Search

SUMMARY Deep focus earthquakes within the underthrust Indian lower crust beneath the Himalaya occur in very specific regions and have distinct source characteristics. The study of the source mechanisms of these earthquakes provides valuable constraints on the kinematics of deformation of the underthrust Indian Plate, and its influence on the active deformation of the overlying Himalayan wedge. One of the most significant regions of these deep focus earthquakes is beneath the Sikkim and Bhutan Himalaya. We study the source characteristics of the 2011 September 18 (Mw 6.9) deep focus Sikkim main shock and its major aftershocks using global, regional and local waveform data. We determined the focal mechanism of the main shock using moment tensor inversion of global P and SH waveforms, and ascertained the earthquake fault plane using rupture directivity from regional P-wave spectra. The main shock originated at 53 ± 4 km depth and ruptured at least 20 km thickness of the underthrust Indian lower crust. Faulting occurred on a near vertical dextral strike-slip fault oriented NW-SE (strike 127 ◦ ,d ip 81 ◦ and rake 167 ◦ ), oblique to the local strike of the Himalayan arc. The rupture initiated from the SE end of the fault and propagated to the northwest. The main shock was followed by 20 small-to-moderate aftershocks (mb > 3.0), which we relocated using phase arrival times. We computed the focal mechanisms of the larger ones (mb ≥ 3.5) using local waveform inversion. We find that all aftershocks originated SE of the main shock, between depths of 12 and 50 km, and have dominantly strike-slip mechanisms. Our results, combined with the source mechanisms of earthquakes from previous studies, reveals that the entire underthrust Indian crust is seismogenic and deforms by dextral strikeslip motion on oblique structures beneath the Sikkim and Bhutan Himalaya. These active oblique structures with transverse motion possibly mark the western boundary of the clockwise rotating ‘microplates’ in northeast India observed from GPS geodesy.

S U M M A R Y We explore the use of on-land GPS observations to detect deformation due to tsunami propagation near source regions of large interplate earthquakes. Here, we focus on the Mw 9 Tohoku-oki earthquake, which occurred around 14:46 (JST) on 2011 March 11. We consider GPS data in the time span 14:54–15:22 (JST) along the Sanriku coast, where the tsunami had the largest amplitude. The displacement data shows the signatures of large aftershocks as well as post-seismic fault slip (afterslip). These effects are particularly evident in the east component. From the horizontal displacement vectors, we construct a simple fault model for the early phase of the afterslip. Mean slip velocity of the early afterslip reaches 0.1 mm s−1. By compiling the early afterslip velocity of recent interplate earthquakes around that region, we find its increasing trend with the main shock magnitude. This scaling relation may reflect higher stressing rates at edges of larger main shock faults. Separately, we forward calculate land deformation due to tsunami height changes based on a tsunami simulation. Tsunami-induced deformation is only evident in the vertical direction at coastal GPS stations. The predicted subsidence amounts at some coastal stations can account for a large portion of the residuals between the observation and the modelled deformation due to the fault slip.

