Hypocentral Parameters Research Articles

Upper crustal structure beneath the Jemez Mountains Volcanic Field, New Mexico, determined by three‐dimensional simultaneous inversion of seismic refraction and earthquake data

The upper crustal structure beneath the Jemez Mountains volcanic field (JVF), New Mexico, was determined using a simultaneous inversion of earthquake and seismic refraction travel time data. One‐, two‐, and three‐dimensional inversions were sequentially applied to derive a detailed velocity model of the upper 10 km of the crust beneath the JVF. Travel time observations from six shot points recorded at as many as 72 temporary seismograph stations and from 27 earthquakes recorded at from 6 to 12 network stations were simultaneously inverted to determine both compressional wave velocity structure and hypocentral parameters. The velocity model was described by a grid of velocity values (unknowns in the inversion procedure) with velocities between grid points given by linear interpolation. Ray tracing through two‐dimensional slices of the three‐dimensional model was performed to calculate theoretical travel times for each source‐receiver pair. A damped least squares iterative inversion was used to derive the velocity model and hypocenters from the differences between observed and theoretical travel times. The relocation of the earthquakes resulted in changes in epicentral location of up to 16 km, although most events were moved less than 7 km. Hypocentral depths of the relocated events ranged from sea level to 17 km, averaging about 8 km. The final three‐dimensional model provided a good fit to the travel time observations. The velocity model displays a prominent low‐velocity, approximately cylindrically shaped body beneath the Valles caldera within the JVF. The velocity anomaly is defined by a lateral decrease in velocity from about 6.0 km/s to as low as 5.6 km/s. The low‐velocity body has a diameter of about 15 km. We interpret the velocity anomaly beneath the Valles caldera to be the result of the combined effects of a silicic intrusion in the upper crust and velocity reduction caused by elevated temperature.

Uniform reduction error analysis

By taking advantage of asymptotic results from the statistical literature, it is possible to develop marginal confidence intervals and joint confidence regions for hypocentral parameters derived from Jeffrey's uniform reduction earthquake location algorithm. The method currently used to compute confidence intervals is ad hoc and can result in substantial errors. Monte Carlo experiments have been used to indicate typical deviations from the asymptotic distribution for finite numbers of observations. Also, the statistical theory can be applied to estimating the efficiency of the uniform reduction algorithm when applied to various models of the travel-time residual probability density function. Further, the nonuniqueness of the uniform reduction algorithm can be understood and eliminated by appealing to well-known results from robust statistical theory. Finally, it is possible to derive a weighting function which is asymptotically efficient for any presumed residual model.

Simultaneous estimation of hypocenters and velocity parameters for an inclined-layer velocity model.

A method of simultaneously estimating hypocenter parameters, station corrections, and velocity parameters has been developed for a velocity model with an arbitrary dipping layer boundary. The seismic ray is traced two-dimensionally by defining an equivalent ray path in a flat-layered velocity model for each combination of source and receiver. The partial derivatives of the travel time with respect to the perturbations in the hypocenter parameters and velocity model parameters are found to be expressed in analytical forms. These advantages are utilized in developing a time-saving computational procedure for the simultaneous estimation. The computer program coded according to the above procedure is tested for an artificial data set. The arrival-time data are stably inverted for true hypocenters and a true velocity model. Thus, it is concluded that the method works well, at least for an ideal data set. The present method may be used in estimating the gross image of the lateral variation in the crust of the velocity structure.

Aftershock distribution of the 1983 Nihonkai-chubu (Japan Sea) earthquake determined from relocated hypocenters.

