Three-dimensional Seismic Structure Research Articles

A Method for Determination of the Three-Dimensional Seismic Velocity Structure from Local Earthquake Data.

We have developed an inverse method to estimate a three-dimensional seismic wave velocity structure using seismic arrival time data obtained from local earthquakes. The method is characterized by a simultaneous determination of a two-dimensional depth distribution of a velocity boundary, a three-dimensional velocity distribution, station corrections and hypocenters. The depth distributions of velocity boundaries and the distributions of the slowness perturbations for P and S waves are modeled by power series of latitude, longitude, and depth, This representation of the three-dimensional seismic structure is advantageous not only in reducing the number of unknown parameters, but also in analytically evaluating the boundary depths, velocities, and partial derivatives of travel times with respect to the unknown parameters. Ray tra-jectories connected between hypocenters and stations are determined by the so-called shooting method for ray tracing scheme under a simplified velocity distribution with the exact boundary depth distribution. Numerical experiments have proved that our inverse method can successfully estimate the three-dimensional velocity structure.

Simultaneous Inversion For 3-D Crustal Structure and Hypocentres Including Direct, Refracted and Reflected Phases-I. Development, Validation and Optimal Regularization of the Method

SUMMARY A new technique for the simultaneous inversion for three-dimensional (3-D) seismic structure and hypocentres (SSH) is developed and validated. The seismic forward problem is solved with a new non-linear 3-D ray tracing procedure that includes direct (as first or later arrivals), refracted and reflected phases from various crustal discontinuities. The non-linear inversion problem is formulated in terms of a modified, iterative Levenberg-Marquardt (LM) technique. It is demonstrated that this method, in spite of its computational inexpediencies, has various properties that make it best suited for the solution of both the general non-linear inverse problem (near-quadratic convergence, due to the approximate restoration of the full Hessian of the objective function) and its linearized subproblems [where LM is equivalent to classical damped least squares (ridge regression)]. Therefore LM is able to play a dual role for the optimal regularization (OR) of both the non-linear and the linear ill-posed SSH problem. The SSH method is validated through simulated inversions of local traveltimes from the Rhine Graben region. The results demonstrate that the usual trade-off between focal depths and seismic velocities can be reduced strongly through the combined use of all crustal phases. A new non-linear statistical F-test shows that the Newton-like LM method gives a quasi-linear ‘appearance’ of the SSH problem: i.e. convergence is essentially obtained already in the first iteration. This is further confirmed from the results of the OR performed within both the non-linear and the linear model space employing the regularization techniques (RT’s) of Tikhonov, the ridge regression, the stochastic inverse, the method of Backus & Gilbert, and a new method (Backus subjective). Some of these RTs rely on various forms of a priori model information so that they can, in principal, be derived from a general Bayesian formulation. However, the results of the OR illustrate the limitations of this approach, which is attributed to difficulties in specifying the appropriate covariance matrices in model and data space and discrepancies between the theoretical optimal model and its numerical realization. Therefore, those ORTs have been found the most useful, that rely on some form of graphical trade-off curves to systematically explore the model and data space. Based on their theoretical foundations and practical performances, the ORTs of Backus (subjective) and of Tikhonov are being chosen as the reference for the other RTs. Using the ORT of Backus, the minimum of the true error (MSE) is computed, after an a priori upper bound on the model has been selected. This allows one to estimate the statistical bias of the model which is proved to be formally equivalent to the resolution error, as epitomized in the Backus & Gilbert formalism. The Tikhonov ORT, which is used in its original form (the objective one) and as a new version (the subjective approach), allows to constrain best the optimal solution from both the model- and data viewpoints. All of the other Bayesian ORT’s and that of Backus & Gilbert have not been found satisfactory, because they are not able to delimit the optimal model space appropriately.

Three-dimensional seismic attenuation structure beneath the Japanese islands and its tectonic and thermal implications