SUMMARY Earthquake source parameter determination is of great importance for hazard assessment, as well as for a variety of scientific studies concerning regional stress and strain release and volcano-tectonic interaction. This is especially true for poorly instrumented, densely populated regions such as encountered in Africa, where even the distribution of seismicity remains poorly documented. In this paper, we combine data from satellite radar interferometry (InSAR) and teleseismic waveforms to determine the source parameters of the Mw 5.9 earthquake that occurred on 2008 February 3 near the cities of Bukavu (DR Congo) and Cyangugu (Rwanda). This was the second largest earthquake ever to be recorded in the Kivu basin, a section of the western branch of the East African Rift (EAR). This earthquake is of particular interest due to its shallow depth and proximity to active volcanoes and Lake Kivu, which contains high concentrations of dissolved carbon dioxide and methane. The shallow depth and possible similarity with dyking events recognized in other parts of EAR suggested the potential association of the earthquake with a magmatic intrusion, emphasizing the necessity of accurate source parameter determination. In general, we find that estimates of fault plane geometry, depth and scalar moment are highly consistent between teleseismic and InSAR studies. Centroid-moment-tensor (CMT) solutions locate the earthquake near the southern part of Lake Kivu, while InSAR studies place it under the lake itself. CMT solutions characterize the event as a nearly pure double-couple, normal faulting earthquake occurring on a fault plane striking 350° and dipping 52° east, with a rake of –101°. This is consistent with locally mapped faults, as well as InSAR data, which place the earthquake on a fault striking 355° and dipping 55° east, with a rake of –98°. The depth of the earthquake was constrained by a joint analysis of teleseismic P and SH waves and the CMT data set, showing that the earthquake occurred in the shallow crust, at approximately 8 km depth. Inversions of ENVISAT (Environment Satellite) and ALOS (Advanced Land Observation Satellite) data place the earthquake at 9 km. A comparison of the scalar moment (9.43 ± 0.06 × 1017 Nm from seismology and 8.99 ± 0.010 × 1017 Nm from the joint InSAR solution) shows good agreement between the two data sets. Such an agreement is in contrast to the large discrepancies observed (up to an order of magnitude) in other places along the EAR where similar earthquake sequences are associated with magmatic intrusion. From this, we infer that the rupture was brittle and occurred with little aseismic deformation as might be expected from magma injection. Our results provide insights into the style of rifting occurring in the South Kivu Volcanic Province and hence will aid future studies on seismic risk in the context of Lake Kivu and underline the importance of systematic monitoring of the EAR area.

SUMMARY We study a large M w = 7.6 earthquake that occurred on 2007 November 14 in the Northern Chile seismic gap near the city of Tocopilla. Using a variety of seismic data we show that this earthquake ruptured only the lower part of the interplate seismic zone and generated a series of plate interface aftershocks. Two large aftershocks on 2007 November 15 ruptured the interplate zone oceanwards of the Mejillones Peninsula, a major geographical feature in the Antofagasta region. On 2007 December 16, a large M w = 6.8 aftershock, that occurred near the southern bottom of the fault plane of the main event, is shown to be a slab-push earthquake located inside the subducted Nazca Plate and triggered by along slab compression. Aftershocks of this event demonstrate that it occurred on an almost vertical fault. The Tocopilla earthquake took place just after the installation of a new seismological network by Chilean, German and French researchers. The accelerometric data combined with far field seismic data provide a quite complete and consistent view of the rupture process. The earthquake broke a long (130 km) and narrow (about 30–50 km) zone of the plate interface just above the transition zone. Using a non-linear kinematic inversion method, we determined that rupture occurred on two well-defined patches of roughly elliptical shape. We discuss the consequences of this event for models of gap filling earthquakes in Chile proposed in the 1970s.

SUMMARY The 2005 June 13, an intraplate, intermediate depth earthquake (M w = 7.8, 98 km depth) occurred in northern Chile. Previous studies show that this earthquake is a slab-pull event with down dip extensional source mechanism. However, the physical origin and the role in seismic cycle of this kind of event are still under debate. We present a seismological source study from strong motion data, based on the simplified slip patch kinematic parametrization. We find two distinct sources. The main source is almost centred around the hypocentre and triggers the secondary one, situated deeper in the slab. The weaknesses and uncertainties of the kinematic models are partly resolved and discussed by constructing dynamic spontaneous rupture models with two radiative asperities and one embedding zone of non-radiative slip process, a required feature to fulfil the seismic moment constraint. Two kinds of models are compatible with observations: a crack mode model (combined with fast rupture velocity and small zone of non-radiative slip) and a pulse mode model (combined with slower rupture velocity and large zone of non-radiative slip). The fast rupture model is preferred for its physical simplicity. Replacing our result in the geometry and thermal contexts of the subduction, we conclude that the rupture has probably broken the whole lithosphere by encountering little resistance and without being very sensitive to the presence of the double seismic zone. This suggests that the rupture happens along a weak plane, on which large slab instabilities can take place.