Aftershocks of the 1983 Nihonkai-chubu (Japan Sea) earthquake (MJMA 7.7) were relocated using an inclined-layer velocity model, a one-layer crust in dipping contact with underlying half-space mantle. We dealt with aftershocks which occurred from June 5 through June 23, 1983, when the seismographic stations of Hirosaki University were most densely distributed in northern Tohoku, Honshu, Japan. The parameters of the velocity model were determined by applying a method of simultaneous estimation of velocity model parameters and hypocenter parameters to an ensemble of P-wave arrival times from selected aftershocks. The determined Moho interface dips nearly eastward with an angle of 6°-8°, being placed at a depth of 18 km at the western end of the aftershock area and 28 km on the Japan Sea coast of Akita and Aomori Prefectures. The relocated hypocenters concentrated intensively on an easterly dipping plane, thus clearly suggesting the geometry of the main-shock fault. The fault mapped by the aftershock distribution changes its strike halfway along the fault length, striking in N15°-20°E in the southern part and N10°-15°W in the northern part. This change in fault strike agrees with a change in the general trend of geology in the aftershock area. The dip of the fault, which had previously been uncertain, was reliably estimated to be dipping in an easterly direction with an angle of about 20° for both the northern and southern parts. The bottom of the fault is probably limited within the crust because the relocated aftershocks are confined to a depth range shallower than 20 km. When compared with hypocenters of aftershocks determined for the inclined-layer velocity model, those determined for laterally homogeneous velocity models were much more scattered. Reduction of the scatter of hypocenters has been accomplished by the introduction of the inclined-layer velocity model which adequately accounts for the lateral variation of velocity structure in and around the aftershock area.

微小地震観測網データの自動処理システム

An automatic event detection and location system has been developed to process microearthquake data collected by the seismic network of Tohoku University, which covers northeastern Honshu, Japan. The system is composed of three subsystems. The first one (TDE) is a pre-processing system whose main function is event detection. Preliminary hypocenter determination is also made in this pre-processing system. Automatic picking of first arrivals and hypocenter location for events detected by TDE are performed by the second subsystem (RPS). Earthquake hypocenters are located by an iterative method. First, P arrivals are picked and are used to locate the event. Then the system picks S arrivals by using the expected arrival times from the hypocenter already determined from P arrivals. Hypocenter location is made again by using both P and S arrivals. Subsequently arrivals which do not agree with this location or are missed in the first picking are repicked and the event is relocated. This process is repeated until reasonable result is obtained. In the location procedure included is an algorithm which can remove wrong data or can discriminate two or more events occurred at the same time. The third subsystem (BPS) is a batch processing system for checking the results obtained by RPS.In June 1984 the system was linked with the telemetering network system of Tohoku University, and since then it has been working in real time. Hypocenter parameters determined from the automatic system are compared to those determined by the manual processing system. It is found that more than 90% of the events with magnitude larger than 2.0 located in and around the network by manual processing are well located by the automatic system. Differences between hypocenter locations obtained by the automatic and manual processing systems are mostly within a standard deviation of determined hypocenters. Hypocenter distribution obtained from the automatic system clearly shows double-planed deep seismic zone beneath the network, which demonstrates hypocenter locations by the automatic system are quite reliable.

Source processes of the 1981 Gulf of Corinth earthquake sequence from body-wave analysis

On 24 February 1981 (20 h 53 m), an offshore earthquake of magnitude M, = 6.7 occurred in the eastern part of the Gulf of Corinth, Greece. So far, it is the largest earthquake that has taken place in the area during this century. The shock caused considerable damage in coastal population centers on both sides of the Gulf. Sixteen people lost their lives and several hundred were injured, mostly by falling objects. The main shock was followed by two large aftershocks of magnitude M, = 6.4 on 2.5 February (02 h 35 m) and 4 March (21 h 58 m). These aftershocks caused additional severe damage in areas to the east of the main shock epicenter and were accompanied by surface faulting. Hypocenter parameters of the three shocks are listed in Table 1. Teleseismic long-period body-waves from the main shock on 24 February and its two principal aftershocks are studied to determine source characteristics. The stereographic projections of P-wave first motions for the three shocks are shown in Fig. 1. It is obvious that most of the available teleseismic observations are dilatational motions, indicating that the focal mechanisms are dominated by a 45” dip-slip

Re-evaluation of the large deep earthquake of January 21, 1906

The large deep earthquake of January 21, 1906 is re-evaluated using old seismogram data and updated analysis techniques. From the P and pP-P time data the hypocentre parameters are determined as follows: origin time, 13h 49min 35s; latitude, 33.8°N; longitude, 137.5°E; depth, 340 km. The body-wave magnitude m B is re-evaluated from the amplitude and periods of P, PP and S waves. The average value of 7.4 is obtained. This value is the smallest among any values assigned previously to this shock, and it is denied that the earthquake is the world's largest deep shock in this century. The focal mechanism is estimated from the P-wave first motions and amplitude distribution of P and S waves. Synthetic body waves are used to constrain the mechanism and to determine the seismic moment. The mechanism solution suggests the down-dip compression typical of this region. A seismic moment of 1.5 × 10 27 dyn · cm is obtained. This value and the re-evaluated value of m B are consistent with the moment- B relation obtained for other deep earthquakes.

