华北地区地震环境噪声特征研究

利用华北流动地震台阵观测的垂直分向连续波形数据, 通过计算功率谱密度和相应的概率密度函数, 对华北地区地震环境噪声特征进行了分析研究. 结果表明, 东部平原和沉积盆地2 Hz以上的高频环境噪声水平与全球新高噪声模型（NHNM）相近, 周期3——18 s的平均噪声水平低于NHNM和新低噪声模型（NLNM）的平均值; 山区及西部高原的高频噪声水平明显低于NHNM, 周期1——18 s的噪声水平大多明显低于NHNM和NLNM的平均值; 不同区域18 s以上周期的噪声水平差异相对较小. 流动地震台阵部分台站的环境噪声存在明显的昼夜变化, 个别台站噪声水平明显高于周边台站, 表明这些台站受人类活动干扰较大. 不同台站的噪声水平分析表明, 将台站布设在摆坑内, 能在一定程度上降低高频和低频段的噪声水平. 台站环境噪声特征的研究结果可为流动地震台阵观测数据质量的定量评估, 观测期间的台站优化调整等提供重要依据.      

Very broadband seismic background noise analysis of permanent good vaulted seismic stations

This paper describes the results of a preliminary study conducted to analyze seismic background noise at sites of recently deployed very broadband stations of the Egyptian National Seismological Network (ENSN). The main purpose of the study is to assess the effects of permanent seismic vault construction and also to establish characteristics and origin of seismic noise at those sites. Another goal of this study is to determine the time needed for noise at those sites to stabilize. The power spectral densities of background noise at short period band (SP), very broadband (VBB), and ultra long period band (ULP) for each component of each broadband seismometer deployed in the different investigated sites are calculated. A MATLAB code has been developed that manages data processing and data analysis and compares the results with the high-noise model (NHNM) and low-noise model (NLNM) of Peterson (1993). Based on the obtained analysis, the noise stability and the efficiency of each station to record regional and teleseismic events are measured. The results of this study could be used in the future to evaluate station quality, to improve those processes that require background noise values, such as automatic association, and to improve the estimation of station and network detection and location thresholds.     

Tilt Observations in the Normal Mode Frequency Band at the Geodynamic Observatory Cueva de los Verdes,Lanzarote

We have conducted observations with the aid of a seismo-tiltmeter station, which is based on the Ostrovsky pendulum and installed at the Geodynamic Observatory Cueva de los Verdes at Lanzarote Island since 1995. In this station the signal is separated into two frequency bands – tidal tilts (from 0 to 5 mHz) and ground oscillations in the frequency range of free Earths normal modes (from 0.2 to 5 mHz). The later band, called accelerometer channel, has additional amplification. We analyzed the background records in the frequency range of Earths free oscillations from August 2000 to September 2001, as well as, Earths normal modes after strong earthquakes. We found several distinctive persistent peaks in the spectra of background oscillations. Both amplitudes of distinguished peaks and noises have seasonal variations. We found that spectra of background oscillations are different in the frequency interval between 1.4 and 2.5 mHz for North- South and East-West components.     

NEW FILTERING METHODS WITH “VIBROSEIS”*

In order to obtain high resolution correlograms, it is of importance amongst other things to get reflection signals with large bandwidth. An advantage of the VIBROSEIS *** *** Trade Mark and Service book of the Continental Oil Company.
method is that the frequencies radiated by the vibrators can be matched to the transmission response of the subsurface involved. By choosing the right frequency range, the highest possible amplitude and most favourable form may be given to the reflection signals. In a reflection correlogram, individual signals cannot be considered in isolation. Signals of different origin are interfering with one another. They very often have different amplitudes, so that it may be desirable in many cases to filter out events of certain apparent velocity. With the VIBROSEIS method this may be achieved quite simply. All frequencies of the noise signal are uniformly suppressed. The advantage is that noise signals, e.g. refraction signals, which cannot be sufficiently attenuated by wavelength filtering, may be completely eliminated by this velocity filtering without affecting the bandwith of the desired signal. The total dynamic range of the tape recording can be used for the registration of wanted events. To perform this kind of filtering several vibrators are necessary in the field; each of them is controlled by an individual signal. There is an unavoidable error of static and dynamic corrections which causes the results of reflection measurements to deteriorate when using multiple coverage. High frequency components especially are seriously affected by destructive interference. This difficulty can be avoided by using a VIBROSEIS signal with high frequency component amplitudes supported. For the probability of error of corrections a normal distribution is assumed. A smoothed amplitude characteristic may be achieved after stacking. The amplitude characteristic of seismic devices is commonly reduced to about 100 cps bandwidth. For further improvement of resolution of VIBROSEIS correlograms it is necessary to apply special filtering methods. This is of particular interest when any kind of gain control is used to display weak events more clearly. With increasing amplification the sidelobes of the strong signals may reach the size of the weak events. In order to eliminate this effect, the amplitude characteristic of the VIBROSEIS signal is adjusted for optimum suppression of side-lobes.     