SUMMARY investigation of the SH wave seismograms indicates that rupture occurred on a shallowly dipping fault, propagating in all directions from the hypocentre. The average rupture velocity in the downdip and updip directions are ∼2.0-3 and ∼1.0 km s −1 , respectively. The maximum slip is found to be ∼12 m and is located downdip of the earthquake's hypocentre. Taken over the fault as a whole, the average stress drop is ∼0.55 MPa. Used in conjunction with relocated aftershocks, we find that two barriers constrain the faulting extent of slip, creating a rupture area of ∼165 km × 140 km. These barriers coincide with submarine topographic features, are domain boundaries for the seismic gaps along the arc, and have high concentrations of large aftershocks near them. The high slip regions align with the orientation of the subducting plate, and resemble in size and shape the many seamounts about to subduct. The 2003 Tokachi-Oki earthquake was the largest earthquake to occur globally in that year and reruptured the portion of the Kuril- Japan Trench that last broke in the great Tokachi-Oki earthquake of 1952 (Hamada & Suzuki 2004). The abundance of local data in this region coupled with the large magnitude of the event has led to a range of models for the rupture being proposed. These solutions are summarised in Table 1. All of these solutions have the total moment for the earthquake significantly below that obtained from the long-period seismological data by the global CMT (GCMT) solution, with some solutions having less than one third of the GCMT moment. This study aims to address this anomaly, by first investigating the long-period moment tensor solution and then by finding a rupture model which is consistent with both broad-band teleseismic data and the long-period solution.

SUMMARY We investigate the 2007 Pisco, Peru, earthquake (M8.0) using seismological and geodetic data. Analysis of teleseismic records indicates that the earthquake is characterized by a steadily increasing moment rate, with the maximum occurring between about 55 and 70 s and the associated slip occurring about 40 km south of the epicentre. We combine InSAR data from L-band (ALOS) and C-band (Envisat) satellites to produce a 3-D image of the ground displacements caused by the earthquake. Distributed slip inversions of the interferograms identify a single large patch of slip in the location of the maximum moment rate identified by seismology. Despite a clear spatial coincidence between the location of the coseismic slip and the anomalously high topography of the Paracas Peninsula, we find no evidence that the earthquake was directly responsible for motion on any upper plate faults.

We present an acoustic emission (AE) monitoring technique to study high-pressure (P > 1 GPa) microseismicity in multi-anvil rock deformation experiments. The application of this technique is aimed at studying fault mechanisms of deep-focus earthquakes that occur during subduction at depths up to 650 km. AE monitoring in multi-anvil experiments is challenging because source locations need to be resolved to a submillimetre scale due to the small size of the experimental assembly. AEs were collected using an 8-receiver array, located on the back truncations of the tungsten carbide anvils. Each receiver consists of a 150-1000 kHz bandwidth PZT transducer assembly. Data were recorded and processed using a high-speed AMSY-5 acquisition system from Vallen-Systems, allowing waveform collection at a 10 MHz sampling rate for each event signal. 3-D hypocentre locations in the assembly are calculated using standard seismological algorithms. The technique was used to monitor fault development in 3 mm long x 1.5 mm diameter olivine cores during axisymmetric compression and extension. The faults were generated during cold compression to similar to 2 GPa confining pressure. Subsequent AEs at 2-6 GPa and 900 degrees C were found to locate near these pre-existing faults and exhibit high pressure stick-slip behaviour.