Microearthquakes beneath the Median Valley of the Mid‐Atlantic Ridge near 23°N: Hypocenters and focal mechanisms

We report hypocenters and focal mechanisms of microearthquakes located by an ocean bottom seismic network deployed in the median valley of the Mid‐Atlantic Ridge near 23°N during a 3 week period in early 1982. The network consisted of seven ocean‐bottom hydrophones and three three‐component ocean bottom seismometers. The instrument coordinates were acoustically determined to within 25 m at the 1 standard deviation level of confidence. The hypocentral parameters of the 26 largest microearthquakes are reported; 18 of these events have epicenters arid focal depths which are resolvable to within ±1 km formal error at the 95% confidence level. Microearthquakes occur beneath the inner floor of the median valley and have focal depths generally between 5 and 8 km beneath the seafloor. Composite fault plane solutions for two spatially related groups of microearthquakes beneath the inner floor indicate normal faulting along fault planes that dip at angles of 30° or more; these solutions are similar to the mechanisms of nearby large earthquakes. Microearthquakes also occur beneath the steep eastern inner rift mountains. The rift mountain earthquakes have nominal focal depths of 5–7 km and epicenters as distant as 10–15 km from the center of the median valley, but these hypocenters have larger uncertainties because of the possible effects of large topographic relief and associated lateral heterogeneity in velocity structure. We interpret the depth distribution and source mechanisms of these microearthquakes to indicate that this segment of ridge axis is undergoing brittle failure under extension to a depth of at least 7–8 km. We infer that the entire crustal column has been cooled to temperatures within the brittle field of behavior and that significant time has elapsed since the most recent episode of volcanic or shallow magmatic activity along this ridge segment.

Earthquake location inverse problem: a review and an application to the aftershocks of the Southern Italy 23 November 1980 earthquake

In the last few years many improvements have been realized on the application of inverse methods to seismology. Increasing attention has been payed to simulataneous inversion methods, allowing us to determine hypocentral parameters and velocity models using data from a set of earthquakes recorded by dense seismic arrays. Since the first studies by Peters and Crosson, Crosson and Aki and Lee, recently Thurber, Pavlis, Spencer-Gubbins developed new techniques which are more efficient and practicable also with moderate-size computers. In the first part of this paper a review of all these recent inversion methods in presented starting with the simple location algorithms, to come to the most sophisticated simultaneous inversion methods. For shortness the discussion is restricted to the so-called «overdetermined» problem in which the velocity model is parametrized by a finite number of parameters. A brief appendix about the most powerful mathematical inversion techniques is also presented. In the second part, we present a study on the precision and the accuracy of some location algorithms, used to relocate aftershocks by using a temporary seismic array installed soon after the 1980 Southern Italy earthquake. This study remarks the important role of network geometry and the use of accurate algorithms for identifying earthquakes rupture zones.

Travel time inversion for three-dimensional P-wave velocity anisotropy.

Inversion method for estimating the three-dimensional isotropic seismic velocity structure developed by Aki and his colleagues is extended to include the effect of velocity anisotropy. Assuming weakly inhomogeneous anisotropic media with some symmetry in anisotropy, we reduce the number of unknown parameters. We present two sets of linearized equations of quasi P-wave travel time residuals suitable for inversion. One set is for a weakly anisotropic medium with a hexagonal symmetry about a horizontal axis, and the other for a medium with a spheroidal P-wave velocity surface. Using these equations, we invert P-wave travel time residual data to estimate the isotropic and anisotropic velocity perturbations in the subdivided three-dimensional blocks within the modeling space. For local earthquakes, the hypocentral parameter perturbations should also be included in the inversion. This three-dimensional anisotropic velocity inversion is a natural extension of the previous isotropic velocity inversion. The three-dimensional anisotropic inversion method is applied to seismic data in Southwest Japan to determine velocity perturbations in the blocks which were well-determined in the previous isotropic study. The isotropic velocity perturbations are well determined, but the anisotropic velocity perturbations are less resolved. In spite of the rather poor resolution, our preliminary analysis shows some interesting features. Within the descending Pacific and Philippine Sea plates, the fast directions of P-wave velocity seem to be perpendicular to the magnetic lineations in the sea-floors near the trenches, suggesting that the descending lithosphere maintains the anisotropic structures which were formed before subducting.

