Search

To understand steeply dipping events in seismic reflection interferometry (SRI), we derived an expression that describes the difference in travel time (Δτ) from a diffractor to two receivers in two dimensions. For a fixed receiver interval, the expression shows that Δτ is zero when the diffractor is at the midpoint of the paired receivers, increases with an apparent velocity of half the medium velocity as the diffractor moves toward either receiver, and remains constant for a diffractor located on the same side of both receivers. The horizontal portion of Δτ is slightly skewed during the normal moveout correction, yielding a maximum peak of the horizontally stacked trace at a slightly smaller time than Δτ. Accordingly, the diffracted waves have an apparent velocity slightly higher than half of the medium velocity in a horizontally stacked image. This conformed to virtual data for an elastic two-layer model with a vertical boundary. We then generalized the expression to three dimensions, in which listric travel time curves were predicted for an oblique edge diffractor, a vertex diffractor offline from the receiver pair, or a buried diffractor. Based on both two- and three-dimensional analyses of the edge diffractor, we tentatively interpreted the linear and listric dipping events observed in the passive SRI image across the Korean Peninsula to have been caused by diffractors near the intersection of the profile and geologic boundaries.

Seismic data are increasingly used to monitor subsurface velocity changes associated with tectonic and environmental processes that occur in different depth sections. To clarify the differences between effects associated with shallow and deep changes of properties, we conduct numerical experiments using simple layered models that include low velocities, low attenuation coefficients and stress-sensitivity of cracked rocks in the shallow crust. We find significant phase-velocity drops in the period range of 5–20 s when large structural changes occur in the top 1–3 km. The apparent velocity changes (δv/v) measured from the first part of the synthetic Rayleigh waves with a cross-correlation based technique show significant velocity drops in the period bands of 5–10 s and 10–20 s that are consistent with reported values of changes at seismogenic depth. The results highlight the importance of accounting for low velocities, attenuation coefficients and stress-sensitivity of parameters in the top 1–3 km in studies aiming to determine the source region of temporal changes. Analyses using different frequency ranges and calculations of apparent delay times over multiple period bands are essential for resolving the depth range of temporal changes of properties. For temporal changes occurring at seismogenic depths, the measured δv/v values at 5–10 s are significantly larger than those of 10–20 s, which are not observed for shallow changes.

[1] Forecasting the location of an eruption is of primary importance for risk management in volcanic regions. Locating the underground structural changes associated with a potential eruption is also a key issue to better understand the dynamics at work in a volcano. Using recent results in wave physics, we develop an imaging procedure that is based on the sensitivity of multiply scattered waves to weak changes in heterogeneous media. This procedure allows to locate changes in both mechanical and scattering properties of the studied medium. We study ambient seismic noise from 19 broadband stations at the active volcano Piton de la Fournaise on Reunion Island, recorded from June to December 2010. During this period, two volcanic eruptions occurred at two different locations. We calculate the noise cross correlations and study two types of changes in the coda: apparent velocity variations related to changes in the elastic properties of the medium; and, waveform decoherence associated with variations in the scattering, and thus the geological structures. We observe that the temporal variations of both of these parameters provide potential precursors of volcanic eruptions at Piton de la Fournaise. The locations determined from the preeruptive and coeruptive changes in both parameters are in good agreement with the actual eruptive activities. These data demonstrate that the coda of ambient noise correlations contains deterministic information on the locations of the eruptive processes in an active volcano. Our analysis offers an original and significant constraint for the localization of forthcoming volcanic eruptions.

We present a method of separation of P- and S-wave components from three-component (3C) crosswell seismic data. Based on differences between P and S waves in polarization and apparent velocity, this method separates P and S waves in a common source gather without estimating emergence angles and by means of data processing techniques including band-pass filter for preprocessing, high-resolution τ–p transform for removing tube waves and separating up-going and down-going waves, and covariance polarization analysis for correcting the direction of geophone and separating polarization. To separate P- and S-wave data, this method requires only the input of original 3C crosswell data and the first arrivals of P waves. We verify and demonstrate effectiveness and practicability of this method using 2C modeling tests as well as real 3C crosswell data.

シールドトンネルなど地下構造物の縦断方向の耐震検討では，構造物に沿った地震波のみかけの伝播速度の評価が重要である．既往の研究では表面波の位相速度を利用することで安全側の評価を行う方法などが提案されているが，地盤の軸ひずみに関して実状に即した評価を行うことを目的とする場合に，みかけの伝播速度としてどの程度の値を用いるべきであるかについては，これまで十分な検討が行われてこなかった．そこで，本研究では，M5.3～M8.0の13の地震に対するシールドトンネルにおける地震観測結果，および，セグメントリング間の相対変位の計測結果に基づき，地震波の見かけの伝播速度に関する検討を行い，見かけ入射角依存性など，見かけの伝播速度に関するいくつかの特徴を明らかにした．

We use relative arrival times and locations for similar earthquake pairs that are found using a cross-correlation method to analyze the time dependence of P and S station terms in southern California from 1984 to 2002. We examine 494 similar event clusters recorded by Southern California Seismic Network (SCSN) stations and compute absolute arrival-time variations from the differential arrival-time resi- duals obtained following event relocation. We compute station terms from the robust means of the absolute arrival-time residuals from all events recorded by each station at 3-month intervals. We observe nine stations with abrupt offsets in timing of 20- 70 msec, which are likely caused by equipment changes during our study period. Taking these changes into account could improve the relative location accuracy for some of the event clusters. For other stations, we generally do not see systematic temporal variations greater than about 10 msec. Analysis of residuals along individual ray paths does not reveal any clear localized regions of apparent velocity changes at depth. These results limit large-scale, long-lasting temporal variations in P and S ve- locities across southern California during this time period to less than about � 0:2%. However, there is an increased fraction of individual travel-time residuals exceeding 20 msec immediately following major earthquakes from source regions near the main- shock rupture.

