首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 46 毫秒
1.
The problem of discriminating between earthquakes and underground nuclear explosions is formulated as a problem in pattern recognition. As such it may be separated into two stages, feature extraction and classification. The short-period (SP) features consist of mb and autoregressive parameters characterising the preceding noise, signal and coda. The long-period (LP) features consist of LP power spectral estimates taken within various group velocity windows. Contrary to common usage we have extracted features from horizontal Rayleigh waves and Love waves as well as vertical Rayleigh waves. The classification is performed by approximating the statistical distribution of earthquake and explosion feature vectors by multivariate normal distributions.The method has been tested on a data base containing 52 explosions and 73 earthquakes from Eurasia recorded at NORSAR between 1971 and 1975. Several of these events are difficult on the mb : Ms diagram [mb(PDE) and Ms (NORSAR) have been used]. The data set was divided into a learning and an independent data set. All of the events both from the learning data set and the independent data set were correctly classified using the new procedures. Furthermore, the increase in separation as compared to the mb : Ms discriminant is significant.  相似文献   

2.
Abstract

This paper describes a fuzzy rule-based approach applied for reconstruction of missing precipitation events. The working rules are formulated from a set of past observations using an adaptive algorithm. A case study is carried out using the data from three precipitation stations in northern Italy. The study evaluates the performance of this approach compared with an artificial neural network and a traditional statistical approach. The results indicate that, within the parameter sub-space where its rules are trained, the fuzzy rule-based model provided solutions with low mean square error between observations and predictions. The problems that have yet to be addressed are overfitting and applicability outside the range of training data.  相似文献   

3.
基于样本增强的卷积神经网络震相拾取方法   总被引：2，自引：2，他引：0       下载免费PDF全文
李安  杨建思  彭朝勇  郑钰  刘莎 《地震学报》2020,42(2):163-176
为了快速、高效地从地震数据中识别地震事件和拾取震相,本文利用基于样本增强的卷积神经网络自动震相拾取方法,将西藏林芝地区L0230台站3个月数据作为训练集,该区内另外6个台站连续1个月的波形数据作为测试集,采用高斯噪声、随机噪声拼接、随机挑选噪声、随机截取地震事件等4种样本增强的方法扩增训练集,以提高自动震相拾取技术的准确率。结果显示:样本增强前模型在测试集上的地震事件识别准确率为80%,样本增强后提升至97%,表明样本增强有效地提高了模型的泛化性能和抗干扰能力;在0.5 s误差范围内,震相自动拾取准确率高于81%,在1.0 s误差范围内,准确率高于95%;利用基于样本增强的卷积神经网络震相拾取方法能够检测出人工拾取震相中误标和漏检的震相。   相似文献   

4.
The complex seismotectonic studies of the pleistoseist area of the Ilin-Tas earthquake (Ms = 6.9), one of the strongest seismic events ever recorded by the regional seismic network in northeastern Russia, are carried out. The structural tectonic position, morphotectonic features of present-day topography, active faults, and types of Cenozoic deformations of the epicentral zone are analyzed. The data of the instrumental observations are summarized, and the manifestations of the strong seismic events in the Yana–Indigirka segment of the Cherskii seismotectonic zone are considered. The explanation is suggested for the dynamical tectonic setting responsible for the Andrei-Tas seismic maximum. This setting is created by the influence of the Kolyma–Omolon indenter, which intrudes into the Cherskii seismotectonic zone from the region of the North American lithospheric plate and forms the main seismogenic structures of the Yana–Indigirka segment in the frontal zone (the Ilin-Tas anticlinorium). The highest seismic potential is noted in the Andrei- Tas block—the focus of the main tectonic impacts from the Kolyma–Omolon superterrane. The general trend of this block coincides with the orientation of the major axis of isoseismal ellipses (azimuth 50°–85°), which were determined from the observations of macroseismic effects on the ground after the Uyandina (Ms = 5.6), Andrei-Tas (Ms = 6.1), and Ilin-Tas (Ms = 6.9) earthquakes.  相似文献   

5.
This paper describes the seismic activity of Vulcano - an active volcano of the Aeolian Islands group - from 1977 to 1980. Two main swarms of low energy seismic events are recognizable on the basis of the entire set of data. The events, localized very near the crater of Vulcano where a fumarole field is present, have been classified into two classes:i) very shallow events (as far as 1 km in depth); andii) deeper events (as far as 3 km in depth). Double events and small earthquakes localized on the rim of the old caldera have also been recorded in a little percentage. The seismic events of Vulcano can be attributed to an intensive fumarolic activity.  相似文献   