VELOCITY STRUCTURE BENEATH THE HIDAKA MOUNTAINS IN HOKKAIDO, JAPAN

One- and three-dimensional velocity structures down to 80 km beneath the Hidaka Mountains in Southern Hokkaido, Japan, were derived by a simultaneous inversion of the P-arrival time data from local earthquakes. Arrival time data for the derivation were obtained from the observation of earthquakes recorded at 27 seismic stations. In order to obtain an appropriate starting P-wave velocity structure and hypocentral parameters to be used in three-dimensional inversions, one-dimensional inversions were first carried out. The three-dimensional inversions were made with a simplified algorithm for the calculation of travel times and their partial derivatives with respect to the hypocentral parameters.One of the most significant features of the three-dimensional velocity structure obtained is the existence of an inclined low velocity zone. This zone has P-wave velocities that are as much as 10% slower than the velocities in the surrounding regions. This low velocity zone is 20-25 km thick. It is located beneath the area from the western to the central portion of the Hidaka Mountains, extends in depth from 10 to 65 km, and possesses a large dip angle of 60°. A remarkably high velocity zone is also found to stretch in a strikingly northwest-southeast direction just beneath the crest of the Hidaka Mountains. A portion of this high velocity zone is found to extend to the south-west so as to overlie the low velocity zone. The thickness of the high velocity zone is estimated to be about 10 km.Comparison of the obtained three-dimensional velocity structure and the seismicity shows that the low velocity zone is seismicly active, while the high velocity zone is relatively inactive.The small value of P-wave velocity in the inclined low velocity zone, a value which was estimated to be only 7 km/s at a depth of 60 km, allows us to infer that the material in the low velocity zone originates from the crust. This inference leads to the possibility that the inclined low velocity zone is subducted crust resulting from collision between the Kurile and Northeastern Japan arcs. The three-dimensional velocity structure also suggests the possibility that the descending Pacific plate has been deformed so as to be confined to the region below the inclined low velocity zone.

Bayesian estimation of hypocenter with origin time eliminated.

In this study a new approach to hypocenter location is formulated from a Bayesian point of view. Observational data for hypocenter location are the arrival times of phases at seismic stations. Unknown parameters are the spatial coordinates and origin time of a quake focus. In general, we have some prior information about the spatial coordinates, but not about the origin time. Then, according to Bayes' theorem, the probability density function (pdf) of hypocenter parameters posterior to observed data is proportional to a product of the likelihood and the prior pdf of the spatial coordinates. Integrating the posterior pdf over the whole range of origin time, we can eliminate the origin time from the location problem, and obtain the marginal pdf of only the spatial coordinates. The best estimate of a hypocenter is given by a set of the spatial coordinates which maximizes the marginal pdf. Supposing Gaussian errors in both observational and prior data, we obtain a simple and robust algorithm to invert arrival time data for a hypocenter. Estimation errors of parameters are evaluated by an asymptotic covariance matrix, which gives a good approximation to exact covariance when the estimate is linearly close to a true hypocenter. The advantages of this approach are as follows : i) the use of prior data resolves unknown parameters completely within finite errors, and ii) the elimination of origin time suppresses the instability of estimates.

Use of two inverse techniques. Application to a local structure in the New Hebrides island arc

Summary. Two inversion techniques (stochastic inverse and linear programming) are used in the determination of hypocentre and velocity model parameters simultaneously by using P-wave arrivals from local earthquakes at a local seismograph array. These procedures are applied to P-wave data obtained from a combined land and ocean bottom seismographs network in the New Hebrides island arc. Two hundred and fifty-two residuals data are inverted simultaneously with 46 earthquake parameters. The displacements of each initial hypocentre location are practically the same with the two methods and are small (2–4km). The main result for the velocity model parameters is the existence at a low-velocity zone between depths of 10 and 19km and may be deeper. An estimate of the degree of confidence of the solution is determined and can confirm that the extremal values of the 7.3 km s-1 are 7.2 and 7.5 km s-1. These results may be related to an intrusion of ultrabasic magma from the mantle beneath Malekula and Efate Islands.