NEW FILTERING METHODS WITH “VIBROSEIS”*

In order to obtain high resolution correlograms, it is of importance amongst other things to get reflection signals with large bandwidth. An advantage of the VIBROSEIS *** *** Trade Mark and Service book of the Continental Oil Company.
method is that the frequencies radiated by the vibrators can be matched to the transmission response of the subsurface involved. By choosing the right frequency range, the highest possible amplitude and most favourable form may be given to the reflection signals. In a reflection correlogram, individual signals cannot be considered in isolation. Signals of different origin are interfering with one another. They very often have different amplitudes, so that it may be desirable in many cases to filter out events of certain apparent velocity. With the VIBROSEIS method this may be achieved quite simply. All frequencies of the noise signal are uniformly suppressed. The advantage is that noise signals, e.g. refraction signals, which cannot be sufficiently attenuated by wavelength filtering, may be completely eliminated by this velocity filtering without affecting the bandwith of the desired signal. The total dynamic range of the tape recording can be used for the registration of wanted events. To perform this kind of filtering several vibrators are necessary in the field; each of them is controlled by an individual signal. There is an unavoidable error of static and dynamic corrections which causes the results of reflection measurements to deteriorate when using multiple coverage. High frequency components especially are seriously affected by destructive interference. This difficulty can be avoided by using a VIBROSEIS signal with high frequency component amplitudes supported. For the probability of error of corrections a normal distribution is assumed. A smoothed amplitude characteristic may be achieved after stacking. The amplitude characteristic of seismic devices is commonly reduced to about 100 cps bandwidth. For further improvement of resolution of VIBROSEIS correlograms it is necessary to apply special filtering methods. This is of particular interest when any kind of gain control is used to display weak events more clearly. With increasing amplification the sidelobes of the strong signals may reach the size of the weak events. In order to eliminate this effect, the amplitude characteristic of the VIBROSEIS signal is adjusted for optimum suppression of side-lobes.     

Rescaled range (R/S) analysis on seismic activity parameters

IntroductionSeismicity analysis is an important branch of seismology, which focuses statistically on the study of time, space and magnitude distribution of a group of earthquakes (FU, 1997). Earthquakes as instability phenomena of the lithosphere, however, involve very complicated mechanics. The seismic activity level is not steady, but fluctuating time by time, which shows the alternative change of periods of quiescence and activity (FU, 1986; Mogi, 1984; ZHANG, 1987). Because of such com…     

地震活动性参数的变尺度（Rem>/Sem>）分析

将改变时间标度的方法应用于地震活动性分析中，选择具有不同空间尺度范围、时间范围和地震活动背景的现代仪器地震目录作为研究对象，分析了几个地震活动性参数——地震频度和地震发生时间间隔与时间标度的关系．这些地震活动性参数的极差和均方差之比与时间标度呈现出幂指数变化方式，且幂指数都大于0.5，表明地震的发生并非无记忆的泊松过程，而是具有随机和规律的双重特性．犎指数偏离0.5的程度可以衡量随机与规律成分所占比重，偏离越大，该序列的规律性成分越多，反之则越少．通过时间标度变换，可以较短时间的观测为基础，对变量未来的发生情况做出保守的估计．      

P-wave spectra of the Füzesgyarmat, eastern Hungary earthquake sequence

On 29–30 September 1996, an earthquake sequence occurred in the Füzesgyarmat region in eastern Hungary. The main shock had a magnitude of M_L = 3.2 and was felt with a maximum intensity 4 MSK. It was preceded by a foreshock with a magnitude of M_L = 2.8 and was followed, within six hours, by five aftershocks with magnitudes 2.1M_L 3.1. The dynamic source parameters of the Füzesgyarmat earthquake sequence have been derived from P-wave spectra of the Hungarian seismograph stations. The average of the obtained values at different stations shows that the main shock occurred on a fault length of 610 m, with relative displacement of 1.13 cm, stress drop of 7 bar and seismic moment of 3.96*10₂₁ dyne.cm. The main shock was small to yield data for a full mechanism solution and no reliable single fault plane solution could be obtained due to the low signal to noise ratio at the recording stations. The parameters of the foreshock are fault length of 560m, seismic moment of 2.09*10₂₁ dyne.cm, stress drop of 5.53 bar and relative displacement of 0.73 cm. The five aftershocks show source parameters similar to the foreshock stress drops (5.26 5.76 bar), fault lengths (415 L 600 m), seismic moments (8.36*10²⁰ M_o 2.31*10₂₁ dyne.cm) and relative displacements (0.52 0.91 cm).     