Using a short-baseline seismic array (U.S. Geological Survey Parkfield Dense Seismograph Array [upsar]) about 12 km west of the rupture initiation of the 28 September 2004 M 6.0 Parkfield, California, earthquake, we have observed the movement of the rupture front of this earthquake on the San Andreas fault. The sources of high-frequency arrivals at upsar, which we use to identify the rupture front, are mapped onto the San Andreas fault using their apparent velocity and back azimuth. Measurements of apparent velocity and back azimuth are calibrated using aftershocks, which have a compact source and known location. Aftershock back azimuths show considerable lateral refraction, consistent with a high-velocity ridge on the southwest side of the fault. We infer that the initial mainshock rupture velocity was approximately the Rayleigh speed (with respect to slower side of the fault), and the rupture then slowed to about 0.66 β near the town of Parkfield after 2 sec. The last well-correlated pulse, 4 sec after S, is the largest at upsar, and its source is near the region of large accelerations recorded by strong-motion accelerographs and close to northern extent of continuous surface fractures on the southwest fracture zone. Coincidence of sources with preshock and aftershock distributions suggests fault material properties control rupture behavior. High-frequency sources approximately correlate with the edges of asperities identified as regions of high slip derived from inversion of strong-motion waveforms.

We analyze a long time–space series of Stromboli volcanic tremors. A very-low-frequency content in the range of 0.02–0.5 Hz has been found by using spectral analysis and independent component analysis. Independent component analysis is an entropy-based technique. We observe the occurrence of a component having a period of 30 sec. Polarization analysis shows that the wave field comes mainly from the crater area, well evidenced by seismometers located around the summit ring, whereas the radiation becomes increasingly scattered at stations located around the base of this volcano. Based on its apparent velocity, the 30-sec component appears to be a slow wave, related to inhomogeneities of the source and/or gas-pressure fluctuations inside the shallow plumbing system.

Seismological monitoring of the Comprehensive Test Ban Treaty (CTBT) requires detailed knowledge of travel-time characteristics of seismic phases across large areas. We use first-arrival travel times from several of Russian Deep Seismic Sounding (dss) profiles, primarily those from Peaceful Nuclear Explosions (pnes), to construct a 3D regional travel-time model of northern Eurasia. The method used, which can be viewed as a generalization of the existing regionalization techniques, is based on apparent-velocity-based spatial interpolation of the travel times picked from dss records. The resulting travel-time field is described in terms of an apparent velocity model in the 3D space of geographic coordinates and ray parameter. In the travel-time calibration, this model could be used to construct approximate source-specific station correction surfaces for any location within the region and also as a region-specific reference model to be used as a background for further calibration effort. This approach allows incremental refinement of the model as additional travel-time data become available. By converting the travel-time model into depth, a 3D regional velocity model is obtained, providing a description of the general features of the upper mantle in northern Eurasia.

We present here two epochs of Space VLBI Observations at 18 cm of the BL-Lac type object Mkn 501. Thanks to the high resolution of these new data we have found that the inner jet is centrally brightened at its beginning but becomes extended and limb brightened at ∼ 8 mas from the core. Moreover a comparison between the two epochs shows the presence of a proper motion with apparent velocity = 6.7c. Observational data have been used to constrain the jet velocity and orientation.

Seismic signals are often irregularly sampled along spatial coordinates, leading to suboptimal processing and imaging results. Least‐squares estimation of Fourier components is used for the reconstruction of band‐limited seismic signals that are irregularly sampled along two spatial coordinates. A simple and efficient diagonal weighting scheme, based on the areas surrounding the spatial samples, takes the properties of the noise (signal outside the bandwidth) into account in an approximate sense. Diagonal stabilization based on the energies of the signal and the noise ensures robust estimation. Reconstruction by temporal frequency component allows the specification of varying bandwidth in two dimensions, depending on the minimum apparent velocity. This parameterization improves the reconstruction capability for lower temporal frequencies. The shape of the spatial aperture affects the method of sampling the Fourier domain. Taking into account this property, a larger bandwidth can be recovered. The properties of the least‐squares estimator allow a very efficient implementation which, when using a conjugate gradient algorithm, requires a modest number of 2-D fast Fourier transforms per temporal frequency. The method shows signicant improvement over the conventionally used binning and stacking method on both synthetic and real data. The method can be applied to any subset of seismic data with two varying spatial coordinates.

We study the early evolution of active regions in the corona by analyzing 56 emerging flux regions (EFRs). We find that the initial apparent velocity of the expansion of the EFRs is 0.4–4.6 km/s, which is much lower than that inferred from Ha observations of arch filament systems or from theory.