SUMMARY We present results from forward modelling to study the feasibility of using S-to-P converted phases to image the seismic discontinuity structure of the crust and upper mantle. We show that a significant level of P-wave energy arriving before the direct S-wave arrival can interfere with the S-to-P converted phases of interest and may result in Sp receiver function phases that do not represent true earth structure. The source of this P-wave energy is attributable to a number of phases, including those that have undergone multiple reflections off the Earth’s surface. For deep focus earthquakes (300‐600 km deep), a significant amount of P-wave energy is observed from pPPP, pPPPP and sPPPP phases, and arrives within the same time window as predicted for S-to-P converted phases from the direct S phase arrival. Furthermore, for earthquakes at all depths, interfering P-wave energy arrives within the same time window as predicted for S-to-P converted phases from the SKS phase arrival, limiting the usefulness of SKSp receiver functions for upper mantle imaging. To isolate true Sp receiver function phases from contamination due to other P-wave phases, we find it necessary to stack receiver functions from a range of epicentral distances and depths in order to aid the suppression of noise and other unwanted phases. We provide constraints on the noise levels to be expected as a function of epicentral distance and earthquake depth. We find that the lowest noise levels are achievable by restricting epicentral distance to less than 75 degrees and the depth of earthquakes used to less than 300 km.

SUMMARY The ability of seismological criteria to identify earthquakes from underground explosions depends partly on the orientation of the earthquake source. Well-determined double-couple moment tensor solutions for a large number of earthquakes have been published in the Harvard centroid moment tensor (CMT) and United States Geological Survey (USGS) catalogues. Statistical analyses of these catalogues indicate that the distribution of the orientation of earthquake mechanisms is not random. The distribution of the T axes shows significant clustering around the downward vertical, indicating that a larger number of earthquake mechanisms radiate compressional P-wave energy to teleseismic distances from near the maximum of the radiation pattern than is predicted if earthquake sources are randomly oriented double couples. The clustered T axes correspond to compressional dip-slip mechanisms, and it is this type of mechanism which is believed to cause both the mb:kfs (the ratio of body-wave to surface-wave magnitude) and first-motion criteria to misidentify an earthquake as an explosion.

SUMMARY We determine the isotropic component of large deep earthquakes by inversion for their full six-component moment tensors using the normal-mode data. We show that it is possible to reduce the correlation between the isotropic component and other components and to determine the isotropic component independently by analyses of normal-mode data at periods longer than 500 s. We find no significant isotropic component for the earthquakes we studied; the magnitudes of the estimated isotropic components are comparable to the uncertainty due to the mislocation of the centroid. The magnitude of the isotropic component is at most 5 per cent of the deviatoric seismic moment if it exists. This suggests that, if a rapid phase change of mantle minerals occurs during the rupture of deep earthquakes, the mantle material that transformed would be confined within an extremely thin layer. We also conduct moment tensor inversion using body waves and surface waves in the period band between 20 and 500 s. We find that there is a relatively consistent non-double-couple component of the deviatoric moment tensor for the Japan Sea earthquake (1994 July 21). This is likely to be caused by source complexity

SUMMARY The observability of the isotropic component of a general moment tensor is investigated by carefully examining its resolvability from the other components. It is shown that the portion of seismograms between the first P-wave arrival and just before the arrival of the first surface wavetrain, which contains many different body-wave phases, can be used to constrain the isotropic component of deep earthquakes. CMT inversion using this portion of the seismogram for major deep earthquakes has revealed that there is no significant isotropic component for deep earthquakes. This conclusion contradicts some earlier studies, for example Dziewonski & Gilbert (1974), who found significant isotropic components of deep earthquakes by analysing normal-mode data. It is shown that the isotropic component of deep earthquakes cannot be resolved independently from the vertical CLVD component of the moment tensor by analysing the type of normal-mode data above 2mHz that previous researchers used. At the same time, however, the possibility of resolving the isotropic component using normal-mode data below 2 mHz is suggested.