Sealing law of far-field spectra based on observed parameters of the specific barrier model

A set of acceleration source spectra is constructed using the observed parameters of the specific barrier model of Papageorgiou and Aki. The spectra show a significant departure from the ²-model at the high frequency range. Specifically, the high frequency spectral amplitudes of seismic excitation are higher as compared to the level predicted by the ²-model. This is also supported by other observational evidence. The high frequency amplitudes of acceleration scale proportionally to the square root of the rupture areaS, to the rupture spreading velocityv, and to the local strain drop (/) (=strain drop in between barriers). The local strain drop in between barriers is not related in a simple way to the global strain drop, which is the strain drop estimated by assuming that it is uniform over the entire rupture area. Consequently, the similarity law does not apply. Using the source spectra which we constructed, we derive expressions for high frequency amplitudes of acceleration such asa _rms anda _max. Close to the fault both are independent of fault dimensions and scale as (/µ)(f)^1/2, while away from the fault plane they scale asW ^1/2(/µ)(f)^1/2, whereW is the width of the fault and f is the effective bandwidth of the spectra.     

New Seismic Noise Models Obtained Using Very Broadband Stations

It has been two decades since the last comprehensive standard model of ambient earth noise was published Peterson (Observations and modelling of seismic background noise, US Geological Survey, open-file report 93–322, 1993). The PETERSON model was updated by analyzing the absolute quietest conditions for stations within the GSN (Berger et al. in J Geophys Res 109, 2005; Mcnamara and Buland in Bull Seism Soc Am 94:1517–1527, 2004; Ringler et al. in Seismol Res Lett 81(4) doi:10.1785/gssrl.81.4.605, 2010). Unfortunately, both the original model and the updated models did not include any deployed station in North Africa and Middle East, which reflects the noise levels within the desert environment of those regions. In this study, a survey was conducted to create a new seismic noise model from very broadband stations which recently deployed in North Africa. For this purpose, 1 year of continuous recording of seismic noise data of the Egyptian National Seismic Network (ENSN) was analyzed in order to create a new noise model. Seasonal and diurnal variations in noise spectra were recorded at each station. Moreover, we constructed a new noise model for each individual station. Finally, we obtained a new cumulative noise model for all the stations. We compared the new high-noise model (EHNM) and new low-noise model (ELNM) with both the high-noise model (NHNM) and low-noise model (NLNM) of Peterson (Observations and modelling of seismic background noise, US Geological Survey, open-file report 93–322, 1993). The obtained noise levels are considerably lower than low-noise model of Peterson (Observations and modelling of seismic background noise, US Geological Survey, open-file report 93–322, 1993) at ultra long period band (ULP band), but they are still below the high-noise model of Peterson (Observations and modelling of seismic background noise, US Geological Survey, open-file report 93–322, 1993). The results of this study could be considered as a first step to create permanent seismic noise models for North Africa and Middle East regions.     

The IDC Seismic,Hydroacoustic and Infrasound Global Low and High Noise Models

The International Data Centre (IDC) in Vienna, Austria, is determining, as part of automatic processing, sensor noise levels for all seismic, hydroacoustic, and infrasound (SHI) stations in the International Monitoring System (IMS) operated by the Provisional Technical Secretariat of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO). Sensor noise is being determined several times per day as a power spectral density (PSD) using the Welch overlapping method. Based on accumulated PSD statistics a probability density function (PDF) is also determined, from which low and high noise curves for each sensor are extracted. Global low and high noise curves as a function of frequency for each of the SHI technologies are determined as the minimum and maximum of the individual station low and high noise curves, respectively, taken over the entire network of contributing stations. An attempt is made to ensure that only correctly calibrated station data contributes to the global noise models by additionally considering various automatic detection statistics. In this paper global low and high noise curves for 2010 are presented for each of the SHI monitoring technologies. Except for a very slight deviation at the microseism peak, the seismic global low noise model returns identically the Peterson (1993) NLNM low noise curve. The global infrasonic low noise model is found to agree with that of Bowman et al. (2005, 2007) but disagrees with the revised results presented in Bowman et al. (2009) by a factor of 2 in the calculation of the PSD. The global hydroacoustic low and high noise curves are found to be in quantitative agreement with Urick’s oceanic ambient noise curves for light to heavy shipping. Whale noise is found to be a feature of the hydroacoustic high noise curves at around 15 and 25 Hz.     