Velocity estimation is examined in 3-D layered structures formed by plane and curved interfaces. The applied technique of coherency inversion tests the layer velocity through the repeating sequence of ray migration/coherency measurement. The reconstructed velocity‐depth model fits zero‐offset reflection times and maximizes semblance on input common midpoint (CMP) gathers. The correctness of layer velocity analysis disregarding the three‐dimensionality of the structures is under consideration. Using the 2-D coherency inversion technique, velocity is correctly determined in the upper layer of the examined structures. Two‐dimensional analysis in the deeper layer gives biased velocity estimates. The errors in the 2-D velocity estimates vary with the profile azimuth and appear in the form of the apparent velocity anisotropy. The inaccuracy of 2-D velocity estimation is analytically considered for the profile oriented along the refractor strike direction. The derived equation relates the velocity error to structure geometry and to the velocity contrast above and below the refractor. Three‐dimensional velocity analysis in the examined structures reveals that the layer velocity resolution is affected by the refractor shape. Below the convex refractor the velocity resolution deteriorates compared with that below the plane.

Waveforms from the NORESS array were analyzed for 147 industrial explosions during the 1985 to 1988 period, along a profile running east from Oslo (NORESS) to Helsinki to Leningrad (OHL profile). The events were 250 to 1300 km from NORESS and had local magnitude in the range 2.0 to 3.5. Event locations and origin times constrained by the University of Helsinki's regional seismic network provide a reliable basis for travel-time estimation at NORESS. We also used data recorded by NORSAR in 1979 for three shots on the FENNOLORA north-south, long-range seismic profile, which were near the OHL profile. Analysis of mantle P-wave signals from the explosions showed that first arrivals could be traced continuously to a distance of 750 to 800 km, where there is a cutoff and shift of approximately 2.0 to 2.5 sec in the travel-time curve and an increase in average apparent velocity. Interpretation of the observed travel times and waveforms for this profile suggests a low-velocity zone from approximately 105 to 135 km depth. Combined analysis of the seismic data with a Bouguer gravity map indicates the presence in the upper mantle of a high-velocity, high-density body of linear extent approximately from 200 to 300 to 500 to 600 km east of the NORESS array. It is postulated that this body may represent the root of an ancient volcanic system, in which lighter, silicic constituents were depleted from the upper mantle during the eruptive phase.

In the Tohoku District, the northeastern part of Honshu, Japan, a conspicuous phase is frequently observed between P and S phases in seismograms for events in the lower seismic plane of the double-planed deep seismic zone. This phase is dominant in the vertical component and has an apparent velocity almost equal to or slightly smaller than the P phase. We have never found this phase in seismograms for events in the upper seismic plane. From these characteristics, the phase is interpreted as SP, converted from S to P at the upper boundary of the descending oceanic plate (UBP). Using the travel times of the SP waves, we estimated the location of the UBP beneath the Tohoku District to be just above the upper seismic plane of the double-planed deep seismic zone. This position is consistent with those determined from the analyses of the travel times of the ScSp and PS waves, which are respectively the seismic wave converted from ScS to P and that from P to S at the UBP.

Downgoing waves in multicomponent VSP experiments are used to obtain seismic P- and S-wave velocities as a function of depth and angle of incidence. If P and SV waveforms do not overlap in time at the depth of interest, local velocities of the medium are obtained by separate analysis of these events. The apparent velocity of the event (P or SV) is computed from the moveout across several neighboring depth locations. The angle of incidence of the same event is computed from the particle‐motion hodogram within an appropriately chosen time window. Then, the local medium velocity (P wave or S wave depending on the chosen event) is given by the apparent velocity multiplied by the cosine of the angle of incidence. Layer interfaces with reasonably sharp velocity contrasts are efficient P-wave to SV-wave converters, even at moderate angles of incidence. In offset VSP experiments, converted SV waves are generated with varying strengths at practically all depths. Consequently, the converted SV waveforms partially overlap with the direct P waveforms, making the separate event analysis difficult and inaccurate. These overlapping waveforms can be handled properly by modeling the data in a given time window as a superposition of several events. In particular, the downgoing data at each depth level are modeled as a superposition of a P wave and an SV wave, with local P and S velocities, angles of incidence, and waveforms as model parameters. These parameters are then estimated by minimizing the squared error between the observed data and the model‐generated data. The unknown waveforms are eliminated from the minimization problem, leaving only four nonlinear parameters (velocities and angles) for estimation. Once these four parameters are found, least‐squares estimates of waveforms are obtained by evaluating a simple expression.

[1] A vapor cloud explosion occurred at an oil depot near Buncefield in the UK in 2005. Three infrasound arrays in the Netherlands detected various stratospheric phases. Some of these phases appeared with celerities, i.e., horizontal propagation velocity, in the conventional range of 0.28 to 0.31 km/s. Exceptionally fast arrivals, infrasonic forerunners, were identified with celerities of 0.31 to 0.36 km/s. These phases could be explained by head-wave-like propagation in a high velocity acoustic channel between 40 and 50 km height, where stratospheric zonal winds reached values of 120 m/s. The manifestation of infrasonic forerunners is validated by modeling with raytracing through actual atmospheric models and determining the celerity, apparent velocity and back azimuth. One phase occurred with a celerity of 0.25 km/s. Hence, we propose a new celerity range for fast stratospheric phases of 0.31 to 0.36 km/s and to lower the limit of the conventional range to 0.25 km/s.