SUMMARY In this paper we present revised locations and original focal mechanisms computed for intermediate and deep earthquakes that occurred within the Southern Tyrrhenian subduction zone between 1988 and 1994, in order to improve our knowledge of the state of stress for this compressional margin. In particular, we define the stress distribution within a large portion of the descending slab, between 40 and about 450 km depth. The seismicity distribution reveals a continuous 40-50 km thick slab that abruptly increases its dip from subhorizontal in the Ionian Sea to a constant 70" dip in the Tyrrhenian. We computed focal mechanisms for events with magnitudes ranging from 2.7 and 5.7, obtaining the distribution of P- and T-axes for many events for which centroid moment tensor (CMT) solutions are not available, thus enabling the sampling of a larger depth range compared to previous studies. We define three portions of the slab characterized by different distributions of P- and T-axes. A general down-dip compression is found between 165 and 370 km depth, whereas in the upper part of the slab (40-165 km depth) the fault-plane solutions are strongly heterogeneous. Below 370 km the P-axes of the few deep events located further to the north have a shallower dip and are not aligned with the 70" dipping slab, possibly suggesting that they belong to a separated piece of subducted lithosphere. There is a good correspondence between the depth range in which the P-axes plunge closer to the slab dip (-70") and the interval characterized by the highest seismic energy release (190-370 km).

SUMMARY The high resolution of broad-band seismic data is highly effective in studying ruptures associated with deep-focus earthquakes. In this study, we report results from an analysis of P and SH waveforms from eight of the largest deep-focus earthquakes that occurred between 1987 and 1992 along the north-western Pacific. In this region, broad-band data recorded at close-in distances, directly above the source zone, offer an unusual source-receiver geometry that complements observations at teleseismic distances. Among the eight earthquakes studied, six showed clear indications of directivity. The best constraint on the extent of rupture came from the steeply dipping Izu-Bonin Wadati-Benioff zone, where subhorizontal ruptures typically originate in the interior of subducted slab and propagate toward the top at high angles. In particular, rupture of the 1988 September 7 event consists of two en echelon line sources. At a depth of approximately 480 km, the overall rupture length of this earthquake constrains the minimum thickness of the seismogenic zone to be 27 ± 12 km. At this depth, our estimated thickness of the seismogenic zone seems remarkably close to the maximum thickness (˜35 km) predicted by the metastable olivine wedge model of seismogenesis. This model interprets deep-focus earthquakes as a result of transformational faulting within a region bounded approximately by the 700°C isotherm. The extent of rupture during the largest events in this region seems to be only slightly larger than the 20-25 km spacing between the two layers of a double seismic zone, found at depths between 300 and 400 km in the same general area. Beneath the Island of Sakhalin, a rupture speed greater than that of the shear wave is not required for the large, deep-focus earthquake (˜615 km) of 1990 May 12. Almost all complexity in observed waveforms for this event can be explained by a northward propagating, subhorizontal rupture composed of only two subevents.

SUMMARY Analysis of moment tensor inversions using various passbands of very broadband seismic data provides clear evidence that some non-double couple moment tensors of intermediate-depth and deep earthquakes result from the superposition of different double couple sources whose predominant principal axes are aligned with the predominant strain state within the subducting slabs. The analyses are performed for three intermediate-depth and deep earthquakes: 1984 January 1 south of Honshu (386 km), 1985 April 23 Luzon (181 km), and 1987 May 7 northern Sea of Japan (417 km). Consistent non-double couple moment tensors obtained using different sets of seismic waves in various low frequency bands suggest that these three significant non-double couple components are not caused by unmodelled propagation errors of seismic waves in the inversion procedure, because the various sets of seismic waves traverse very different ray paths. For the south of Honshu and the Luzon events, two major arrivals in the broadband P-wave diplacement seismograms are observed, with varying relative amplitudes or polarities from station to station. The different double couple mechanisms which model the two phases combine to produce the significant non-double couple moment tensors obtained at long periods. The principal axes of the subevents, closest in orientation to the predominant strain states within the slabs, tend to be quite stable, whereas the other two principal axes rotate between subevents. This observation may explain the global nature of non-double couple components in relation to the strain regime within the slab. For the northern Sea of Japan event, two major phases are observed in the P-wave displacement waveforms, but the variation in mechanism of two subevents that we model is not enough to explain the large non-double couple component observed at long periods. However, since the non-double couple component is exceptionally large compared with those for other deep earthquakes, changes of focal mechanism with shorter delays than we can resolve appear to explain the overall radiation most simply.

