NG41A-0140
Modeling of Active Monitoring with Harmonic Vibrational Signals
The active seismic monitoring is based on the sounding of the geological medium by the seismic waves from controllable vibrational sources. Vibrational sources can generate stationary harmonic wave fields, which can be effectively used to monitor the variability of the medium in time. A stationary harmonic wave field is formed in the medium due to the long time radiation by the vibrator of a harmonic signal with constant characteristics, such as frequency, amplitude, and phase. The direct problem of modeling of active monitoring with harmonic vibrational signals is solved. It is determined the change in the characteristics of the stationary wave field recorded at the surface cause by the changes in the density and velocity of seismic waves in some inner area of the medium. The results of a mathematical modeling of vibroseismic monitoring of changes in the elastic characteristics in the interior Earth's crust zone in the approximation of the wave equation and of a model of the Earth's crust-mantle system in the form of a layer at a half-space with different velocity values of elastic waves are presented. It is assumed that the vibrational source is a point and harmonic one, with a constant oscillation frequency, and that the zone of changes of the characteristics in the medium is spherical. The wave field in the medium is calculated in the ray approximation. Wave field variations in the medium and at the free surface are determined for the case of small velocity changes in the spherical region by calculating the beam pattern of a fictitious 3D source in diffraction approach. As a result of the modeling the estimation of sensitivity of active monitoring method with harmonic vibrational signals is done.
NG41A-0141
Vibroseismic Investigations of Mud Volcano, Experimental Results and Mathematical Modeling
The theoretical and experimental fundamentals of the monitoring system of living volcanoes with the use of powerful vibroseismic sources are presented. A method to calculate Green's function for inhomogeneous models of media with the employment of the reciprocity principle was developed for the numerical modeling of arbitrary structures, in particular, magma chambers. The medium is assumed to be consisting of arbitrary blocks. Some part of the medium can be a layered pack with an arbitrary number of layers. The thickness of the layers and the velocity parameters of the media can be arbitrary quantities in both the layered pack and the block medium. It should be noted that the calculation of Green's function allows simultaneous calculation of the wave fields for an arbitrary number of sources and receivers. This is important also for problems of area seismic prospecting and for solving inverse problems of geophysics by using optimization methods. The results of first experimental works for the Taman mud volcano province are presented. The medium's model in the mud volcano Shugo region obtained on the basis of vibroseismic sounding data was refined by numerical modeling.
NG41A-0142
Vibroseismic Observations in Seismically Active Zones of Siberia
High-power vibration sources were developed in Siberian Branch of the Russian Academy of Sciences during 30 years for use in active seismology and studies of Earth's deep structure. Most of data is obtained using eccentric 40- and 100-ton vibrators. Current research involving these sources covers Baikal rift zone, Altay-Sayan folded area and Okhotsk-Chukotski regions in Russia. Using one week to 15-20 days recording intervals, the vibroseismic observations were repeated for several years in the south of Baikal Lake and in Novosibirsk region in the framework of active monitoring technology. The total area of vibroseismic monitoring in the south of Lake Baikal exceeds 20000 km2. Such spatial scale allows to control stress changes in the zones of large faults (Obruchevski, Primorski, Bolsherechenski), which repeatedly activate during rift genesis. Near Novosibirsk, the system of vibroseismic monitoring is focused on changes of the Earth crust physical characteristics caused by seasonal changes of water levels in Novosibirsk reservoir. It is oriented transversely and along the spread of reservoir at 50-70 km distances from the source location. Variations of elastic waves correlate with process of earthquake preparation in 2002 and were detected in the south of Lake Baikal. Results of long-term experiment near Novosibirsk suggest that the observed 0.3-0.6% variations in velocities of longitudinal and shear waves are connected with seasonal changes of water levels (up to 5m) in the biggest reservoir. Overall results suggest possibility of vibroseismic monitoring with small number of high-power vibrators and a large number of recording stations.