During last two decades a lot of new seismic experiments were carried out for the upper mantle structure investigation. They covered different tectonic regions of continents and oceans and enable to determine principal features of deep structures and their correlation with geological history and geodynamics. All the studies showed a heterogeneous velocity structure of the upper mantle, existence of the first-order discontinuities, complicate heterogeneous layers, low and high-velocity zones and velocity anisotropy [1,2,3]. Some correlations were determined between these inhomogeneities and thermal regimes, ages of structures and gross tectonic features. But from the velocity models published at that time it was difficult to see if the upper mantle stratification has some regional features, and if there are some layers or seismic boundaries more pronounced or more significant than others. Even the lithosphere-asthenosphere boundary was difficult to distinguished from the very complicate layered models. The most part of the models were limited by a depth of 70–100 km.

Deep seismic soundings (DSS) from long-range recordings along profiles in northern Eurasia are used for the two-dimensional inversion and interpretation of lateral variations of the lithosphere-asthenosphere system. The major steps in developing a suitable model for travel-time inversion are shown. Modelling of nuclear-seismic data from the refraction — wide-angle reflection line QUARTZ in western Eurasia using a 2-D travel-time inversion method for layer boundaries and elastic wave velocities simultaneously, has quickly and efficiently resolved major structural differences between the East European Platform to the northwest of the Urals and the Siberian Platform to the southeast. Differences in Moho depths range between 36km under the European platform in the northwest and about 50km below the center of the Siberian basin. Subcrustal lithospheric velocities vary laterally from 7.90km/s under the Altai — Sayan folded region, 8.21km/s in the Siberian basin, 8.15km/s southeast of and below the Urals mountain belt to an average apparent velocity of 8.24km/s below the European platform. The type of boundary between the lithosphere and the asthenosphere, a sharp first order discontinuity in the northwest and transition zone in the southeast, and strong variations in the mantle transition zone depths have been resolved by minimizing the travel-time residuals for the model with a weighted least-squares algorithm. Nevertheless further modelling is needed to explain all structures of the two dimensional wavefield, as shown in an example for shotpoint 123 recorded to the south.

A small L-shaped short-period seismic array was set up at Etna Volcano to measure the wave velocity and the incoming azimuth of the volcanic tremor generated by the volcanic activity at Etna in non eruptive-conditions. Cross-correlation techniques were used to measure the phase-shifts between the array stations. An approximately N-S direction of the propagation direction was estimated, coincident with the azimuth of the crater area of Etna. An apparent velocity of less than 500 m/s was estimated, suggesting a surface propagation of the tremor waves. Spectral analysis, confirming previous results, shows velocity spectra peaked at approximately 2 Hz. A localized source, coincident with the crater area, agrees with our data and could generate the volcanic tremor at Etna in non-eruptive stages.

A limited, experimental seismic refraction survey over the Elura massive lead?zinc sulphide orebody obtained travel time data and time?distance plots with distinctive features. These can be regarded as a seismic signature of a zone of low seismic velocity, limited in dimensions by the higher velocity weathered host rock, and the much higher velocity fresh host rock and ore at depth. At Elura, these are produced by the low velocity plug of gossan which extends down through the weathered bedrock to the level of the fresh host rock and ore at a depth of about 90 m. The signature shows low seismic apparent velocities across the area of limited surface gossan. These apparent velocities are only slightly greater than the velocity of the surface material. In addition, there occur marked negative apparent velocity segments on the travel time?distance plots. These only occur on the further edge of the low velocity (gossan) zone to that of the location of the shotpoints. It is considered that these features or seismic signatures are produced by diffraction effects around the low velocity zone.

A direct measurement of apparent velocities for oceanic paths was made with an array of sensitive ocean bottom seismographs. The measurement was performed by recording waves from shallow earthquakes which occurred in the area close to trench axes and which were accurately located by the land seismological network in Japan. The range of epicentral distances is from 500 to 1,800 km. The observed P travel times are less than those in the Jeffreys-Bullen tables by 6–10 s for the range of distances. Since the dimension of the OBS array is about 400 km, the apparent velocities are determined quite precisely and show little dependence on the epicentral distances. The average value of the apparent velocities for the range 500–1,700 km is 8.64 ± 0.13 km/s. An offset of travel times, which is thought to be associated with a low-velocity layer underneath the oceanic lithosphere, has been observed. These results indicate that a high-velocity layer with a velocity of 8.6 km/s exists in the lower part of the oceanic lithosphere. Beneath the 8.6-km/s layer there is a thin low-velocity layer under which the velocity of the P wave is again 8.6 km/s.

Over 200 earthquakes in the distance range 30°–90° and azimuthal range 0°–360°, recorded at Indian WWSSN stations, have been used in the present study. We have treated the four WWSSN recording stations i.e. New Delhi, Poona, Shillong and Kodai-Kanal, as Super Large Aperture Seismic Array (SLASA) Network with New Delhi being its cross-over point. Short period P-wave data as obtained from these stations have been analysed using a least square technique. Slowness and azimuthal anomalies have been computed for all these events. Relative time residuals have also been calculated. A velocity model has been derived on the basis of the slowness and travel-time data. The results do not indicate presence of any triplication in the travel-time curve. Variations in the relative residuals refer to the tectonic features beneath the recording stations. The P-wave velocity increases continuously in the lower mantle region and there is no indication for the presence of any appreciable velocity gradient.