6.
Preliminary reference Earth model   总被引：29，自引：0，他引：29  
A large data set consisting of about 1000 normal mode periods, 500 summary travel time observations, 100 normal mode Q values, mass and moment of inertia have been inverted to obtain the radial distribution of elastic properties, Q values and density in the Earth's interior. The data set was supplemented with a special study of 12 years of ISC phase data which yielded an additional 1.75 × 106 travel time observations for P and S waves. In order to obtain satisfactory agreement with the entire data set we were required to take into account anelastic dispersion. The introduction of transverse isotropy into the outer 220 km of the mantle was required in order to satisfy the shorter period fundamental toroidal and spheroidal modes. This anisotropy also improved the fit of the larger data set. The horizontal and vertical velocities in the upper mantle differ by 2–4%, both for P and S waves. The mantle below 220 km is not required to be anisotropic. Mantle Rayleigh waves are surprisingly sensitive to compressional velocity in the upper mantle. High Sn velocities, low Pn velocities and a pronounced low-velocity zone are features of most global inversion models that are suppressed when anisotropy is allowed for in the inversion.The Preliminary Reference Earth Model, PREM, and auxiliary tables showing fits to the data are presented.  相似文献   

7.
Results of fractal analysis of ultra-low-frequency (ULF) emissions registered at a low-latitude observatory, Guam (geomagnetic coordinates Φm=9°N, Λm=225°), and at a high-latitude drifting station, North Pole-30 (Φm=75°N, Λm=172°), are presented. The first set of data covers a long period of observations (20 months) including the strong (Ms=8) Guam earthquake of 8 August 1993. The second set of data covers a short period of observations (21 days) in April 1989 during the preparation phase of the big magnetic storm of 25 April 1989. Definite peculiarities in the behavior of ULF emission scaling (fractal) characteristics have been found, which are discussed on the basis of the self-organized criticality concept. The principal common peculiarity for magnetosphere–ionosphere and lithosphere systems is detection of flicker noise (β∼1, D0∼2) in a certain frequency range on the preparation phase of strong magnetospheric and seismic events.  相似文献   

8.
Free oscillation and body wave data are used to construct average Q models for the earth. The data set includes fundamental and overtone observations of the radial, spheroidal and toroidal modes, ScS observations and amplitudes of body waves as a function of distance. The preferred model includes a low-Q zone at both the top and the bottom of the mantle. In these regions the seismic velocities are likely to be frequency dependent in the “seismic” band. Absorption in the mantle is predominantly due to losses in shear. Compressional absorption may be important in the inner core.A grain-boundary relaxation model is proposed that explains the dominance of shear over compressional dissipation, the roughly frequency independent average values for Q and the variation of Q with depth. In the high-Q regions, the lithosphere and the midmantle (200–2000 km), Q is predicted to be frequency dependent. However, the low-Q regions of the earth, where Q is roughly frequency independent, dominate the observations of attenuation.  相似文献   

9.
Tomographic Imaging of Lg and Sn Propagation in the Middle East   总被引：1，自引：0，他引：1  
?—?Observations based on relatively limited data recorded by sparsely distributed stations have indicated that regional seismic phase propagation (Lg and Sn) is very complex in the Middle East. Accurate characterization of regional seismic wave propagation in this region necessitates the use of a large number of seismic stations. We have compiled a large data set of regional and local seismograms recorded in the Middle East. This data set comprises approximately four years of data from national short-period networks in Turkey and Syria, data from temporary broadband arrays in Saudi Arabia and the Caspian Sea region, and data from GSN, MEDNET, and GEOFON stations in the Middle East. We have used this data set to decipher the character and pattern of regional seismic wave propagation. We have mapped zones of blockage as well as inefficient and efficient propagation for Lg, Pg, and Sn throughout the Middle East. Two tomographic techniques have been developed in order to objectively determine regions of lithospheric attenuation in the Middle East.¶We observe evidence of major increase in Lg attenuation, relative to Pg, across the Bitlis suture and the Zagros fold and thrust belt, corresponding to the boundary between the Arabian and Eurasian plates. We also observe a zone of inefficient Sn propagation along the Dead Sea fault system which coincides with low Pn velocities along most of the Dead Sea fault system and with previous observations of poor Sn propagation in western Jordan. Our observations indicate that in the northern portion of the Arabian plate (south of the Bitlis suture) there is also a zone of inefficient Sn propagation that would not have been predicted from prior measurements of relatively low Pn velocities. Mapped high attenuation of Sn correlates well with regions of Cenozoic and Holocene basaltic volcanism. These regions of uppermost mantle shear-wave attenuation most probably have anomously hot and possibly thin lithosphere.  相似文献   

