远程事件次声监测中的信号参数二维子空间计算方法

越来越多的观测发现,在地震、火山爆发、泥石流等重大自然灾害发生前,常产生异常的次声信号,这为地震及其他自然灾害的预报工作增加了一种可能的信息;同时,次声还是监测大气层、浅地表爆炸的有效手段.在自然灾害和爆炸事件次声监测中,慢度和方位角等参数对于源信号传播、定位以及源性质识别等工作具有重要意义.然而,目前的慢度和方位角等参数的算法——频率波数(FK)分析法,尚存在精度和分辨率不高等问题,特别是对多源次声信号的识别能力较差.为提高次声信号的监测精度,基于次声信号和噪声的子空间不相关性,构建了次声信号慢度和方位角二维子空间计算模型,并在此基础上提出了一种高分辨率次声信号二维子空间算法,仿真实验和实际数据的对比分析结果表明:本文提出的方法在精度和分辨率方面明显优于FK法,且能够更好地分离多源次声信号.     

Analysis of Signals from an Unique Ground-Truth Infrasound Source Observed at IMS Station IS26 in Southern Germany

Quantitative modeling of infrasound signals and development and verification of the corresponding atmospheric propagation models requires the use of well-calibrated sources. Numerous sources have been detected by the currently installed network of about 40 of the final 60 IMS infrasound stations. Besides non-nuclear explosions such as mining and quarry blasts and atmospheric phenomena like auroras, these sources include meteorites, volcanic eruptions and supersonic aircraft including re-entering spacecraft and rocket launches. All these sources of infrasound have one feature in common, in that their source parameters are not precisely known and the quantitative interpretation of the corresponding signals is therefore somewhat ambiguous. A source considered well-calibrated has been identified producing repeated infrasound signals at the IMS infrasound station IS26 in the Bavarian forest. The source results from propulsion tests of the ARIANE-5 rocket’s main engine at a testing facility near Heilbronn, southern Germany. The test facility is at a range of 320 km and a backazimuth of ~280° from IS26. Ground-truth information was obtained for nearly 100 tests conducted in a 5-year period. Review of the available data for IS26 revealed that at least 28 of these tests show signals above the background noise level. These signals are verified based on the consistency of various signal parameters, e.g., arrival times, durations, and estimates of propagation characteristics (backazimuth, apparent velocity). Signal levels observed are a factor of 2–8 above the noise and reach values of up to 250 mPa for peak amplitudes, and a factor of 2–3 less for RMS measurements. Furthermore, only tests conducted during the months from October to April produce observable signals, indicating a significant change in infrasound propagation conditions between summer and winter months.     

Detection and Analysis of Near-Surface Explosions on the Kola Peninsula

Seismic and infrasonic observations of signals from a sequence of near-surface explosions at a site on the Kola Peninsula have been analyzed. NORSAR’s automatic network processing of these events shows a significant scatter in the location estimates and, to improve the automatic classification of the events, we have performed full waveform cross-correlation on the data set. Although the signals from the different events share many characteristics, the waveforms do not exhibit a ripple-for-ripple correspondence and cross-correlation does not result in the classic delta-function indicative of repeating signals. Using recordings from the ARCES seismic array (250 km W of the events), we find that a correlation detector on a single channel or three-component station would not be able to detect subsequent events from this source without an unacceptable false alarm rate. However, performing the correlation on each channel of the full ARCES array, and stacking the resulting traces, generates a correlation detection statistic with a suppressed background level which is exceeded by many times its standard deviation on only very few occasions. Performing f-k analysis on the individual correlation coefficient traces, and rejecting detections indicating a non-zero slowness vector, results in a detection list with essentially no false alarms. Applying the algorithm to 8 years of continuous ARCES data identified over 350 events which we confidently assign to this sequence. The large event population provides additional confidence in relative travel-time estimates and this, together with the occurrence of many events between 2002 and 2004 when a temporary network was deployed in the region, reduces the variability in location estimates. The best seismic location estimate, incorporating phase information for many hundreds of events, is consistent with backazimuth measurements for infrasound arrivals at several stations at regional distances. At Lycksele, 800 km SW of the events, as well as at ARCES, infrasound is detected for most of the events in the summer and for few in the winter. At Apatity, some 230 km S of the estimated source location, infrasound is detected for most events. As a first step to providing a Ground Truth database for this useful source of infrasound, we provide the times of explosions for over 50 events spanning 1 year.     

