation is around 10%. The results of two goodness-of-fit tests show that the Gamma and lognormal distributions are relatively suitable to represent the AMEM around Taiwan among five common probability distributions. Using the proposed approach, the recurrence probability is 4% for an earthquake with magnitude greater than 7.5 in a 1-year period around Taiwan. More site-specifically, the probability is around 5% in Central Taiwan for such an earthquake to occur in a 50-year period.     

Neural network-based model for assessing failure potential of highway slopes in the Alishan, Taiwan Area: Pre- and post-earthquake investigation

This paper is aimed at creating an empirical model for assessing failure potential of highway slopes, with a special attention to the failure characteristics of the highway slopes in the Alishan, Taiwan area prior to, and post, the 1999 Chi-Chi, Taiwan earthquake. The basis of the study is a large database of 955 slope records from four highways in the Alishan area. Artificial neural network (ANN) is utilized to “learn” from this database. The developed ANN model is then used to study the effect of the Chi-Chi earthquake on the slope failure characteristics in the Alishan area. Significant changes in the degrees of influence of several factors (variables) are found and possible reasons for such changes are discussed. The novelty of this paper lies in the fact that the developed ANN models are used as a tool to investigate the slope failure characteristics before and after the Chi-Chi earthquake.     

基于ANN与GIS技术的区域岩溶塌陷稳定性预测———以桂林西城区为例

岩溶地面塌陷是岩溶区常见的一种地质灾害, 塌陷区域预测是进行国土规划、资源开发与灾害防治的必要工作.由于岩溶塌陷的影响因素众多且相互作用, 发展过程复杂, 加之各评价因子的数值获取困难, 致使长期以来塌陷区域定量预测成为一个难以解决的课题.现行的区域预测模型不能描述塌陷形成模式的非线性特征, 也难以克服评价因子权重确定过程中人为经验因素的影响.神经网络技术的自学习、自适应与高度非线性映射特点显示了其在塌陷区域预测领域中应用的前景.根据研究区内地面塌陷空间聚集分布的特征, 提出了不同因子组合条件下塌陷发生可能性的定量化方法, 结合选定的评价因子类别确定了神经网络预测模型的结构, 利用312个塌陷点样本中的292个进行网络训练, 余下的20个样本的校验结果表明该模型具有较高的可信度.运用GIS技术将研究区进行评价单元划分, 并获取各评价因子的取值, 输入到训练好的网络中进行预测.将各单元的输出值进行归并处理后得到研究区岩溶塌陷的稳定级分区图.      

Bayesian estimation of seismic hazard for two sites in Switzerland

Seismic hazard analysis is based on data and models, which both are imprecise and uncertain. Especially the interpretation of historical information into earthquake parameters, e.g. earthquake size and location, yields ambiguous and imprecise data. Models based on probability distributions have been developed in order to quantify and represent these uncertainties. Nevertheless, the majority of the procedures applied in seismic hazard assessment do not take into account these uncertainties, nor do they show the variance of the results. Therefore, a procedure based on Bayesian statistics was developed to estimate return periods for different ground motion intensities (MSK scale).Bayesian techniques provide a mathematical model to estimate the distribution of random variables in presence of uncertainties. The developed method estimates the probability distribution of the number of occurrences in a Poisson process described by the parameter . The input data are the historical occurrences of intensities for a particular site, represented by a discrete probability distribution for each earthquake. The calculation of these historical occurrences requires a careful preparation of all input parameters, i.e. a modelling of their uncertainties. The obtained results show that the variance of the recurrence rate is smaller in regions with higher seismic activity than in less active regions. It can also be demonstrated that long return periods cannot be estimated with confidence, because the time period of observation is too short. This indicates that the long return periods obtained by seismic source methods only reflects the delineated seismic sources and the chosen earthquake size distribution law.     

Statistical analysis of the largest possible earthquake magnitudes on the Ecuadorian coast for selected return periods

ABSTRACT

In this work, we have studied the largest earthquake magnitudes on the Ecuadorian coast by using the principles of Extreme Value Analysis based on its two approaches: Block Maxima and Peaks-over-Threshold. First, before modelling the recorded earthquakes, the K-means clustering technique was applied to determine a classification according to the level of magnitude of the earthquakes. Then, models based on the Extreme Value theory of earthquake magnitudes were developed for each of the four clusters that were found, and finally, the best-fitted models were those known as Fréchet and Gumbel ones. The zone with the greatest earthquake magnitudes on the Ecuadorian coast is located between the north of the province of Manabí and the south of the province of Esmeraldas, with a return period of 50 years for an earthquake with magnitude greater than 7.7 M_W.     