Approximate expression for the hilbert transform of a certain class of functions and their application to the ray theory of seismic waves

Summary Approximate expressions for the Hilbert transform of the functionf(t)=exp(- ₀ ² t ²/²) cos( ₀ t+v) are determined. This function, given a suitable choice of the three parameters ₀, and v, approximates a wide class of seismic signals very well. The approximate expressions for the Hilbert transform enable very simple formulae to be given for the elementary seismograms of the individual seismic body waves (in the zero approximation of the ray theory). This accelerates the computation of ray theoretical seismograms considerably.     

Expert Assessment of the Displacements Provoked by Seismic Events: Case Study for the Sofia Metropolitan Area

— The paper discusses the very recent seismotectonic and geological investigations for the Sofia valley. It also reviews available historical and recent seismological data. One possible variant of seismic zoning (interpretation of available geological and seismological data) is used to evaluate uniform probability Fourier spectra of strong ground motion at Sofia City. The distribution of the maximum expected displacements, using 1-D models and SHAKE91 techniques for 50 years of exposure time is mapped. The maps of eigensolution results for 30 m, 50 m depths to the rock are illustrated. These issues are important for definition of design parameters for evaluation of seismic safety of underground infrastructure, as well as for tall structures.     

黑龙江省地震科学台阵背景噪声特征分析

2019年黑龙江省完成"一带一路"地震科学台阵项目中台址勘选工作,基于科学台阵中136个台址的地面运动噪声数据,通过计算不同频段范围内背景噪声记录的加速度功率谱密度,研究不同环境噪声下科学台阵记录数据的地噪声特征及其台基响应。结果表明：黑龙江西北和东南部地区地面运动噪声水平低,观测环境较好;中部和东北部地区噪声水平较高,大庆地区尤为严重。勘选结果真实反映了黑龙江区域内的背景噪声分布,使我们对本区域地噪声水平和干扰因素有了新的认识。     

地震噪声异常实时监测

本文采用福建省85个测震台站2012年全年噪声资料的垂直向记录作为研究对象,将噪声记录以每5min为单位进行分段,求出每小段的功率谱,应用概率分布函数方法绘出台站的PDF图,之后利用网格概率法确定出台站的高低噪声参照线。另外,根据85个台站的PDF图异常,将噪声异常分成四类：缺数异常、低噪处异常、高噪处异常、中噪处异常。依据四类异常的特征分别研究出四类异常的挑选方法,再将这四种挑选方法结合形成地震噪声实时监测系统。选取福建省85个测震台站2013年7月份的噪声记录进行验证,结果表明：85个台站应用地震噪声实时监测系统识别出来的异常正确率都达到90%以上,挑选效果很好,并可应用于台站噪声实时监测。     

Source spectra and RMS acceleration of Mexican subduction zone earthquakes

Strong motion (SM) data of six Mexican subduction zone earthquakes (6.4M _S8.1) recorded near the epicentral zone are analyzed to estimate their far-field source acceleration spectra at higher frequencies (f0.3 Hz). Apart from the usual corrections such as geometrical spreading (1/R), average radiation pattern (0.6), free surface amplification (a factor of 2), and equal partitioning of the energy into two orthogonal horizontal components (a factor of 1/ ), the observed spectra are corrected for a frequency dependentQ(Q=100f), a site dependent filter (e ^–kf ), and amplification ofS waves near the surface (a factor of about 2 atf2Hz). We takeR as the average distance from the rupture area to the site. If we model the high frequency plateau (f1 Hz) of the source spectra, by a point source ^–2-model, and interpret them in terms of Brune's model we obtain between 50 and 100 bars for all earthquakes. The low-frequency broadband teleseismicP wave spectra, corrected witht ^*=1.0 s, agrees within a factor of two with SM source spectra near 1 Hz. The ^–2-model is inadequate to explain the observed source spectra in a broad frequency range; these resemble spectra given byGusev (1983) with some differences.SM source acceleration spectra require significant corrections to explain observed spectra and RMS acceleration (a_rms) (a) at farther coastal sites for extended sources due to directivity effect and (b) at inland sites (100R200 km) because of unaccounted path and site amplification and/or invalidity of body-wave approximation. The observed spectra and a_rms at these sites are significantly greater than the predicted values from the estimated source spectra.     