NG41A-0143
Powerful Vibro-Sources for Active Seismology and Deep Earth Sounding
Success of active seismology in deep probing of seismically-active zones depends on parameters of used sources. It is unacceptable to use strong explosions or powerful stationary vibration sources for this purpose because of ecologic reasons as well as poor spatial coverage. Low-power vibration sources (which are commonly used for seismic prospecting purposes) are also ineffective, because of low translucence aperture and small penetration depths. Field investigations using powerful (40-60 tons) transportable vibrators have been carried out by Siberian Branch of the Russian Academy of Sciences since 1989. For the period of 20 years, a considerable data volume is accumulated in various regions of Siberia (Sayan region, Altay-Sayan and Okhotsk-Chukotski regions). Use of powerful transportable vibro-sources was developed into effective working technology. Optimal transportation unit was developed for work in hard-to-reach areas on the base of high performance cross-country vehicles. Data processing revealed quality wave fields for both longitudinal and shear waves reflected from a reference boundary in the Erath crust and Moho which were recorded for 0 – 300-400 km source-receiver offsets. Data show high stability and repeatability. Many data sets were recorded using both explosions and vibro-sources at different offsets and in presence of tectonically active zones. A database of records registered in near-field zone was accumulated for various geology (from low-velocity sediments to crystal rocks). We present some examples of using powerful transportable vibrators, including deep sounding and monitoring data, recording equipment, and observation systems.
NG41A-0144
Instant Variations in Velocity and Attenuation of Seismic Waves in a Friable Medium Under a Vibrational Dynamic Loading
Instant variations of the velocities and attenuation of seismic waves in a friable medium subjected to dynamic loading have been studied by new experimental techniques using a powerful seismic vibrator. The half-space below the operating vibrator baseplate was scanned by high-frequency elastic waves, and the recorded fluctuations were exposed to a stroboscopic analysis. It was found that the variations of seismic velocities and attenuation are synchronous with the external vibrational load but have phase shift from it. Instant variations of the seismic waves parameters depend on the magnitude and absolute value of deformation, which generally result in decreasing of the elastic-wave velocities. New experimental techniques have a high sensitivity to the dynamic disturbance in the medium and allow one to detect a weak seismic boundaries. The relaxation process after dynamic vibrational loading were investigated and the results of research are presented.
NG41A-0145
Observations of the Vibrational Wave Fields Variation in the Baikal Rift Zone
Vibroseismic monitoring of the seismic active central part of Baikal rift zone is carried out since 2004 with the use of the vibrator CV-100 located on geophysical observatory "Souhoi Rouchei" near the shore of lake Baikal. The vibrator operates one night every month and radiates the harmonic signals with the frequencies of 7.0 Hz, 8.0 Hz and 9.0 Hz and duration of 10 minutes. The control system of vibrator provide the stability of the parameters of radiating signal about 4 percent( or 14 degrees) for phase and 1 percent for amplitude. The synchronization errors of the systems of radiation and recording are not exceed 1 millisecond. The procedure of complex averaging-out with 4096 points window are used for calculation of a phase and amplitude of the monochromatic signal. The records of the seismic stations of the local network are used for the determination of the amplitude- phase characteristics of coherent radiation of the vibrator CV-100. Seismic stations "Khuramsha", "Turuntaevo", "Tirgan", "Ongureni", "Zakamensk" have the highest signal-to-noise ratio for the harmonic signals on South Baikal region and provide the precise determination of the small variations of the signal parameters. The smooth decrease of phases and increase of amplitudes in summer season are allocated for the first time by means of long-term observations. These fluctuations have cyclic (seasonal) nature. It confirms necessity to take into account the annual variations of amplitude-phase characteristics of stationary vibroseismic field when the geodynamic processes are investigated by the methods of vibroseismic interferometry. The structured database of monitoring sessions of 2004-2006 for the seismic interferometry method in Baikal region is presented.
NG41A-0146
Geysers Characteristics before and after Landslide of June 3-rd, 2007 (Geysers Valley, Kamchatka, Russia)
Since 1990 cycling characteristics of five geysers (Maly, Bolshoy, Shel, Velican, Troynoy) were contentiously monitoring using automatic telemetric system (V A Drosnin, http://www.ch0103.emsd.iks.ru/ ). The most powerful geyser Velikan erupted steam clouds at 300 m height. 1:20 UTC June 3-rd, 2007 lower basin of the Geysers Valley was in a few minutes buried under 10 mln m3 of mud, debris, and blocks of rocks. Some indications were found, that landslide triggered by steam eruption in the upstream area of Vodopadny creek. As a result of this three famous geysers (Pervenets, Sakharny,Troynoy) located at lower elevations were sealed under 10-30 m thick caprock as well as Vodopadny hot creek, a rock dumb trap Geysernaya river and lifted water into 20 m deep lake, which flooded three famous geysers (Conus, Bolshoy and Maly) terminating their cycling activity. Nevertheless Bolshoy and Maly activity continues in a form of discharge of water circulated in the former geysers channels and a clear plume at a lake surface above exits observed. Shortly after landslide continuous monitoring of the cycling characteristics of the upper basin geysers, including Velikan and lake level, accomplished by temperature loggers – restarted. There are some indications time periods of the geysers cycling decrease.