The refinement of hypocenter parameters on the basis of three-dimensional velocity models of focal zones

A complex approach to the processing of seismological data to determine hypocenter coordinates and to construct three-dimensional velocity fields of focal zones is considered. The iterative process includes the determination of focal coordinates under the assumption of a laterally homogeneous medium, the construction of a three-dimensional velocity model, and subsequent redefinition of the focal coordinates taking into account the inhomogeneous structure of the medium. The principal results of calculations of three-dimensional velocity fields and refined earthquake hypocenters are presented for focal zones in Vrancea (Carpathians), the Caucasus, and Kamchatka.

Simultaneous determination of P and S velocities of the crust and the sub-Moho mantle with earthquake hypocenters in central Honshu,Japan.

P and S arrival times of crustal and subcrustal earthquakes, obtained from three-component wide-band seismographs of the 10-station telemetered network of Nagoya University, were inverted by a damped least squares method to obtain the P and S velocity structure of the and the in the Chubu district, Honshu, Japan. The hypocentral parameters and the P and S station anomalies were determined simultaneously. The solutions showed satisfactory stability and resolution. The and the were divided into four layers, based on results of refraction studies. The uppermost of 4km thickness is the sedimentary layer which was not incorporated into the inversion. The second with a thickness of about 20km is the crust for which P and S velocities of 6.1 and 3.6km/sec were obtained with a VP/VS ratio of 1.69. This P velocity is consistent with the results of refraction studies. When this was divided into two parts at a depth of 12km, the sums of squared residuals were reduced dramatically, suggesting a subdivision of the upper curst at this depth. The upper portion possesses P and S velocities of 6.0 and 3.55km/sec and the portion, of 6.3 and 3.7km/sec, respectively. Both portions are characterized by a VP/VS ratio of 1.69. The third of about 10km thickness includes the lower crust there in, for which P and S velocities of 6.9-7.0 and 3.8-3.95km/sec were obtained. The corresponding VP/VS ratio is 1.78-1.81, in marked contrast to the upper crustal VP/VS ratio. The fourth is the sub-Moho mantle for which we obtained P and S velocities of 7.7-7.8 and 4.4km/sec that give a VP/VS ratio of 1.76-1.77, a value similar to that for the crust. The inverted P velocity is in agreement with the results of refraction studies. The Moho boundary probably lies at a depth of around 33km. The boundary between the upper and is not well constrained by the data. The consistency of our P velocity structure with those derived from explosion seismic studies indicates the feasibility of the present method for obtaining an S velocity structure. Poisson's ratio changes significantly across the boundary between the upper and but not across the Moho.

Some seismological features from an analysis of the recent seismic activity in the Messina Strait

abstract An analysis of the recent seismic activity at the Strait of Messina during the last 30 yr, made on the basis of collected observation data, is reported. For about 2 yr this activity has been better observed since a short-period seismic network [Messina Strait Network (MSN)] has been installed on both sides of the strait. The use of a compatible crustal model and of the HYPOSTRE program, elaborated to determine the hypocentral parameters of local events, gives results which can be correlated with the structural characteristics of the graben of the strait. Besides, some elements arising from a previous seismological analysis, made in the same area before the MSN had been activated, are confirmed. In particular, the existence of a microearthquake source, having its epicenter situated a little south of Messina, is here reproposed. Furthermore, the increase in the number of stations on both sides of the reveals that the absorption of the seismic energy propagating parallel to the graben structures is different from that observed when the energy travels orthogonally to them. This fact, together with the results obtained using HYPOSTRE for calculating the focal depth of some local events, suggest the existence of an “energy trap” affecting the more superficial crustal layers. This is correlated with the existence of the dynamic discontinuity of the strait.