Recent Advances and Difficulties of Infrasonic Wave Investigation in the Ionosphere

Acoustic waves have a remarkable ability to transfer energy from the ground up to the uppermost layers of the atmosphere. On the ground, there are many permanent sources of infrasound, and also pulsed and/or sporadic sources (e.g., sea waves, infrasonic and sonic noise of cities, lightning, earthquakes, explosions, etc.). The infrasonic waves carry away the major part of their energy upwards through the atmosphere. What are the consequences of the upward energy transfer? What heights of the atmosphere are supplied by energy from various sources of an infrasonic wave? In most cases, the answers to these questions are not well known at present. The only opportunity to monitor the propagation of an infrasonic wave to high altitudes is to watch for its influence on the ionospheric plasma. Unfortunately, most of standard equipment for ionospheric sounding, as a rule, cannot detect plasma fluctuations in the infrasonic range. Besides, the form of an infrasonic wave strongly varies during propagation due to nonlinear effects. However, the development of the Doppler method of radiosounding of the ionosphere has enabled progress to be made. Simultaneously, the ionospheric method for sensing aboveground and underground explosions has been developed. Its main advantage is the remote observation of an explosion in the near field zone by means of short radio waves, i.e., the radio sounding of the ionosphere directly above the explosion. The theory of propagation of an acoustic pulse produced by an explosion on the ground up to ionospheric heights has been developed better than the theory for other sources, and has been quantitatively confirmed by experiments. A review of some advances in the area of infrasound investigations at ionospheric heights is given and some current problems are presented.     

Detection of Impact Points of Fragments of Spent Launch Vehicle Stages Using Infrasound Direction-Finding Methods

The paper describes the principles and techniques used to detect signals propagating in the atmosphere in the infrasonic frequency range. Such signals can be generated by different sources: ground and atmospheric explosions, as well as objects moving in the atmosphere at supersonic speed (aircraft, rockets, bolides, fragments of spent stages of launch vehicles). Portable infrasound monitoring stations are described, each of which includes three spaced infrasonic microphones. Each such station makes it possible to determine three basic parameters of the detected infrasound signal, which are subsequently used to solve the direction- finding problem: the time of arrival of an infrasonic wave, the azimuth to the source in the horizontal plane, and the wave approach angle from the source of infrasonic waves to the Earth’s surface in the vertical plane. An acoustic detector used to extract useful signals against a noise background is described. The detector is based on an algorithm similar to the STA/LTA detection algorithm known in seismology. Examples of the operation of an acoustic detector with data obtained during real measurements are given. Passive infrasound direction-finding technology is described. It is based on mathematical modeling of the of infrasonic wave propagation in the atmosphere, which are generated by objects moving along possible trajectories; comparison of theoretical signals with real ones recorded by monitoring stations; and determination of the realized trajectories. The paper gives examples of experimental verification of the effectiveness of passive infrasound direction-finding technology for determining the impact points of the first and second stages of launch vehicles. It is shown that infrasound direction-finding systems makes it possible to reduce the estimated search area for launch vehicle fragments that fall to the Earth, significantly decrease the time and costs for their search and utilization, and mitigate the negative environmental impact of the rocket and space industry.     

Seismo-acoustic signals associated with degassing explosions recorded at Shishaldin Volcano,Alaska, 2003–2004