NG41A-0147
Time-Evolution of Later Phases in the Transfer Functions Obtained by a Permanent Seismic Source and Receivers (ACROSS)
Repeated transmission of a controlled seismic signal and a long-term continuous observation of the seismic waves enable precise monitoring of time-evolving physical properties of the Earth's interior. An observation using the ACROSS (Accurately-Controlled, Routinely-Operated Signal System) source was conducted to establish a method to monitor the state of coupling of the plate boundary and to understand the time-evolving phenomena such as slow-slips and deep low-frequency tremors (DLFTs). We carried out a long-distance seismic monitoring experiment using ACROSS for 10 months in the Tokai distinct, Japan. Our attempt was to detect reflected phases from the top of the subducting Philippine Sea plate and to detect their time evolution. The results have been presented in the past AGU meetings (Yamaoka et al., 2004, Soma et al., 2005, Watanabe et al., 2005, Ikuta et al., 2006.) A seismic array composed of 12 seismometers was deployed at 55 km away from the source. Stacking and deconvolution of continuous data retrieved transfer functions, which correspond to time-domain waveform, with improved S/N ratio. We applied an array analysis to identify the later phases that are coherent among the array elements. By comparing the synthetic waveform and the coherent phases extracted from the semblance analysis, we concluded that the later phases included the reflected waves from the top of Philippine Sea plate, Moho discontinuity and other layer boundaries. Using borehole seismometer (Hi-net) data that were free from environmental effects, the time-evolutional changes of each phase found in the transfer function were examined. The first arrival showed small variation, on the other hand, the later phases show larger variation. The time evolving changes indicate the existence of time- variant scatterers in deep crust or the effect of multiple scattering. An interesting feature is that the time-variation of the later phases shows a correlation with the episodic activity of the DLFTs in this area. This implies the possibility of seismic monitoring of seismogenic plate-subduction zone using ACROSS.
NG41A-0148
Interpretation of Frequency Dependence in Transfer Functions Acquired by Seismic ACROSS
The Accurately Controlled Routinely Operated Signal System (ACROSS) is an effective tool for active monitoring of subsurface geophysical properties. Seismic ACROSS provides us with continuous record of transfer function between the transmitter and the receiver, which is interpreted as the tensor Green's function sampled at finite discrete frequencies. Hasada et al. (2006 Fall Meeting) reported that there are significant frequency dependence in the transfer function acquired by seismic ACROSS and suggested that the frequency dependence may be caused by multipath arising from heterogeneous structure of propagating medium rather than material dispersion. In order to interpret the mechanisms and locate the structure causing the frequency dependence, we have analyzed the ACROSS transfer functions observed at several seismic stations including borehole seismometers and a surface seismic array. First we draw spectrograms from the observed transfer functions and notice some conspicuous patterns in both of P and S wave packets. Location of the structure causing each spectral pattern can be (1) near the transmitter, (2) on the propagation path, or (3) near the receiver. In the case 1, the spectral pattern should appear in the data observed at all stations. Utilizing the data acquired by seismic array enables us to distinguish the case 2 or 3, though there is difficulty due to limited distribution of receivers. For example, we take notice at the significant spectral peak lies around 16Hz in the P wave observed by the seismic ACROSS transmitter at Toki and the receiver at ~60km southeast of the transmitter. The similar feature is found in the data from another station at ~10km north of the transmitter, indicating that the location of the structure causing 16Hz peak of P wave is expected to be near the transmitter. More complicated frequency dependence exists in coda parts following P and S arrivals, which may contain randomly scattered waves. Array analysis helps us to interpret such features.