A three-dimensional (3-D) seismic attenuation structure beneath the Japanese Islands is determined by inversion of seismic intensity data. The result clearly reveals a remarkable contrast in the attenuation structure and reflects the lateral variation of thermal structure beneath the Japanese Islands. For the inversion, the method developed by Hashida and Shimazaki ∗ ∗ (1984). is used: the intensity which is assumed to be a measure of the maximum acceleration of the S-wave at the observation point is inverted to estimate the attenuation structure and the acceleration at the seismic source. The data consist of about 15,000 intensity readings for about 800 earthquakes reported by the Japan Meteorological Agency between 1951 and 1983. By combining 3-D structures of eight districts in Japan, 3-D attenuation maps consisting of three layers are constructed to a depth of 90 km. The obtained structure reveals the following features: first, a distinct difference exists in the crustal structure between NE and SW Japan. In NE Japan, low-Q (high-attenuation) areas on the continental side of the arc are dominant whereas some high-Q (low-attenuation) areas exist on the Pacific side of the arc. In SW Japan, except for the Kyushu district, high-Q areas are dominant and small low-Q areas exist mainly on the Pacific coast. These crustal features generally continue down to the uppermost mantle. Second, the obtained crustal structure shows that attenuation is age dependent. low-Q areas correspond to the distribution of Quaternary volcanic materials and high-Q areas correspond to the geological provinces where pre-Cenozoic rocks are distributed. Since the thermal structure strongly depends on the crustal age, this feature suggests that the attenuation structure reflects the thermal structure. Third, in the upper mantle, low-Q loci are found, which correspond to the distribution of active and other Quaternary volcanoes forming the arc-volcanic chain. It should be noted that even on the continental side of the volcanic front these loci do not exist beneath areas where no volcano exists. This suggests that the upper mantle attenuation is controlled by the existence of upwelling diapirs, which are considered to lie beneath a volcano or a group of volcanoes. Fourth, high-Q zones in the upper mantle exist on the Pacific side of the volcanic front along the Kurile-northeast Honshu arc and along Ryukyu arc. The high-Q mantle lying above the sinking slab seems to reflect a cooling effect of the subducting cold oceanic plates. The thickness of the lithosphere may reach 60 km or more in such outer arcs, although the thickness has been considered to be about 30 km or so. In SW Japan, where the high-Q uppermost mantle also lies, a thick lithosphere of about 60 km may exist. Possible thermal cross sections are proposed for NE and SW Japan, based on the obtained 3-D attenuation structure and other geophysical considerations. In the uppermost mantle of NE Japan a higher temperature is expected than in SW Japan. The thermal structures reflect the difference of tectonic settings in the two regions, such as the subducting plate, overriding lithosphere and volcanic activity.

Three-dimensional seismic attenuation structure beneath the Aegean region and its tectonic implication

The three-dimensional attenuation structure beneath the Aegean Sea and the surrounding regions is determined by inversion of seismic intensity data. A large number of seismic intensity data have been accumulated in a uniform scale in the Aegean region, where the seismic activity is much higher than that of the other Mediterranean regions. Nearly 7000 seismic intensity data from 50 earthquakes that have occurred in these regions are used to determine seismic attenuation structure and source acceleration of the earthquakes. The resultant structure reveals a remarkable contrast of attenuation. In the top layer (depth 0–40 km), low Q is dominant in the Aegean Sea, while high Q is dominant in the surrounding land areas, except for the southwestern Turkey. The low- Q regions correspond to areas of Neogene-Quaternary grabens where the high seismicity of shallow earthquakes appears. In the upper mantle, high- Q zones corresponding to the subducting African Plate dominate along the Hellenic arc. The irregular shape of high- Q zones might reflect the split or disintegration of the African slab. Some low- Q spots corresponding to the distribution of volcanoes exist along the volcanic arc. The low- Q spots might correspond to diapirs causing the subduction volcanism.

Determination of seismic attenuation structure and source strength by inversion of seismic intensity data: Tohoku district, northeastern Japan arc.

As the first application of the method of Hashida and Shimazaki (J. Phys. Earth, 32, 299-316, 1984), a three-dimensional seismic attenuation structure and source strengths are estimated by inversion of seismic intensity data for earthquakes that occurred in the Tohoku district, Japan. We carefully selected 1, 630 intensity data from 101 earthquakes so that the intensities are consistent with accelerations converted from Kawasumi's relation and so that a well-resolved attenuation structure is obtained. A reasonable fit of the formula proposed in our preceding paper to the actual intensity data guarantees that a systematically biased attenuation structure is not obtained. A comparison of the obtained attenuation structure with velocity structures estimated by previous studies shows that high (low) Q nearly corresponds to high (low) V. The correlation of both structures indicates that the attenuation structure estimated by the proposed method is reliable. The resultant attenuation structure shows a remarkable contrast in the attenuation coefficient and two prominent features. The first feature is low-Q zones down to a depth of 90 km, which corresponds to the distribution of volcanoes. The second is high-Q zones that correspond to the subducting Pacific slab. The high-Q slab is in contact with the high-Q zone in a depth range of 30-60 km, which lies on the east side of the volcanic front. The presence of high-stress earthquakes in this depth range, such as the 1978 Miyagi-ken-oki earthquake, is explained by a model in which the contact of the underthrusting Pacific plate with the surface high-Q zone accumulates higher stress and thus causes stronger seismic coupling. The estimated source strength, which is expressed as a point source acceleration, correlates well with earthquake magnitude. Normalized source acceleration, which is an average acceleration over a source area, is estimated. The acceleration suggests that the stress drop of an earthquake becomes higher with magnitude and with depth. The relation between JMA magnitude (MJ) and seismic intensity at a hypocentral distance of 100 km (I100) is found to be I100=1.5 MJ-6.5 for crustal events. This I100 is in agreement with the value reported by Utsu (Bull. Earthq. Res. Inst., 59, 219-233, 1984) which was determined from events excluding those which show anomalous distributions of intensity data. This agreement suggests that our method of estimating earthquake magnitude from intensity data is effective for removing the effect of structure.