Seismic reliability analysis of earth slopes under short term stability conditions

Different models were developed for evaluating the probabilistic three-dimensional (3-D) stability analysis of earth slopes and embankments under earthquake loading. The 3-D slope stability model assumed is that of a simple cylindrical failure surface. The probabilistic models evaluate the probability of failure under seismic loading considering the randomness of earthquake occurrence, and earthquake induced acceleration and uncertainties stemming from the discrepancies between laboratory-measured and in-situ values of shear strength parameters. The models also takes into consideration the spatial variabilities and correlations of soil properties. The probabilistic analysis and design approach is capable of obtaining the 2-D and 3-D static and dynamic safety factors, the probability of slope failure, the earthquake induced acceleration coefficient, the yield acceleration coefficient, the earthquake induced displacement, and the probability of allowable displacement exceedance taking into account the local site effect. The approach is applied to a well known landslide case: Congress Street Landslide in Chicago. A sensitivity analysis was conducted on the different parameters involved in the models by applying those models to the Congress Street landslide considering different levels of seismic hazard. Also, a sensitivity analysis was carried out to study the sensitivity of computed results to input parameters of undrained shear strength, and corrective factors. A comparison was made between the different models of failure. The parametric study revealed that the hypocentral distance and earthquake magnitude have major influence on the earthquake induced displacement, probability of failure and dynamic 2-D and 3-D safety factors.     

Spatial mapping of earthquake hazard parameters in the Hindukush–Pamir Himalaya and adjacent regions: Implication for future seismic hazard

The study deals spatial mapping of earthquake hazard parameters like annual and 100-years mode along with their 90% probability of not being exceeded (NBE) in the Hindukush–Pamir Himalaya and adjoining regions. For this purpose, we applied a straightforward and most robust method known as Gumbel’s third asymptotic distribution of extreme values (GIII). A homogeneous and complete earthquake catalogue during the period 1900–2010 with magnitude M_W ⩾ 4.0 is utilized to estimate these earthquake hazard parameters. An equal grid point mesh, of 1° longitude X 1° latitude, is chosen to produce detailed earthquake hazard maps. This performance allows analysis of the localized seismicity parameters and representation of their regional variations as contour maps. The estimated result of annual mode with 90% probability of NBE is expected to exceed the values of M_W 6.0 in the Sulaiman–Kirthar ranges of Pakistan and northwestern part of the Nepal and surroundings in the examined region. The 100-years mode with 90% probability of NBE is expected to exceed the value of M_W 8.0 in the Hindukush–Pamir Himalaya with Caucasus mountain belt, the Sulaiman–Kirthar ranges of Pakistan, northwestern part of the Nepal and surroundings, the Kangra–Himanchal Pradesh and Kashmir of India. The estimated high values of earthquake hazard parameters are mostly correlated with the main tectonic regimes of the examined region. The spatial variations of earthquake hazard parameters reveal that the examined region exhibits more complexity and has high crustal heterogeneity. The spatial maps provide a brief atlas of the earthquake hazard in the region.     

Probabilities of earthquake occurrences in Mainland Southeast Asia

The frequency–magnitude distributions of earthquakes are used in this study to estimate the earthquake hazard parameters for individual earthquake source zones within the Mainland Southeast Asia. For this purpose, 13 earthquake source zones are newly defined based on the most recent geological, tectonic, and seismicity data. A homogeneous and complete seismicity database covering the period from 1964 to 2010 is prepared for this region and then used for the estimation of the constants, a and b, of the frequency–magnitude distributions. These constants are then applied to evaluate the most probable largest magnitude, the mean return period, and the probability of earthquake of different magnitudes in different time spans. The results clearly show that zones A, B, and E have the high probability for the earthquake occurrence comparing with the other seismic zones. All seismic source zones have 100 % probability that the earthquake with magnitude ≤6.0 generates in the next 25 years. For the Sagaing Fault Zone (zones C), the next Mw 7.2–7.5 earthquake may generate in this zone within the next two decades and should be aware of the prospective Mw 8.0 earthquake. Meanwhile, in Sumatra-Andaman Interplate (zone A), an earthquake with a magnitude of Mw 9.0 can possibly occur in every 50 years. Since an earthquake of magnitude Mw 9.0 was recorded in this region in 2004, there is a possibility of another Mw 9.0 earthquake within the next 50 years.     