NG41A-0149
Temporal changes in emission characteristic of elastic wave by vibrator source and its correction
We are developing an Accurately Controlled Routinely Operated acoustic Signal System (Acoustic-ACROSS) to observe time evolution of crustal structure. Acoustic-ACROSS source continuously radiates accurately controlled elastic signal by rotating eccentric mass. If the emission of the ACROSS signal does not vary with time, we can attribute the time evolutions in the observed signal to the change of propagation properties along the path. But it is pointed that the radiated signal by ACROSS sources changes due to surface conditions such as atmospheric temperature and rainfall. So we should correct these source instabilities to extract the temporal changes of the signal due to the changes of the propagation properties in the deeper part of the earth. We conducted a 15 months long-term monitoring experiment using a couple of ACROSS sources deployed near Nojima fault, Awaji Island. Receiver seismometer was deployed at the bottom of an 800 m deep borehole dug beside the sources. We have proposed an analytical approach to correct the effect of the radiation instability by referring to records of 12 seismometers deployed near the sources (Yamaoka et al. 2001 and Ikuta et al. 2004). We assumed that the records S(t) obtained by the near-source sensors represent signal input. So the wave field Y(t) recorded by the borehole sensor is expressed by a linear combination of S(t): Y(t)=G(t)S(t), in which the combination coefficient G(t) is transfer function between the input and the borehole sensor. This equation can be rewritten in the following form: Y(t)=[G0+dG(t)][S0+dS(t)], in which G0 and S0 denote the time-invariant parts of G(t) and S(t), respectively. We also assumed that the time-variant terms dG(t) and dS(t) were small. Ignoring higher- order fractions, this is rewritten as Y(t)=G0[S0+dS(t)]+dG(t)S0. Subtracting G0dS(t) from the right-hand side makes the observed wave field Y(t) independent from the radiation instability. We obtained G0 by applying least square method to fit Y(t) with S0+dS(t). The residual of this fitting is dG(t)S0. Now the sum of G0S0 and the residual is what we want to know. We corrected observed wave field as Y'(t) =G0S0+dG(t)S0, which does not include the source input variation. However, this method was not effective for a record longer than a few months. This method could not correct the long-term record all together because the G0 was not stable through the term. We adopted another new method for correction. We separated the records into the short-term and the long-term variations by applying a moving average with an interval of 20 days. Next we corrected both variations individually and then added the results again. For the individual corrections, we adopted a new model in which the time- variant part in the records dY(t) is fitted by linear combination of dS(t): dY(t)=G0dS(t)+e. We can obtain G0 by applying least square method to fit dY(t) with dS(t) and the residual e represents dG(t)S0. This is none other than corrected observed variation by the borehole sensor dY'(t). We applied this method to the record in the whole period of the 15 months experiment. We successfully reduced the variation in the borehole records due to radiation instability by this correction method. The variation in the borehole records reduced twice as good as the previous method. After the correction, a seasonal changes in the records disappeared and a variation in amplitude and travel-time of P and S waves corresponding to a distant earthquake remained clearly.
NG41A-0150
Recent Progress in Electromagnetic (EM-) ACROSS in Tokai Region
A brief review is presented on the current state of the ACROSS (Accurately Controlled Routinely Operated Signal System), which has been developed for active monitoring of the physical states within the seismogenic zone for almost 10 years. An essential point of ACROSS is a noise-robust observation system of acquiring the transfer functions of both electromagnetic and seismic waves between a pair of signal source and receiver by means of accurately controlled continuous transmission of sinusoidal signals. The first generation experiments of EM (electromagnetic) ACR0SS are made successfully at Tono Geoscience Center (JAEA), Toki, Gifu prefecture. The signal source was a set of two electric dipoles of 150m-4A for transmission of a vector signal to obtain the tensor transfer function within a specified frequency range up to the distance of 50 km by data stacking. At present, two EM transmitters are being operated; one at Horonobe in Hokkaido with a set of electric dipoles of 180m-10A continuously for more than 2 years, and another at Shizuoka with a larger dipole moment of 600m-20A for long distance transmission. We will report the recent results obtained by Shizuoka transmission experiments to determine the transfer functions and to estimate the delay time of signal propagation up to about 20km distant electric potential and magnetic field observation sites. Further we present a prospect towards the monitoring of effects of water in the anticipated Tokai Earthquake focal region located beneath the Shizuoka prefecture. As a matter of course, we need more powerful transmitters and also an observation network designed specially for our approach.