In summer 2003, a Chaparral Model 2 microphone was deployed at Shishaldin Volcano, Aleutian Islands, Alaska. The pressure sensor was co-located with a short-period seismometer on the volcano’s north flank at a distance of 6.62 km from the active summit vent. The seismo-acoustic data exhibit a correlation between impulsive acoustic signals (1–2 Pa) and long-period (LP, 1–2 Hz) earthquakes. Since it last erupted in 1999, Shishaldin has been characterized by sustained seismicity consisting of many hundreds to two thousand LP events per day. The activity is accompanied by up to ∼200 m high discrete gas puffs exiting the small summit vent, but no significant eruptive activity has been confirmed. The acoustic waveforms possess similarity throughout the data set (July 2003–November 2004) indicating a repetitive source mechanism. The simplicity of the acoustic waveforms, the impulsive onsets with relatively short (∼10–20 s) gradually decaying codas and the waveform similarities suggest that the acoustic pulses are generated at the fluid–air interface within an open-vent system. SO₂ measurements have revealed a low SO₂ flux, suggesting a hydrothermal system with magmatic gases leaking through. This hypothesis is supported by the steady-state nature of Shishaldin’s volcanic system since 1999. Time delays between the seismic LP and infrasound onsets were acquired from a representative day of seismo-acoustic data. A simple model was used to estimate source depths. The short seismo-acoustic delay times have revealed that the seismic and acoustic sources are co-located at a depth of 240±200 m below the crater rim. This shallow depth is confirmed by resonance of the upper portion of the open conduit, which produces standing waves with f=0.3 Hz in the acoustic waveform codas. The infrasound data has allowed us to relate Shishaldin’s LP earthquakes to degassing explosions, created by gas volume ruptures from a fluid–air interface.     

Characterizing complex eruptive activity at Santiaguito,Guatemala using infrasound semblance in networked arrays

We implement an infrasound semblance technique to identify acoustic sources originating from volcanic vents and apply the technique to the generally low-amplitude infrasound (< 3 Pa at 1 km) signals produced by Santiaguito dome in Guatemala. Semblance detection is demonstrated with data collected from two-element miniature arrays with ~ 30 m spacing between elements. The semblance technique is effective at identifying a range of eruptive phenomena, including pyroclastic-laden eruptions, vigorous degassing events, and rockfalls, even during periods of high wind contamination Many of the detected events are low in amplitude (tens of mPa) such that they are observed only by select arrays positioned with proximity and line-of-sight to the source. Larger events, such as the pyroclastic-laden eruptions, which occurred bi-hourly in 2009, were detected by all five arrays and produced an infrasonic signal that was correlated across the network. Network correlated events can be roughly located and map to the summit of the Caliente Vent where pyroclastic-laden eruptions originate. In general, the degree of Santiaguito infrasound event correlation is poor across the network, suggesting that complex source geometry contributes to asymmetric sound radiation.     

Analysis of infrasonic and seismic events related to the 1998 Vulcanian eruption at Sakurajima

We present results from a detailed analysis of seismic and infrasonic data recorded over a four day period prior to the Vulcanian eruptive event at Sakurajima volcano on May 19, 1998. Nearly one hundred seismic and infrasonic events were recorded on at least one of the nine seismic–infrasonic stations located within 3 km of the crater. Four unique seismic event types are recognized based on the spectral features of seismograms, including weak seismic tremor characterized by a 5–6 Hz peak mode that later shifted to 4–5 Hz. Long-period events are characterized by a short-duration, wide spectral band signal with an emergent, high-frequency onset followed by a wave coda lasting 15–20 s and a fundamental mode of 4.2–4.4 Hz. Values of Q for long-period events range between 10 and 22 suggesting that a gas-rich fluid was involved. Explosive events are the third seismic type, characterized by a narrow spectral band signal with an impulsive high-frequency onset followed by a 20–30 second wave coda and a peak mode of 4.0–4.4 Hz. Volcano-tectonic earthquakes are the fourth seismic type. Prior to May 19, 1998, only the tremor and explosion seismic events are found to have an infrasonic component. Like seismic tremor, infrasonic tremor is typically observed as a weak background signal. Explosive infrasonic events were recorded 10–15 s after the explosive seismic events and with audible explosions prior to May 19. On May 19, high-frequency impulsive infrasonic events occurred sporadically and as swarms within hours of the eruption. These infrasonic events are observed to be coincident with swarms of long-period seismic events. Video coverage during the seismic–infrasonic experiment recorded intermittent releases of gases and ash during times when seismic and acoustic events were recorded. The sequence of seismic and infrasonic events is interpreted as representing a gas-rich fluid moving through a series of cracks and conduits beneath the active summit crater.     