Poisson's ratios of the upper and lower crust and the sub-Moho mantle beneath central Honshu, Japan, are investigated using three independent methods that are based on S to P ratios of apparent velocities, the Wadati diagrams and an inversion of P and S arrivals. Shallow earthquakes at distances of 200—500 km from the Nagoya University Telemeter Network are used for the apparent velocity ratio method. Crustal and subcrustal earth-quakes under the network are used for the other two methods. The network consists of wide-band seismometers with three components which are particularly suitable for detecting S waves. The three different methods give a consistent result for Poisson's ratio σ, that is, (1) σ = 0.23 ± 0.01 in the upper crust, (2) σ = 0.26−0.28 in both the lower crust and in the sub-Moho mantle. The result indicates a sharp contrast in σ between the upper and the lower crust rather than at the Moho. The low σ in the upper crust can only be explained by the presence of a substantial amount of free quartz, indicating granitic rocks. A higher σ in the lower crust suggests that this portion is presumably less saturated in silica and may be even undersaturated, pointing to intermediate to mafic rocks. The sub-Moho σ is almost equal to the σ averaged over the entire upper mantle that has been estimated from the Wadati diagrams of deep shocks beneath Japan but is higher than those calculated from Pn and Sn velocities in oceanic and stable continental regions.

We present here results of three refraction profiles obtained during the “Minos” cruise (March–April, 1986) in the northern Red Sea (above 25°N) with our new marine seismic observation system, the Ocean Bottom Vertical Seismic Array (OBVSA). In the first part of the paper, we briefly describe the instrument and give some of its main characteristics. In the second part, we present a velocity-depth model for each one of the three non-reversed profiles. The model is first inferred from interpretation of the hodochron, using the slope intercept method and assuming plane interfaces, and iteratively modified by application of a ray tracing program. Our results support other data collected durng the cruise with Expanding Spread Profiles and reported elsewhere in this issue by Gaulier et al. (1988). The two northern profiles are about 50 km long and are oriented N15 °E. The recording instrument was located at the southern extremity of the profiles at 26°35′N, 34°46′E and 27°00′N, 34° 26′E respectively. The basement velocity was found to be in the 5.2–5.8 km s −1 range. The Moho is located at about 14 km below the sea level. The upper mantle apparent velocity is about 7.3 km s −1 . The southern profile is 40 km long, oriented S45° W with the OBVSA located at its eastern extremity, at 26°06'N, 35°23′E. The Moho depth is about 11.5 km. The crustal structure under the eastern part of the profile is interpreted as being of the oceanic type.

By limiting data input to the forward‐inverse tau‐p transform pair, velocity filtering which varies with time and offset can be achieved. The limiting procedure, called hyperbolic velocity filtering, suppresses coherent noise while retaining reflection events. It is based on knowledge of the stacking velocities, but only approximate values are needed. A suitable application for hyperbolic velocity filtering is in hard water‐bottom marine areas where laterally propagating noise, diffracted by shallow scatterers, gives rise to reflection‐like events with high apparent velocity in the CMP gathers. Such diffractions do not, however, generally have reflection‐like moveouts in shot or receiver gathers. Hyperbolic velocity filtering offers moveout discrimination at all offset distances and traveltimes. When it is applied to shot (and receiver) gathers from a hard water‐bottom area such as offshore Florida, a striking reduction of noise occurs in the stacked section. Comparison with similar application of fixed‐limit (f-k) velocity filtering shows the hyperbolic filtering to be more effective for this data set.

A 275-km-long reversed refraction profile in the Oregon Cascades, two shallow earthquakes of magnitude 5 in southern Washington, a shallow earthquake of magnitude 4.6 in northern California, and a previously published analysis of the Bouguer gravity field are used to develop a crustal P wave velocity model for the Oregon Cascades. Travel time analysis of the refraction profile indicates a crustal structure characterized by surface layers with P wave velocities that vary from 2.9 to 5.2 km/s and thicknesses that vary from 2.5 to 5.0 km, upper crustal velocities of 6.1 to 6.5 km/s between the depths of 3 and 29 km, lower crustal velocities near 7.0 km/s between the depths of 29 and 44 km, and a mantle reflector at a depth of 44 km. Comparison with synthetic seismograms supports this structure and shows that the lower crustal and Mono transitions can be better modeled by continuous velocity gradients than by first-order discontinuities. Arrival times from two shallow earthquakes in southern Washington in 1981 across 14 telemetry stations in the Oregon Cascades show an apparent velocity of 7.62 km/s. On the basis of a previously published gravity analysis, the mean north-south component of the dip to the crust-mantle boundary is estimated at 1° down dip to the south. The apparent velocity from the earthquakes and the dip estimate from gravity indicate that the true Pn velocity is 7.70 km/s. Arrival times from a shallow earthquake in northern California in 1978 across telemetry stations in the Washington Cascades are consistent with this model and indicate that the upper mantle velocity reaches 8.20 km/s at a depth of approximately 100 km.

Seismic data from explosions and local earthquakes are used to determine shallow crustal structure near Managua, Nicaragua, and to derive a complete crustal model for northern Costa Rica. A new method, referred to as the minimum apparent velocity (MAV) method, has been developed and applied in the analysis of local earthquake data. The results indicate that the MAV method of multilayer analysis, when combined with well-determined station corrections, can be quite useful in regions of high seismicity. Based upon the analysis presented here, the total thickness of the crust beneath the central volcanic province of northern Costa Rica is about 43 km. Four crustal layers are identified with compressional-wave velocities of 2.6, 5.1, 6.2, and 6.6 km/sec. The upper mantle velocity is 7.9 km/sec. The major features of this model closely resemble those of other seismically active, marginal zones of the Pacific.