Determination of the three-dimensional seismic structure of the crust and upper mantle in the Central Midlands of England

Summary. Teleseismic P-wave residuals relative to CWF, a permanent shortperiod seismic station on Charnwood Forest in the Central Midlands of England, have been determined for two small aperture arrays deployed over the Precambrian block of Charnwood and its surrounding Phanerozoic sediments. The data have been inverted to produce a block model of the P-wave velocity variations in the crust and upper mantle beneath the study region. The results are consistent with significant variations penetrating to a depth of at least 50 km. Low velocities are associated with two upper crustal intrusive bodies, the Caledonian Mountsorrel granodiorite and the South Leicestershire diorites. A longer-wavelength variation at lower crustal/upper mantle depths could arise from the Moho dipping to the south-west beneath the study region, and whose strike sub-parallels the dominant Charnian trend of the major basement structures in this part of Central England.

Three-dimensional seismic transmission prospecting of the Mont Dore volcano, France

Summary. The three-dimensional seismic structure of the Mont Dore volcano is studied by inversion of the arrival times of seismic waves. With this aim two new methods are developed. First, the arrival times are those of Moho-reflected waves at a critical distance from artificial sources in different azimuths. Secondly, the inversion uses a technique which does not require the traditional a priori partition of the space into blocks. The resulting picture reveals such features as: (1) a circular caldera within the basement, the rim of which is marked by magnetic anomalies associated with post-caldera activity; (2) a clear lower limit of the volcano-sedimentary sequence under part of the caldera, opposed to low velocity anomalies extending deeper beneath another part and which may have been the site of volcanic material transport, and (3) eeper heterogeneities possibly related to foundered basement blocks.

Three-dimensional seismic structure of the crust and upper mantle beneath the orogenic belts in southern Hokkaido,Japan.

Three-dimensional seismic structure of the crust and upper mantle beneath the orogenic belts in southern Hokkaido,Japan.

Three-dimensional seismic structure beneath southwest Japan: The subducting Philippine Sea Plate

A three-dimensional seismic structure beneath southwest Japan has been investigated in detail by applying an inversion method. This analysis clearly reveals the present, and possibly the past, subduction of the Philippine Sea plate beneath southwest Japan. Beneath the Izu Peninsula, the Philippine Sea plate does not appear to subduct but collides against the continental Asian plate here, and its thickness appears thinnest, at most about 30 km. From the Tokai to the Shikoku regions, the high-velocity zone corresponds well to the zone of subcrustal earthquakes down to a depth of about 60–70 km, indicating that the subduction of the present Philippine Sea plate does not reach below this depth here. However, a small high-velocity portion appears at depths between 100 km and 150 km beneath a part of the Kinki region and probably extends to the Chugoku region, where no corresponding seismic activities exist. These high velocities might be an indication of the remnants of the past subduction of the Philippine Sea plate. In the Kyushu region, the high-velocity zone continues down to a depth of about 150 km or even to 200 km in the southern part, corresponding to a steeply inclined seismic zone. The Philippine Sea plate in this region appears to be thicker than in regions east of the Kyushu—Palau ridge. It was also found that there exist narrow low velocity zones intervening within the Philippine Sea plate, where two aseismic ridges, the Kyushu—Palau and Shichito—Iwojima ridges, extend northwards and the Nankai trough shows warping. These low-velocity zones give support to the “buoyant subduction” hypothesis. It also appears that lateral heterogeneities existing within and around the subducting plate are closely related to the mode of rupture of large earthquakes occurring around there.

Three-dimensional seismic structure of subducting lithospheric plates under the Japan islands

The three-dimensional seismic structure of subducting lithospheric plates under the Japan Islands has been investigated in detail by using an inversion method developed by Aki and co-workers. The present analysis clearly reveals the subduction of the Pacific plate, with a gradual narrowing in breadth down to about 600 km beneath the Sea of Japan, and also gives some indication of the Philippine Sea plate in the uppermost mantle, less than 50 km under the southernmost part of Japan. The upper boundary of the descending Pacific plate shows sharp velocity contrasts with respect to the overlying low-velocity zone, while the lower boundary appears to have a transitional nature. Large positive Bouguer anomalies in northeastern Japan may be explained, in part, by the effects of the subducting Pacific plate, but those over the Sea of Japan cannot be accounted for by lateral heterogeneities in the upper mantle.