Summary Mt Cameroon is an active basanite composite volcano on the West African coast which has erupted five times this century, the latest in October/November 1982. An interesting aspect of Mt Cameroon region seismicity is the regular occurrence of deep, near-constant magnitude, low-frequency microearthquakes near the volcano. The near-regularity of occurrence of these limited magnitude range microearthquakes in a relatively small area, and the closeness of the region to the equator, where tidal deformation is greatest, suggests that a test for possible earth-tide triggering of these events is a good idea. An examination of these deep events for semidiurnal tidal correlation, however, shows conclusively that no correlation exists between their time of occurrence and the tidal phase.

SUMMARY Computer graphics techniques are used to portray the shape of the subducted lithosphere as seen from different views. The configuration of the subducted lithosphere is represented by a surface that approximately fits the upper envelope of the slab-like spatial distribution of earthquake hypocentres in the Benioff-Wadati zone. Regions considered include Tonga-Kermadec, Izu-Bonin-Mariana, JapanKuril-Kamchatka, and Sunda-Banda in the western Pacific. These regions include all known zones of deep focus earthquakes except the tectonically complex regions of New Hebrides, Solomon-New Britain, Molucca Sea (Phillipines), the Mediterranean, and South America. Shown at the same scale and with the effect of Earth's sphericity included, the views are very useful for illustrating variations in the configuration of the subducted lithosphere between different regions and within the same region. The results illustrate the segmentation of the upper several hundred kilometres of subducted lithosphere into major units of relatively uniform configuration. These segments are also clearly defined by major features of surface morphology of the trench and upper plate,, the distribution of subduction-related volcanism, and by the intersection with the subduction zone of major bathymetric features of the subducting plate. In contrast, the deeper portions of the slabs have a more complex and variable configuration. Discordances between the deep and shallow structures are often important on a regional scale and suggest that there are relative horizontal movements between the surface trace of the subduction zone and the material at depth into which the subducted lithosphere is sinking. These relative movements appear to be intimately associated with the tectonic evolution of the region and indicate a regional aspect of subduction that is often overlooked in an attempt to explain the global variability of the shapes of subducted lithosphere. Although earthquake activity ceases at 680 km, possibly because of a phase transition, the geometry of the deep portion of subducted lithosphere suggests that subducted lithosphere does penetrate to greater depth.

SUMMARY Using more than 300 recordings of deep Tonga earthquakes from the Warramunga Seismic Array (central Australia), we have searched for short-period S-P conversions from the 670 km discontinuity arriving in the P-wave codas. The array seismograms were stacked and plotted in sections to optimize observation of S670P yielding the following results: For the northern part of Tonga (17.3°–18.4°S), S670P-phases are observed with apparent conversion depths between about 660 and 680 km. For the central part (20.2°–22.3°S), S670P is not clearly observed. For the southern part (23.1°–25.7°S), S670P is observed for conversion depths between about 660 and 700 km. Intermittent occurrence of S-P conversion may result from different characteristic source radiation patterns among the three sections. Overall uncertainties in conversion depths (resulting from unmodelled lateral velocity variations and hypocentral location errors) are perhaps a few tens of kilometres. The S-P rays must travel either within or very close to the subducted slab because of the source/receiver/slab-dip geometry, so we conclude that large (>50 km) deformations of the 670 km discontinuity do not occur beneath Tonga. Our observations appear to be inconsistent with a chemical discontinuity at 670 km, but are easily explained if the 670 km discontinuity is a strongly pressure-dependent and, perhaps, moderately endothermic phase change in both the slab and surrounding mantle.