10.
This study employs an event location algorithm based on grid search to investigate the possibility of improving seismic event location accuracy by using non-Gaussian error models. The primary departure from the Gaussian error model that is considered is the explicit use of non-Gaussian probability distributions in defining optimal estimates of location parameters. Specifically, the class of generalized Gaussian distributions is considered, which leads to the minimization of Lp norms of arrival time residuals for arbitrary p≥1. The generalized Gaussian error models are implemented both with fixed standard errors assigned to the data and with an iterative reweighting of the data on a station/phase basis. An implicit departure from a Gaussian model is also considered, namely, the use of a simple outlier rejection criterion for disassociating arrivals during the grid-search process. These various mechanisms were applied to the ISC phase picks for the IWREF reference events, and the resulting grid-search solutions were compared to the GT locations of the events as well as the ISC solutions. The results indicate that event mislocations resulting from the minimization of Lp residual norms, with p near 1, are generally better than those resulting from the conventional L2 norm minimization (Gaussian error assumption). However, this result did not always hold for mislocations in event depth. Further, outlier rejection and iterative reweighting, applied with L2 minimization, performed nearly as well as L1 minimization in some cases. The results of this study suggest that ISC can potentially improve its location capability with the use of global search methods and non-Gaussian error models. However, given the limitations of this study, further research, including the investigation of other statistical and optimization techniques not addressed here, is needed to assess this potential more completely.  相似文献   

11.
We did a Study of Horizontal-to-Vertical Component Spectral Ratio in the Tehran seismic zone. Micro-earthquakes, microtremors and quarry blasts data were used as an estimation of the site response in the Tehran zone. Site effects were studied based on horizontal to vertical ratios by the Nakamura׳s technique. Also, we used the spectra of signals for three components with the lowest noise levels for spectral slope studies. The analysis used seismic events from a network of 13 seismic stations by the permanent local seismological network of the Tehran Disaster Mitigation and Management Organization (TDMMO) from 2004 to 2007. The number of events used were different for each station. Quarry blast events were with 1.2≤ML≤2.2 and micro-earthquakes were with 1.1≤ML≤4.1.By comparing results for earthquake, microtremor and quarry blast, we could see that there is a significant difference between them. The data showed clear observations, especially in high-frequencies. The H/V spectral ratios indicate dominant frequency for rock/soft site with a higher ratio level for quarry blast ratios, which are comparable to the earthquake results due to their difference sources. The results derived by spectral H/V ratios and spectral analysis may be used to distinguish between local earthquakes and quarry blasts.  相似文献   

12.
Seismic noise attenuation is very important for seismic data analysis and interpretation, especially for 3D seismic data. In this paper, we propose a novel method for 3D seismic random noise attenuation by applying noncausal regularized nonstationary autoregression (NRNA) in f–x–y domain. The proposed method, 3D NRNA (f–x–y domain) is the extended version of 2D NRNA (f–x domain). f–x–y NRNA can adaptively estimate seismic events of which slopes vary in 3D space. The key idea of this paper is to consider that the central trace can be predicted by all around this trace from all directions in 3D seismic cube, while the 2D f–x NRNA just considers that the middle trace can be predicted by adjacent traces along one space direction. 3D f–x–y NRNA uses more information from circumjacent traces than 2D f–x NRNA to estimate signals. Shaping regularization technology guarantees that the nonstationary autoregression problem can be realizable in mathematics with high computational efficiency. Synthetic and field data examples demonstrate that, compared with f–x NRNA method, f–x–y NRNA can be more effective in suppressing random noise and improve trace-by-trace consistency, which are useful in conjunction with interactive interpretation and auto-picking tools such as automatic event tracking.  相似文献   