Relative Pn arrival times across an array of stations at The Geysers-Clear Lake region in northern California indicates that the upper mantle velocity is 8.0 km/sec, and that the Moho dips 5.3° at 57° to the northeast in this area, reflecting thickening of the crust toward the continent. These results agree with regional trends in the Bouguer gravity field. The travel-time residuals with respect to the dipping model suggest a crustal low-velocity zone beneath Mt. Hannah, consistent with reported teleseismic delays, and the low-velocity zone is interpreted as representing partial melt. No delays are observed in The Geysers, precluding the existence of an extensive magma body beneath the steam production area. Azimuthal variations in apparent velocity may reflect upper mantle azimuthal velocity anisotropy, but such an interpretation is uncertain due to the limited azimuthal distribution of earthquakes.

Locations and velocities were calculated for microseisms occurring in samples of rock subjected to triaxial loading and injection of pore fluid. This was accomplished by analyzing arrival times of acoustic emission using an automatic first arrival picker. Apparent velocity anomalies were observed prior to both failure of intact samples and violent slip in samples containing saw cuts. Further analysis revealed that these fluctuations in calculated velocity were not due to changes in the true seismic velocity. Instead, variations in calculated velocity are shown to be related to sampling errors in picking first arrivals. The systematic picking of late first arrivals for small magnitude events was found to be a persistent bias resulting in low calculated velocities. This has encouraged the reexamination of earthquake records to determine how important sampling biases are in contributing to reported velocity anomalies.

Ocean bottom seismograph (OBS) long-range explosion experiments were carried out in the Mariana Basin in 1973 and 1976. Seven large shots (8.5–1.5 ton) as well as several tens of small shots were fired. The maximum range of observation was about 1900 km. As many as 25 OBS stations were deployed in an array of about 800 km. It is found that the sub-Moho P-wave velocity structure is of stratified nature, being composed of alternating high- and low-velocity layers. High-velocity layers with apparent velocities of 8.1, 8.2, 8.4, 8.6 and 8.7 km/s are identified. Low-velocity layers, sandwiched between the high-velocity layers of 8.4, 8.6 and 8.7 km/s, are very prominent. The sub-Moho high-velocity lid with an apparent velocity of 8.4 km/s is very thin. Thinning of this lid, thickening of the low-velocity layer, and the presence under it of another high-velocity layer (8.6 km/s) appear to characterize the uppermost mantle structure beneath the Mariana Basin.

Summary The gross structure of the Earth'sinterior is fairly well known, but we are still lacking information on the finer details, i.e. structural discontinuities of higher orders. A powerful tool in investigations of this type of problem is P-wave travel time, and specifically the parameter DT/DΔ. We have investigated this problem, taking advantage of the concept of continental arrays. Reported arrival time data (seismic bulletins) for the Fennoscandian network have been used for direct measurements of apparent velocity (DT/DΔ) and direction of approach of P-waves from 648 seismic events. Our observations are interpreted in terms of vertical anomalies at depths around 850, 1050, 1250, 1700 and 2600 km where the velocity changes very slowly with depth. The corresponding epicentral distances are 35, 47, 53, 62 and 87 deg. In addition, we have strong evidence for existence of lateral P-velocity variation which amounts to around 0-1 km s–1 in the depth interval 1750–2300 km and the distance range is around 63–80 deg. A comparison of our data with those presented by others favours lateral velocity variations also at depths around 700–800 km and the corresponding distance range is 25–30 deg.

Summary Apparent velocity measurements of P arrivals from Mid-Atlantic Ridge earthquakes across pairs of stations in Iceland and Greenland have been made using bulletin data. They define both the vertical and horizontal extent of the low velocity anomalous mantle zone between 50° and 70° N.

On a group of seismic refraction records there occurs a late arrival with the same apparent velocity as an earlier arrival. It was first thought that the late arrival travelled along the same refractor as the earlier one but had been delayed in the overburden either by multiple reflection or by P‐S (dilatation to shear) conversion. Either of these two mechanisms could be made to fit the time‐distance data. A study of the ratio of the amplitudes of the two events showed that the observed ratio was about ten times that to be expected on either of these hypotheses. Also, the rate of attenuation of the earlier arrival (3.7 ± 1.0 db/1000 ft) was appropriate to a thin layer while the rate for the later arrival (0.74 ± 0.30 db/1000 ft) was appropriate to a thick layer. Accordingly, the later arrival was identified as a refraction from a deeper layer. The subsequent depth section agreed very well with that found by drilling.

The dispersion of shear-coupled PL waves, PL(S) and PL(SS), has been studied for several paths on the Canadian Shield over the distance range 33.4° to 115.5° with the twofold aim of elucidating the manner in which these waves are generated and of examining their value in investigating the structure of the crust and upper mantle. In many cases these waves are composed of several wave trains quasi-stationary in period. These observations can be explained by considering the large amplitudes associated with cusps in the S-wave travel-time curve and the adjacent flat portion of the S-wave apparent velocity curve produced by velocity discontinuities in the upper mantle; little-dispersed PL(S) trains appear to be generated in regions where S waves of very high amplitudes impinge on the crust. In addition, the recovery of the periods of these individual wave trains in the PL(S) record can provide information on the crustal structure in a relatively localized region and on the ray parameter close to each cusp.