A Simplified Technique for Simulating Wide-band Strong Ground Motion for Two Recent Himalayan Earthquakes

A simplified technique for simulation of wide-band strong motion based on simple regression relations and Empirical Greens Function (EGF) technique by Irikura (1986) is presented in this paper. The method uses the acceleration envelope as a shaping window for a filtered white Gaussian noise, to get the synthetic accelerogram from each subfault. Correction factors for slip of large and small events and transmission factors at each boundary of different layers are included in this synthetic accelerogram. The synthetic accelerogram obtained from each subfault is used as the Greens function to get resultant records. Simulations are made for the confirmed models of the Uttarkashi and the Chamoli earthquakes at a number of stations to get wide-band strong ground motion. The comparison of synthetic with the observed records over a wide range of frequencies for two different Himalayan earthquakes establishes the efficacy of the present technique.     

Detection capability of seismic network based on noise analysis and magnitude of completeness

Assessing the detection threshold of seismic networks becomes of increased importance namely in the context of monitoring induced seismicity due to underground operations. Achieving the maximum possible sensitivity of industrial seismic monitoring is a precondition for successful control of technological procedures. Similarly, the lowest detection threshold is desirable when monitoring the natural seismic activity aimed to imaging the fault structures in 3D and to understanding the ongoing processes in the crust. We compare the application of two different methods to the data of the seismic network WEBNET that monitors the earthquake swarm activity of the West-Bohemia/Vogtland region. First, we evaluate the absolute noise level and its possible non-stationary character that results in hampering the detectability of the seismic network by producing false alarms. This is realized by the statistical analysis of the noise amplitudes using the ratio of 99 and 95 percentiles. Second, the magnitude of completeness is determined for each of the nine stations by analysing the automatic detections of an intensive swarm period from August 2011. The magnitude–frequency distributions of all detected events and events detected at individual stations are compared to determine the magnitude of completeness at a selected completeness level. The resulting magnitude of completeness M _c of most of the stations varies between ?0.9 and ?0.5; an anomalous high M _c of 0.0 is found at the most distant station, which is probably due to inadequate correction for attenuation. We find that while the absolute noise level has no significant influence to the station sensitivity, the noise stationarity correlates with station sensitivity expressed in low magnitude of completeness and vice versa. This qualifies the method of analysing the stationary character of seismic noise as an effective tool for site surveying during the seismic station deployment.     

用时频峰值滤波方法消减地震勘探资料中随机噪声的初步研究

时频峰值滤波算法是一种新颖的基于时频分析的信号增强算法,能够有效地消除随机噪声,恢复有效波信息.本文将这种时频分析算法用于消除地震勘探资料中的随机噪声,对淹没于随机噪声下的40道共炮点记录进行时频峰值滤波,恢复出来的共炮点记录可以清楚地表现原始记录同相轴的位置.经过对40道中任选两道（即第21道和第7道）滤波前后的子波形态Wigner\|Ville分布、傅立叶振幅谱等的比较,可知仅在谷值和峰值点误差较大,子波带宽相对误差小于25％.仿真试验表明信噪比可达－7dB,说明该方法可以有效地消减地震资料中的随机噪声.     

流动地震观测背景噪声的台基响应

大规模流动地震台阵技术发展为高分辨率深部结构成像提供了重要基础,背景噪声是影响流动地震观测质量的关键因素. 为掌握流动地震观测噪声规律,发展流动地震观测降噪技术, 编制流动地震观测技术规范, 我们开展了针对不同台基流动地震观测背景噪声的观测实验与分析. 其中,山西省临汾市五个地点架设了共22个对比观测台站, 进行了超过一年半的连续观测. 通过计算不同频段范围内背景噪声记录的加速度功率谱密度, 研究了不同场地条件和环境噪声下流动地震观测台站的噪声特征及其台基响应,分析了不同台基处理方式对噪声的抑制效果. 结果表明:(1)高频人为噪声和长周期自然噪声是影响流动地震观测质量的主要噪声, 可以通过增加台基深度和改善台基处理方式等方法降低其影响; (2)增加台基深度能有效地降低长周期噪声和高频噪声, 2 m深坑能使高人为噪声台站各分量的高频频段和长周期频段分别降低5 dB和10 dB; (3)由于其不稳定性, 沙子台基的水平分量在长周期频段一般要高于摆墩台基5 dB, 流动地震观测中推荐使用摆墩台基; (4) 台站位置、台站内部温度和空气流动都是影响台站噪声的重要因素. 在此基础上提出了不同场地条件和噪声环境下的台基处理建议和适合国情的移动地震台阵台站建设参考方案, 有助于流动地震观测野外工作的标准化和规范化.