Dealing with the complexity of earthquake using neurocomputing techniques and estimating its magnitudes with some low correlated predictors

Earthquake, being one of the most hazardous geophysical events, has fascinated the scientists all over the globe to investigate an efficient predictive methodology. In India, earthquake is not a rare event. Purpose of the present paper is to view the earthquake as a complex system and then to estimate the magnitude of earthquake over Indian subcontinent using artificial neural network with backpropagation learning. The day, month, and year of occurrence of the earthquake, latitude of the place, and the longitude of the place are considered as the predictor variables. After considering various sizes of the hidden layer, it has been found that artificial neural network in the form of multilayer perceptron with two hidden nodes and sigmoid nonlinearity can be used to anticipate the magnitude of earthquake over Indian subcontinent. The paper concludes with an outline for future work.     

Seismic hazard across Bulgaria and neighbouring areas: regional and site-specific maximum credible magnitudes and earthquake perceptibility

A probabilistic seismic hazard assessment is developed here using maximum credible earthquake magnitude statistics and earthquake perceptibility hazard. Earthquake perceptibility hazard is defined as the probability a site perceives ground shaking equal to or greater than a selected ground motion level X, resulting from an earthquake of magnitude M, and develops estimates for the most perceptible earthquake magnitude, M _P(max). Realistic and usable maximum magnitude statistics are obtained from both whole process and part process statistical recurrence models. These approaches are extended to develop relationships between perceptible earthquake magnitude hazard and maximum magnitude recurrence models that are governed by asymptotic and finite return period properties, respectively. Integrated perceptibility curves illustrating the probability of a specific level of perceptible ground motion due to all earthquakes over the magnitude range extending from ?∞ to a magnitude M _i are then developed from reviewing site-specific magnitude perceptibility. These lead on to achieving site-specific annual probability of exceedance hazard curves for the example cities of Sofia and Thessaloniki for both horizontal ground acceleration and ground velocity. Both the maximum credible earthquake magnitude M ₃ and the most perceptible earthquake magnitude M _P(max) are of importance to the earthquake engineer when approaching anti-seismic building design. Both forms of hazard are illustrated using contoured hazard maps for the region bounded by 39°–45°N, 19°–29°E. Patterns are observed for these magnitude hazard estimates—especially M _P(max) specific to horizontal ground acceleration and horizontal ground velocity—and compared to inferred patterns of crustal deformation across the region. The full geographic region considered is estimated to be subject to a maximum credible earthquake magnitude M ₃—estimated using cumulative seismic moment release statistics—of 7.53 M _w, calculated from the full content of the adopted earthquake catalogue, while Bulgaria’s capital, Sofia, is estimated a comparable value of 7.36 M _w. Sofia is also forecast most perceptible earthquake magnitudes for the lowest levels considered for horizontal ground acceleration of M _PA(50) = 7.20 M _w and horizontal ground velocity of M _PV(5) = 7.23 M _w for a specimen focal depth of 15 km.     

Estimation of earthquake hazard parameters for incomplete and uncertain data files

The maximum likelihood estimation of earthquake hazard parameters (maximum regional magnitudem _max, activity rate λ, and theb parameter in the Gutenberg-Richter distribution) is extended to the cases of incomplete and uncertain data. The method accepts mixed data containing only large (extreme) events and a variable quality of complete data with different threshold magnitude values. Uncertainty of earthquake magnitude is specified by two values, the lower and upper magnitude limits. It is assumed that such an interval contains the real unknown magnitude. The proposed approach allows the combination of different quality catalog parts, e.g. those where the assignment of magnitude is questionable and those with magnitudes precisely determined. As an illustration of the method, the seismic hazard analysis for western Norway and adjacent sea area (4–8°E, 58–64°N) is presented on the basis of the strongest earthquakes felt during the period 1831–1889 and three complete catalog parts, covering the period 1890–1987.