The results of a study of seismic waves generated by eight nuclear explosions and recorded at 31 locations between the Nevada test site (NTS) and Ordway, Colorado, are discussed. The line of recording stations crosses the eastern part of the Basin and Range province, the Colorado plateaus, and the southern Rocky Mountains, and it extends into the Great Plains. In the eastern Basin and Range province and the western part of the Colorado plateaus (0≤Δ≤385 km), the time-distance curves for Pg and Pn can be expressed, respectively, as T1 = 0.8 + Δ/6.0 and Ts = 5.8 + Δ/7.6. A third phase, tentatively identified as P*, is represented by the equation T2 = 3.8 + Δ/6.5. Using the crustal structure and Pn velocity (7.9 km/sec) found for the NTS region by other authors, we find that the above relations indicate that the thickness of the crust increases from about 25 km at NTS to about 42 km in the western part of the Colorado plateaus. East of this boundaiy the velocity of P in the upper mantle increases to 8.0 km/sec; depth to the M discontinuity in the Colorado plateaus is approximately constant over the range 435

When a seismic reflector is a dipping plane and the subsurface can be approximated by a single bed, the velocity needed to stack CDP data is higher than the velocity with which the energy travels in the subsurface. The ratio of the stacking velocity to the energy‐travel velocity increases from unity for strike lines to the secant of the dip angle for dip lines. Although the reflection points are not common, reflections stack even for steep dips. Stacking velocities for multiple reflections increase with the order of the multiple, but multiple reflections may also be stacked. To convert from reflection time to depth requires knowledge not given by the stacking velocity. In theory, a split‐spread procedure can furnish the needed information.

Summary The paper describes a radio-linked short-period seismometer network now operating in Scotland and recording on analogue magnetic tape. The network will be used as a permanent network for time-term refraction studies, and for monitoring the minor seismic activity within the area. The magnetic tapes may be digitized in the laboratory, processed in a computer, and the network used as an array of 100-km aperture. The aperture allows fine resolution between beams for determining apparent velocity and azimuth of teleseisms.

Several analytical techniques have been used to derive crustal structure from the data obtained in the 1963 Lake Superior seismic experiment conducted as a part of the United States and Canadian programs in the Upper Mantle Project. The Lake Superior region was found to have an unusual and rapidly varying crustal structure in comparison with the Canadian shield area surrounding it. In general, under the lake, a high-velocity crust (6.67 km/sec) was found beneath 4 to 6 km of Keweenawan sediments and volcanics. The upper mantle velocity of 8.07 km/sec agrees closely with values found previously in adjacent areas. The depth to the M discontinuity varies from about 20 km in the area just west of the lake to 55 km or more in the eastern half of the lake. These values are among the most extreme yet observed in North America. A number of second-order structural features were also found in the M discontinuity. These differences and the major structures persist for all reasonable assumptions about the velocity-depth function. Of the techniques used, the standard reverse profile technique is poor for this complex structure, and the apparent velocity measurements provided limited information about the velocity-depth function. The extended time-term technique developed and used here produced the most complete structural picture, and the general features of this structure are confirmed by a dip projection method derived from apparent velocity data.

Seismic-refraction measurements from chemical explosions near San Francisco, California, and Fallon and Eureka, Nevada, were made along a line extending nearly 700 km inland from San Francisco across the Coast Ranges, Great Valley, Sierra Nevada, and Basin and Range province. The velocity of Pg in the Basin and Range province was found to be 6.0 km/sec. Between Fallon and Eureka the velocity of Pn is 7.8 km/sec, and just east of the Sierra Nevada it is about 7.9 km/sec. Two prominent phases closely following the first arrival between 50 and 250 km from the source in the Basin and Range province were interpreted as reflections from an intermediate layer and from the M discontinuity. The velocity of P in the possible intermediate layer, deduced from the reflected phases because the refracted wave expected from this layer is nowhere a first arrival, seems to be 6.6 km/sec at the top of the layer and probably increases with depth. In the Coast Range northeast of San Francisco, the velocity of P in the crust, 5.4 to 5.6 km/sec, is substantially smaller than in the Basin and Range province. When corrected for abnormal delays in the sediments, first arrivals from San Francisco across the Great Valley suggest a Pn velocity of 7.9 km/sec. The low apparent velocity of first arrivals on this profile beyond the valley indicates that the M discontinuity dips downward beneath the Sierra Nevada. From a depth of about 20 km beneath the Coast Ranges and Great Valley, the M discontinuity descends to at least 40 km beneath the high Sierra Nevada. East of the Sierra Nevada it rises rapidly to about 22 km beneath Carson Sink and then dips downward toward Eureka, where it reaches a depth of 32 km. If the intermediate layer is real, the foregoing depths are increased by 2 to 5 km.

Evidence for a velocity increase at shallow depth within the upper mantle was found during a seismic refraction experiment in the Gulf of Mexico. The shots were fired in the Gulf of Mexico and observed in Mexico and Florida. This paper reports on the analysis of the travel times to the Mexican stations and on their interpretation in terms of crust and upper-mantle structure. A break in apparent velocity from about 8 km/sec (corresponding to Pn) to a velocity of 8.77 km/sec occurred at a distance of about 360 km. This change in apparent velocity is interpreted as arising from a discontinuity, or rapid increase of velocity with depth, at a depth of 57 km. It is suggested that the discontinuity is due to a phase transition. The P amplitudes decrease appreciably beyond 700 km. The reduction in amplitude is interpreted as being due to penetration of the rays into a layer of decreasing velocity. Several models with low-velocity zones between 100 and 200 km depth are consistent with these observations.