Source location variability and volcanic vent mapping with a small-aperture infrasound array at Stromboli Volcano,Italy

Stromboli Volcano in Italy is a persistently active, complex volcanic system. In May 2002 activity was confined to 3 major summit craters within which several active vents hosted multiple explosions each hour. During a 5-day field campaign an array of 3 low-frequency microphones was installed to investigate the coherent infrasound produced by degassing from these vents. Consistent phase lags across the 3 stations indicate distinct sources that are subsequently investigated to determine the associated vent location, apparent depth, and origin time. The cross-correlation routine allows for variations in comparison window length, waveform filtering bandwidth, and correlation and consistency thresholds, allowing for improved detection of certain types of degassing sources. Identification of activity at the various vents could be subsequently corroborated with 3 channels of synchronously acquired thermal data and video. During the May 2002 experiment persistent, energetic infrasound was observed from a passive degassing source within the Central Crater (CC) and transient infrasound, produced by discrete Strombolian explosions, was identified at 4 additional vents. The continuous infrasound produced by the CC exhibits variable frequency-dependent correlation lag times that are interpreted as a diffraction effect due to the acoustic radiators recessed location within a steep-walled crater. Such dispersion has important implications for accurate eruption source modeling because it indicates that infrasonic waveforms may be significantly filtered during propagation. Transient explosion signals from the Northeast Crater (NEC) and Southwest Crater (SWC) vents also exhibit dynamic correlation lag times, but this scatter may be more reasonably attributed to variable epicentral locations. Explosions from the NEC west vent, for instance, appear to emanate from a diffuse zone with a lateral extent in excess of 10 m.Editorial responsibility: R. Cioni     

北京地震前的异常次声波

观测并研究了2011年10月12日发生在北京海淀区的一次小地震前4天,五个次声监测站点接收到的异常次声波信号.这五路信号的波形一致,均为"N"形脉冲波,且持续时间基本一致,约在一个小时左右.基于Wigner-Ville分布方法对信号进行时频分析发现次声波能量主要集中在0.025 Hz的频率以下.五路信号间的相关系数均高达0.8左右.采用波束形成方法对信号源进行成像定位研究,其结果表明:该地震前异常次声波源的位置与地震发生时震中的位置相差约5 km.本文的分析结果说明了地震前可能有低频大气次声波的产生,研究这类次声波可能为地震的预测提供一种有价值的信息.     

Infrasonic tracking of large bubble bursts and ash venting at Erebus Volcano,Antarctica

Explosive degassing at Erebus Volcano produces infrasound that can be used to locate, characterize, and quantify eruptive activity from multiple vents. We use a three element distributed microphone network to pinpoint eruption sources and track the activity at the prominent vents through time. Eruptive mechanisms for both source types are analyzed in conjunction with the telemetered time-synced video imagery. We identify two commonly active vents corresponding to the large (often > 10-m diameter) bubble bursts at the free surface of a persistent phonolitic lava lake (‘Ray Lake’), and the less frequent ash-rich eruptions from a constricted vent (‘Active Vent’) located ∼ 80 m from the lava lake. During a 3-month study interval from 6 January to 13 April 2006 we identified and mapped more than 350 eruptive sources from the lava lake and 20 sources from the ash vent. Lava lake events are characterized by high-amplitude infrasonic transients that reflect rapid (less than a few s) acceleration and rupture of magma bubble films followed by an explosion of pressurized gases. Precise infrasonic localization of the lava lake events to accuracies of a few m indicates variable bubble source locations across a 40 by 50-m region spanning the lava lake. Spatial variability is corroborated by the video data. In contrast, degassing from the ash vent produces longer-duration (tens of s), lower amplitude transients that reflect diminished impulsivity and an extended degassing duration, features that are corroborated by video. Because infrasound networks can operate continuously in all weather conditions and during both diurnal and seasonal polar darkness, and are easily incorporated into automatic processing, they significantly contribute to the completeness and quantification of eruption catalogues for Erebus.     