13.
Seismicity and Casing Failures Due to Steam Stimulation in Oil Sands   总被引：1，自引：0，他引：1  
—This paper describes observations of seismicity and casing failures associated with steam stimulation operations at Imperial Oil Ltd.’s Cold Lake oil field in Alberta, Canada. A total of 11 oil-producing pads were monitored over a 1–2 year period using 3-component geophones cemented at depths ranging from 160 m to 400 m and data acquisition systems with a flat frequency response up to 1.5 kHz. Most of the seismicity was detected during the steaming operations and was located in the formation overlying the oil-bearing layer. Some activity was observed in the shales above, however, the reservoir itself showed almost no evidence of seismicity. The estimated seismic moment of the observed events was in the range 105–107 N·m (?2.7 < M < ?1.3). According to a theoretical model (Talebi et al., 1998) and in situ observations, the seismic source corresponding to casing failure events should be well described by a dipole registering seismic moment in the order of 2 · 106 N·m. Seismic signals of a total of four observed casing failures were analyzed. The partial failures produced seismic moments slightly lower than this value while total failures were stronger by about one order of magnitude. The use of the SV/SH amplitude ratio, in conjunction with accurate source locations, provided a robust technique for the detection of casing failures.  相似文献   

14.
Repeatability of seismic data plays a crucial role in time‐lapse seismic analysis. There are several factors that can decrease the repeatability, such as positioning errors, varying tide, source variations, velocity changes in the water layer (marine data) and undesired effects of various processing steps. In this work, the complexity of overburden structure, as an inherent parameter that can affect the repeatability, is studied. A multi‐azimuth three‐dimensional vertical‐seismic‐profiling data set with 10 000 shots is used to study the relationship between overburden structure and repeatability of seismic data. In most repeatability studies, two data sets are compared, but here a single data set has been used because a significant proportion of the 10 000 shots are so close to each other that a repeatability versus positioning error is possible. We find that the repeatability decreases by a factor of approximately 2 under an overburden lens. Furthermore, we find that the X‐ and Y‐components have approximately the same sensitivity to positioning errors as the Z‐component (for the same events) in this three‐dimensional vertical‐seismic‐profiling experiment. This indicates that in an area with complex overburden, positioning errors between monitor and base seismic surveys are significantly more critical than outside such an area. This study is based on a three‐dimensional three‐component vertical‐seismic‐profiling data set from a North Sea reservoir and care should be taken when extrapolating these observations into a general four‐dimensional framework.  相似文献   

15.
Operational earthquake forecasting (OEF) relies on real-time monitoring of seismic activity in an area of interest to provide constant (e.g., daily) updates of the expected number of events exceeding a certain magnitude threshold in a given time window (e.g., 1 week). It has been demonstrated that the rates from OEF can be used to estimate expected values of the seismic losses in the same time interval OEF refers to. This is a procedure recently defined as operational earthquake loss forecasting (OELF), which may be the basis for rational short-term seismic risk assessment and management. In Italy, an experimental OELF system, named MANTIS-K, is currently under testing. It is based on weekly rates of earthquakes exceeding magnitude (M) 4, which are updated once a day or right after the occurrence in the country of an M 3.5+ earthquake. It also relies on large-scale structural vulnerability and exposure data, which serve to the system to provide continuously the weekly expected number of: (1) collapsed buildings, (2) displaced residents, and (3) casualties. While the probabilistic basis of MANTIS-K was described in previous work, in this study OELF is critically discussed with respect to three recent Italian seismic sequences. The aim is threefold: (1) illustrating all the features of the OELF system in place; (2) providing insights to evaluate whether if it would have been a useful additional tool for short-term management; (3) recognizing common features, if any, among the losses computed for different sequences.  相似文献   

16.
—?A crustal velocity model has been developed for Fennoscandia, the Baltic shield and adjacent areas. This model represents a simplified average of various models developed for parts of this region. We show that P-wave travel times calculated with this model provide an excellent fit to observations at the Fennoscandian, KRSC and IRIS station networks for a set of seismic events with known or very well-constrained locations. The station-event paths cover large parts of Western Russia and the Barents Sea, thus indicating that this model, which we denote the Barents model, is appropriate for this entire region. We show by examples that significant improvements in event location precision can be achieved compared to using the IASPEI model. We finally use the Barents model to calculate locations of recent small seismic events in the Novaya Zemlya region of interest in a CTBT monitoring context.  相似文献   