Three-dimensional seismic structure of Usu volcano Part2

Three-dimensional seismic structure of Usu volcano Part2

Three-dimensional seismic velocity anomalies and their relation to local seismicity

This is a brief review of recent results on three-dimensional seismic velocity structures, determined by the method of Aki, Christoffersson and Husebye, using the teleseismic and local earthquake P-time data. Particular attention was given to the relation between the velocity structure and local seismicity. We found that seismic activity may be associated with either low- or high-velocity anomalies, presumably depending on whether the structural weakness, which invites stress concentration, or the ductility, which cannot sustain the tectonic stress, is causing the velocity low.

Determination of the three-dimensional seismic structure of the lithosphere

A new three-dimensional earth modeling is proposed as a framework to obtain more detailed and accurate information about the earth's interior. We start with a layered medium of classic seismology but divide each layer into many blocks and assign a parameter to each block which describes the velocity fluctuation from the average for the layer. Our data are the teleseismic P travel time residuals observed at an array of seismographs distributed on the surface above the earth's volume we are modeling. By isolating various sources of errors and biases we arrive at a system of equations to determine the model parameters. The solution was obtained by the use of generalized inverse and stochastic inverse methods. Our method also gives a lower limit of the true rms slowness fluctuation in the earth under the assumption of ray theory. Using P wave residual data from the Norwegian Seismic Array (Norsar), we have obtained the map of velocity anomalies at various depths up to a depth of 126 km. The rms slowness fluctuation was found to be at least 3.1%. This is in agreement with estimates obtained from statistical analysis of P time fluctuations based on the Chernov theory. The three-dimensional velocity anomalies are presented both by the generalized inverse and by the stochastic inverse solutions. We prefer the dual presentation because it gives the reader greater freedom in judging the results than a single ‘optimal’ solution. Both methods gave essentially the same results. The discrepancies, when they existed, were always explainable in terms of differences in the smoothing procedure which is explicitly given in the resolution matrix. The dominant features in the obtained three-dimensional velocity image of the lithosphere beneath the Norsar array are low velocities to the west and high velocities to the east. The latter feature may be attributed to rocks of the Baltic shield which are undisturbed by the Caledonian orogeny or by Permian volcanism. Our result conclusively demonstrates the existence of strong small-scale inhomogeneities to the bottom of the lithosphere. More theoretical and experimental works are needed to relate these velocity anomalies with the magma ascent mechanism which caused the Oslo graben.

A large-scale three-dimensional seismic structure under the Japan Islands and the sea of Japan.

A large-scale three-dimensional seismic velocity structure down to 650 km under the Japan Islands and the Sea of Japan has been investigated to clarify the subduction of lithospheric plates by applying the inversion method of Aki and others. About 3500 data of P-wave travel time residuals observed at 330 Japanese and foreign stations from 20 intermediate and deep-focus earthquakes around Japan were used to determine the velocity anomalies in a number of blocks and source corrections simultaneously. This analysis has well revealed lateral heterogeneities in the upper mantle above 450km. A high velocity zone corresponding to the descending Pacific plate has been clearly identified, particularly in the north Honshu arc, indicating good correlations with the horizontal and vertical distributions of a number of earthquakes around there. It was found that intermediate and deep-focus earthquakes take place mainly in the upper portions of the descending plate with a thickness of about 100km. The highest velocity anomalies are +5-6%(50-250km), +3%(350km), +2%(450km) within the plate, and about -2% down to 350km over a region on the continental side of the descending plate.

Three-dimensional seismic structure of the lithosphere under Montana Lasa

abstract Using P-wave residuals for teleseismic events observed at the Montana Large Aperture Seismic Array (LASA), we have determined the three-dimensional seismic structure of the lithosphere under the array to a depth of 140 km. The root-mean-square velocity fluctuation was found to be at least 3.2 per cent which may be compared to estimate of ca. 2 per cent based on the Chernov random medium theory. The solutions are given by both the generalized inverse and stochastic inverse methods in order to demonstrate the relative merit of different inversion techniques. The most conspicuous feature of the lithosphere under LASA is a low-velocity anomaly in the central and northeast part of the array siting area with the N60°E trend and persisting from the upper crust to depths greater than 100 km. We interpret this low-velocity anomaly as a zone of weakness caused by faulting and shearing associated with the building of the Rocky Mountains.