云南腾冲火山区地震/次声台阵建设

在中央级公益性科研院所基本科研业务专项《地震重点监视区地震与次声联合台阵探测对比试验》支持下,在云南腾冲火山区建设了一个由4个子台组成的三角形地震/次声台阵,首次将宽频带地震计及次声传感器联合部署在强地震活跃区及火山区。本文详细介绍了腾冲火山区地震/次声台阵的建设及其观测系统组成,特别是与次声观测有关的次声降噪滤波系统,并提出了下一步研究工作。     

Reventador Volcano 2005: Eruptive activity inferred from seismo-acoustic observation

Reventador Volcano entered an eruptive phase in 2005 which included a wide variety of seismic and infrasonic activity. These are described and illustrated: volcano-tectonic, harmonic tremor, drumbeats, chugging and spasmodic tremor, long period and very long period events. The recording of this simultaneous activity on an array of three broadband, seismo-acoustic instruments provides detailed information of the state of the conduit and vent during this phase of volcanic eruption. Quasi-periodic tremor at Reventador is similar to that observed at other volcanoes and may be used as an indicator of vent aperture. Variations in the vibration modes of the volcano, frequency fluctuations and rapid temporal fluctuations suggest the influx of new material, choking of the vent and possible modification of the conduit geometry during explosions and effusion over a period of six weeks.     

Infrasound Detection of Large Mining Blasts in Kazakstan

—?We describe infrasonic observations recorded since October, 1997, at the Kurchatov Observatory in Kazakstan from large mining blasts in Kazakstan and Siberia. Seismic signals are regularly recorded on a 21-element cross-array from events located at the Ekibastuz mine, 250?km NW of Kurchatov. However, associated infrasonic detections are infrequent and appear to be seasonal, with maximum numbers of detections occurring during November to January. Raytracing through model atmosphere temperature and wind profiles predicts enhanced infrasound reception during the winter months, when the prevailing stratospheric winds blow towards Kurchatov. In addition, raytracing confirms that the first infrasound arrivals at Kurchatov propagate through the troposphere and are followed, some 50–70?s later, by a stratospheric arrival.     

Towards an Automatic Monitoring System of Infrasonic Events at Mt. Etna: Strategies for Source Location and Modeling

Active volcanoes characterized by open conduit conditions generate sonic and infrasonic signals, whose investigation provides useful information for both monitoring purposes and studying the dynamics of explosive processes. In this work, we discuss the automatic procedures implemented for a real-time application to the data acquired by a permanent network of five infrasound stations running at Mt. Etna volcano. The infrasound signals at Mt. Etna consist in amplitude transients, called infrasound events. The adopted procedure uses a multi-algorithm approach for event detection, counting, characterization and location. It is designed for an efficient and accurate processing of infrasound records provided by single-site and array stations. Moreover, the source mechanism of these events can be investigated off-line or in near real-time by using three different models: (1) Strombolian bubble; (2) resonating conduit and (3) Helmholtz resonator. The infrasound waveforms allow us to choose the most suitable model, to get quantitative information about the source and to follow the time evolution of the source parameters.     

2003年日本北海道8.0级地震次声波特征研究

利用在北京昌平安装的次声三点阵, 记录了2003年9月26日日本北海道地震的前兆次声波和震后次声波。 实际记录的P-t曲线及波速波向图, 经快速傅立叶变换得到其三维动态频谱。 结果表明: ① 强震前2～7天能测到振幅很强、 方向可测的地震前兆次声波。 其频谱特点是: 振幅由弱(10～15 Pa)逐渐加强到(50～80 Pa)或更大; 先为短周期波(10～30 min), 紧跟着长周期波(30～50 min), 然后又出现更长的周期波, 最后长短周期的波一起出现连成一片(2～62 min); ② 地震前兆次声波的产生和记录较容易, 而当地次声波和震中次声波记录和识别比较困难; ③ 若两地的三点阵波向相交, 可预测震中位置。 故地震前兆次声波的测量研究, 有可能发展成为临震预报中一种有效的新方法     

Mapping complex vent eruptive activity at Santiaguito,Guatemala using network infrasound semblance

Pyroclastic-laden explosive eruptions from Santiaguito Volcano (Guatemala) are vented from the 200-m diameter Caliente Dome summit and result in a superposition of spatially extensive and temporally sustained (tens of seconds to minutes) acoustic sources. A network of infrasonic microphones distributed on various sides of the volcano record distinct waveforms, which are poorly correlated across the network and suggestive of acoustic interference from multiple sources. Presuming the infrasound wavefield is a linear superposition of spatially and temporally distinct sub-events, we introduce a semblance mapping technique to recover the time history of the spatially evolving sources during successive time windows. Coincident high-resolution video footage corroborates that both rapid dome uplift and individual explosive pulses are likely sources of high semblance infrasound that are identifiable during short (2 s) time windows. This study suggests that complex and network-variable infrasound waveforms are produced whenever a volcanic vent source dimension is large compared to the wavelength of the sound being produced. Non-compact infrasound radiators are probably commonplace at silicic volcanic systems, where venting often occurs across a dome surface.     