Seismograms recorded during the Trans-Scandinavian Profile of 1969 at the stations Kajaani, Oulu, Maaselka and Sodankyla are used to determine theP wave velocities for distances from 400 km to 1300 km. An apparent velocity of 8.50 km/s was found. After making a correction due to the curvature of the earth the velocity of about 8.40 km/s was obtained at an approximate depth of 80 km. Some comparisons are made with other results from Fennoscandian area.

Attention is given in this paper to the display of intensive seismic refraction information in a convenient form similar to spatial display techniques being developed for resistivity methods. An observational space for first arrivals is defined and illustrated with respect to some simple structures and the possible use of such a representation in diagnosis is discussed. The spatial display is suitable both for steeply dipping and gently dipping structures. In the latter case, the low relief space may be subjected to further operations to produce apparent velocity distributions and values simply related to plus times. A practical example of the latter process and the subsequent interpretation is given.

Apparent-velocity variations across the large-aperture seismic array (Lasa) have been studied by using P waves from six teleseisms at different distances and azimuths around Lasa. Data from 13 instruments per subarray were used in the analysis. Average apparent velocities between subarrays, taken in pairs, were measured by using arrival-time data. Apparent velocities at subarrays were estimated by a cross-spectral method involving the whole signal shape. For the 2 events in the southeast direction, there is a large zone of low apparent velocity to the northwest of A0 subarray. For the events in the other azimuths, regions of low and high apparent velocities are clearly contourable across Lasa. It appears that a 30% variation in apparent velocity exists across Lasa for events from all azimuths. The apparent-velocity measurements at subarrays, in general, follow the pattern shown by the measurements based on arrival-time data. However, there are some significant differences, especially at the subarrays in E and F rings. The apparent-velocity measurements indicate that structural complexities over and above those seen in the average crustal models proposed so far exist under Lasa.

To identify small, deep earthquakes in South America quadripartite nets of about one km in size were set up in Toconce, Chile and Cuzco, Peru. During the period of observation the U. S. Coast and Geodetic Survey identified two fairly large, deep earthquakes in northern Argentina within about a month by teleseismic observations, but neither foreshocks nor aftershocks associated with these earthquakes during the period of observation may indicate that there are comparatively few small, deep earthquakes as compared with shallow earthquakes. In Toconce, at the foot of the Andes, there was a systematic azimuthal distortion in apparent velocity and in direction of propagation, whereas in Cuzco, on the Altiplano, such a distortion was not found.

Preliminary travel times of diffracted P waves have been determined to distances of 167° and found to have an apparent velocity at the surface of 24.55±0.08 km/sec, rather than the Jeffreys-Bullen value of 25 km/sec. The diffracted wave velocity has been used with Jeffreys' solution of the Herglotz-Wiechert equation (giving the velocity structure of the mantle) to determine a core size of 3537±9 km and a lower mantle velocity of 13.64±0.045 km/sec. These values are shown to be consistent with a shadow boundary determination of 96°, established by different methods in an earlier study. The dispersion is less than 1% over a 25- to 2.5-sec period range, suggesting absence of structure at the base of the mantle. There is, however, suggestion of circumferential irregularity at the core-mantle interface.

The purpose of the sonic log is to measure formation velocity, but various geometrical factors can cause apparent velocity readings substantially different from the true velocity at the depth of measurement. An understanding of these factors enables one to derive greater value from the log. Bed thickness, hole size, caves, alteration of the formation adjacent to the bore, spacing of the transducer, and whether the tool is centered in the bore, are among the factors influencing the log. Centering the tool can increase the signal strength, but in large holes and slow formations the first arrival may come through the mud. The transmitter‐receiver spacing which is needed to avoid this is given for various hole sizes and formation velocities. The effect of caves is shown with theoretical and actual curves. There is some evidence that action of the mud or stress concentration around the bore can alter the velocity of the formation; the effect on the indicated velocity is shown as a function of transducer spacing and of degree and depth of velocity alteration.

By measuring the differences in phase angle of Fourier spectral densities across a fixed array, apparent velocities may be determined at a number of frequencies. Because the analysis is carried out in the frequency domain, determination may be objectively restricted to regions of the spectrum giving consistent phase relations between array elements. Using an array of approximately 1.25 km length, the apparent velocities of the first arrivals from the 1962 North Carolina experiment were determined for shots from 45 to 155 km. No indication of an increase in velocity with range between 55 and 140 km was found.

This paper contains a brief review of the operations that are involved in digital processing of array seismic recordings by the methods of velocity filtering, summation, cross-multiplication and integration, and by combinations of these methods (the UK method and multiple correlation). Analyses by the several techniques were made on array recordings that were obtained by the U. S. Geological Survey of seismic waves from chemical and nuclear explosions in the western United States. Seismograms were synthesized, using recordings of seismic noise and a Pn signal, such that the signal-to-noise ratio, onset time and apparent velocity of the signal were predetermined for the synthetic records. These records were then analyzed by summation, cross-multiplication and the UK method, and the results were compared. From the examples presented, it appears that for recordings with initial peak-signal-to-rms-noise ratio less than about 0.8, the best improvement can be obtained by the process of velocity filtering and summation of all the traces. For a six-element array improvement can be obtained by this method on records which have an original signal-to-noise ratio of 0.4 or greater. When the initial signal-to-noise ratio is greater than 0.8, multiplicative compounding yields about the same improvement as a process involving additive compounding and subsequent raising of the summation trace to large powers.