Three‐dimensional velocity anomalies in the upper crust near Socorro, New Mexico

262 P‐wave arrival times associated with 40 local events obtained from a movable array of up to eight short‐period seismographs were used to determine hypocentral parameters and an accurate three‐dimensional seismic velocity model for the Rio Grande rift zone near Socorro, New Mexico. The initial attempt to model these data resulted in a representative half‐space velocity of 5.85 ± 0.02 (1 s.d.) km/sec. No azimuthal velocity variation was found that was significantly different from the half‐space solution. The area was then subdivided at depth (4km) and into blocks one‐tenth of a degree on a side. The resulting model showed that a block in the lower layer approximately 15 km west of Socorro had an average velocity of 5.17 ± 0.11 (1 s.d.) km/sec (0.68 km/sec less than the half‐space velocity). Other blocks had smaller, but still significant, anomalies relative to the half‐space velocity. The preferred explanation for the anomalously low velocity is that this area represents a site of magmatic intrusion into the upper crust (4‐10 km deep).

Determination of earthquake source parameters from waveform data for studies of global and regional seismicity

It is possible to use the waveform data not only to derive the source mechanism of an earthquake but also to establish the hypocentral coordinates of the ‘best point source’ (the centroid of the stress glut density) at a given frequency. Thus two classical problems of seismology are combined into a single procedure. Given an estimate of the origin time, epicentral coordinates and depth, an initial moment tensor is derived using one of the variations of the method described in detail by Gilbert and Dziewonski (1975). This set of parameters represents the starting values for an iterative procedure in which perturbations to the elements of the moment tensor are found simultaneously with changes in the hypocentral parameters. In general, the method is stable, and convergence rapid. Although the approach is a general one, we present it here in the context of the analysis of long‐period body wave data recorded by the instruments of the SRO and ASRO digital network. It appears that the upper magnitude limit of earthquakes that can be processed using this particular approach is between 7.5 and 8.0; the lower limit is, at this time, approximately 5.5, but it could be extended by broadening the passband of the analysis to include energy with periods shorter that 45 s. As there are hundreds of earthquakes each year with magnitudes exceeding 5.5, the seismic source mechanism can now be studied in detail not only for major events but also, for example, for aftershock series. We have investigated the foreshock and several aftershocks of the Sumba earthquake of August 19, 1977; the results show temporal variation of the stress regime in the fault area of the main shock. An area some 150 km to the northwest of the epicenter of the main event became seismically active 49 days later. The sense of the strike‐slip mechanism of these events is consistent with the relaxation of the compressive stress in the plate north of the Java trench. Another geophysically interesting result of our analysis is that for 5 out of 11 earthquakes of intermediate and great depth the intermediate principal value of the moment tensor is significant, while for the remaining 6 it is essentially zero, which means that their mechanisms are consistent with a simple double‐couple representation. There is clear distinction between these two groups of earthquakes.

Microearthquake survey in the hot spring area in Parbati Valley (Himachal Pradesh, India)

A number of hot springs occur in the Parbati Valley in Himachal Pradesh in India. Temperatures range from 21 to 96°C, the boiling point of water at that altitude. Geological conditions, temperature variations and chemical composition of spring water in the Parbati Valley hydrogeological unit indicate that the deep thermal fluids are of meteoric origin. The maximum temperature acquired by water during its circulation is estimated to come close to 200°C. In order to assess the possibility of extracting geothermal energy, a seismic survey was arranged to locate the hypocentres of microearthquakes associated with the thermal source. A total of eight microearthquake units was set up at an interstation spacing of about 10 km and two months recording were obtained. During this period an average of 2–3 events per day was recorded with S—P interval less than 5 seconds. The data have been analysed with the help of Hypo 71, a Fortran IV computer program designed to determine the hypocentral parameters of earthquakes from seismic data. The results indicate faulting but there is no apparent spatial relationship to surface manifestations of geothermal energy.

Joint analysis of seismological data by the USSO and DSS stations in the Caucasus region

This paper deals with a procedure of a joint analysis of seismic data from earthquakes and those obtained by DSS. The DSS data are used as a first approximation to construct a two-dimensional model of the medium made up of individual blocks. These models serve as a basis when constructing specific three-dimensional travel-time curves. These travel-time curves are further used for the calculation of hypocenter parameters in a laterally inhomogeneous block medium.