2001年1月强震的前兆次声波测量及机理探讨

频谱分析结果表明 :(1) 2 0 0 1年 1月 (含两次 8级地震 )的测量结果与 1999年 1月～ 7月、 2 0 0 0年 4月～ 2 0 0 1年 4月观测结果具有相同的规律 ,强地震前约 10天内常能测到振幅很强、方向可测的地震前兆次声波。其三维动态频谱的典型特点是振幅由弱 (10Pa～ 2 0Pa)变强 (80Pa以上 ) ;最后的周期范围很宽 (2分～ 6 5分 ) ;波形明显不同于流星雨次声和大风次声。 (2 )强震前兆次声三维动态频谱的最大振幅不仅受震级大小影响 ,而且与震源距离及深度等其它因素都有关。 (3)强震前长周期波的产生机理 ,主要有断层预滑理论、断裂预扩展理论和流体撞击假说 ,而前兆次声波很可能是由地壳中的长周期波耦合到空气中形成。 (4 )岩石破裂实验证实主破裂前在岩石表面和空气中均收到较强的低频波。 (5 )若在北京和昆明各建足够大的三点阵测准波向 ,两直线相交 ,即可预测国内的震中位置 ,这就可能为临震预报提供一种较有效的新方法     

Acoustic source characterization of impulsive Strombolian eruptions from the Mount Erebus lava lake

We invert for acoustic source volume outflux and momentum imparted to the atmosphere using an infrasonic network distributed about the erupting lava lake at Mount Erebus, Ross Island, Antarctica. By modeling these relatively simple eruptions as monopole point sources we estimate explosively ejected gas volumes that range from 1,000 m³ to 24,000 m³ for 312 lava lake eruptions recorded between January 6 and April 13, 2006. Though these volumes are compatible with bubble volumes at rupture (as estimated from explosion video records), departures from isotropic radiation are evident in the recorded acoustic wavefield for many eruptions. A point-source acoustic dipole component with arbitrary axis orientation and strength provides precise fit to the recorded infrasound. This dipole source axis, corresponding to the axis of inferred short-duration material jetting, varies significantly between events. Physical interpretation of dipole orientation as being indicative of eruptive directivity is corroborated by directional emissions of ejecta observed in Erebus eruption video footage. Although three azimuthally distributed stations are insufficient to fully characterize the eruptive acoustic source we speculate that a monopole with a minor amount of oriented dipole radiation may reasonably model the primary features of the recorded infrasound for these eruptions.     

Monitoring Seismo-volcanic and Infrasonic Signals at Volcanoes: Mt. Etna Case Study

Volcanoes generate a broad range of seismo-volcanic and infrasonic signals, whose features and variations are often closely related to volcanic activity. The study of these signals is hence very useful in the monitoring and investigation of volcano dynamics. The analysis of seismo-volcanic and infrasonic signals requires specifically developed techniques due to their unique characteristics, which are generally quite distinct compared with tectonic and volcano-tectonic earthquakes. In this work, we describe analysis methods used to detect and locate seismo-volcanic and infrasonic signals at Mt. Etna. Volcanic tremor sources are located using a method based on spatial seismic amplitude distribution, assuming propagation in a homogeneous medium. The tremor source is found by calculating the goodness of the linear regression fit (R ²) of the log-linearized equation of the seismic amplitude decay with distance. The location method for long-period events is based on the joint computation of semblance and R ² values, and the location method of very long-period events is based on the application of radial semblance. Infrasonic events and tremor are located by semblance–brightness- and semblance-based methods, respectively. The techniques described here can also be applied to other volcanoes and do not require particular network geometries (such as arrays) but rather simple sparse networks. Using the source locations of all the considered signals, we were able to reconstruct the shallow plumbing system (above sea level) during 2011.