17.
For faster and more robust ray tracing in 1-D velocity models and also due to the lack of reliable 3-D models, most seismological centers use 1-D models for routine earthquake locations. In this study, as solution to the coupled hypocenter-velocity problem, we compute a regional P-wave velocity model for southern Iran that can be used for routine earthquake location and also a reference initial model for 3-D seismic tomography. The inversion process was based on travel time data from local earthquakes paired reports obtained by merging the catalogues of Iranian Seismic Center (IRSC, 6422 events) and by the Broadband Iranian National Seismic Network (BIN, 4333 events) for southern Iran in the period 2006 through July 2017. After cleaning the data set from large individual reading errors and by identifying event reports from both networks belonging to same earthquake (a process called event pairing), we obtained a data set of 1115 well-locatable events with a total number of 24,606 P-wave observations. This data set was used to calculate a regional minimum 1-D model for southern Iran as result of an extensive model search by trial-and-error process including several dozens of inversions. Significantly different from previous models, we find a smoothly increasing P-velocity by depth with velocities of 5.8 km/s at shallow and velocities of 6.4 km/s at deepest crustal levels. For well-locatable events, location uncertainties are estimated in the order of ±?3 km for epicenter and double this uncertainty for hypocentral depth. The use of the minimum 1-D model with appropriate station delays in routine hypocenter location processing will yield a high-quality seismic catalogue with consistent uncertainty estimates across the region and it will also allow detection of outlier observations. Based on the two catalogues by IRSC and BIN and using the minimum 1-D model and station delays for all stations in the region, we established a new combined earthquake catalogue for southern Iran. While the general distribution of the seismicity corresponds well with that of the two individual catalogues by IRSC and BIN, the new catalogue significantly enhances the correlation of seismicity with the regional fault systems within and between the major crustal blocks that as an assembly build this continental region. Furthermore, the unified seismic catalogue and the minimum 1-D model resulting from this study provide important ingredients for seismic hazard studies.  相似文献   

18.
In recent year, there has been growing interest in the possible use of electromagnetic observations to study earthquakes and possible precursors prior to seismic activity, in response to the success in United States, Japan, Russia, China, and other countries using seismo-electromagnetic methods. We have established a new experimental setup (i.e., biopotential sensor) in Farah region (geographic coordinates: 27.17°N, 77.47°E), Mathura, India. The setup has started operating and analyzed the data since November 2011. The data have been tested by various methods and a good correlation with seismic events was found; thus, a real-time analysis from 21:00 p.m. through 8:00 a.m. every day was initiated. First, we recorded the amplitude enhancement in bio-potential and found positive correlation with seismic activities (near Delhi and Rajasthan) and analyzed the data with solar flares and magnetic storms during the same period, finding a negative correlation of these events. The studies of these events are in progress with statistical analysis of the data. We chose the observing site in Farah region because this region is well known for being a site of a conductive channel of seismic activity.  相似文献   

19.
Based on the seismological, geodynamic, and seismotectonic data on the strongest series of earthquakes in 2009–2017 in Central Italy, which were collected by many researchers, mostly Italian, the tectonic position of these events is determined and the seismic history of the region over more than 2000 years of observations of seismic manifestations in Italy is traced in the context of the strong events of the beginning of the 21st century. The aftershock processes of these earthquakes are investigated and, as a result, the possibility of a series of strong aftershocks of the earthquake of October 30, 2016 (MW = 6.6) is predicted in advance of the actual occurrence of these events on January 18, 2017.  相似文献   

20.
—?A program package, called HYPOSAT, has been under development that attempts to use the maximum information possible to estimate the hypocenter of a seismic source. The standard input parameters can be used: arrival times of first and later onsets with backazimuths and ray parameters (or apparent velocities). In addition, travel-time differences between different phases observed at the same station can be optionally used. The observed standard deviations are used to weight all input parameters and the inversion is done with a generalized matrix inversion code.¶A starting solution with a priori uncertainties can be calculated as the intersection of all backazimuth observations. If S observations are also available, a preliminary origin time is estimated using Wadati's approach to estimate a source time.¶Global earth models and user-defined horizontally layered local or regional models can be used alone or together to locate seismic events. To gain the best result from all input data, observations of all seismic phases as defined in the IASPEI91 tables can be inverted. Station corrections and corrections for phases with reflection points at the earth's surface can be applied by using local velocity structures.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号