SUMMARY Short-period teleseismic P-waves recorded at Noumea, New Caledonia, from intermediate and deep focus earthquakes in SE Asia and the SW Pacific are used to investigate crustal and upper mantle structure of the Norfolk ridge, which is currently described as a zone of convergence in its caledonian part. Radial components recorded from various source backazimuths show a succession of large P-to-S converted waves whose timing and polarity suggest the presence of a north-dipping low-velocity zone at approximately 60km depth below the station. The inferred strong lateral variations of structure in the upper mantle are also supported by observations of complex horizontal particle motions and large tangential components which exhibit clear P-to-S phases. The high velocity contrast observed across these dipping interfaces is interpreted as a structural remnant of the New Caledonia basin oceanic crust which would have been underthrust beneath the New Caledonia crust. Using geometrical ray theory in dipping structure, a model valid for the Noumea station has been derived. The crust is apparently quite homogeneous. It is 26 ±1 km thick beneath the station and seems to thicken eastward as deduced from the orientation of the Moho. The low average crustal velocities suggest a continental nature for New Caledonia. Toward the thrust zone, the Moho is a thick transition zone whereas it can be modelled by a sharp boundary elsewhere. The inferred structure can be interpreted in terms of either a dead subduction zone or an underthrusting zone. According to the history of compressive deformations observed in land, the underthrusting would likely have been syntectonic of the New Caledonia late Eocene obduction.

Summary. A systematic study of deep seismicity is performed using waveform data from the GDSN between 1977 and 1980 and the Centroid Moment Tensor method of Dziewonski et al. Two hundred new CMT solutions are found for events deeper than 100km; their moments, obtained using the same algorithm, range between 7 x and 1.4 x lo2' dyne cm - a factor of 20 000. A statistical analysis of the CMT parameters provides important conclusions. Among these, the shift in origin time for deep events is found to be almost independent of the seismic moment, providing an indication of the complexity of deep earthquakes. The deviation from the double couple model appears to be correlated to the state of stress of the subducting slabs; mechanisms showing down-dip compression are generally dominated by the eigenvalue corresponding to the compression axis, and down-dip tension correlates with the predominance of the tensional eigenvalue. Comparison with moment tensors obtained by other authors using different waveform data shows good agreement between different methods. Comparison with published focal plane mechanisms confirms the precision and reliability of the CMT method. Examples of CMT solutions used to identify patterns of seismicity associated with deep fault planes are shown.

Summary Anomalous high-frequency arrivals from deep-focus earthquakes have been observed for epicentral distances of 13–21 degrees in Japan, Tonga and South America. Their arrival times are up to a minute later than those of S, their durations are longer than those of the direct phases and they have an apparent phase velocity of 4.7 km s-1. Isacks and Barazangi have proposed a model which fits the data for arrivals of this type observed at NNA in South America. Their model assumes the existence of structure heretofore not predicted by independent observation. Assuming the uniqueness of their model in its ability to fit the data, they have used it as evidence for the existence of these structures. We have found, however, alternative models which fit the data for arrivals of this type observed at ARE, CUS and TRU in addition to those at NNA. According to these models, a P phase travels from the focus to some discontinuity where it is converted to an S phase which travels on to the station. The conversion points predicted by these models lie near a boundary of the inclined seismic zone which, we speculate, defines a dipping lithospheric slab. We conclude that these anomalous arrivals can be interpreted using seismic-ray calculations based on laterally homogeneous velocity-depth models and a dipping interface.

Summary Regional distribution of anelastic attenuation beneath the United States was investigated using amplitudes and dominant periods of short period P and S waves originating from deep focus earthquakes in South America and the Circumpacific seismic belt, and recorded at LRSM (Long Range Seismic Measurement) stations. The observed regional distribution pattern shows high attenuation in the western United States, including California, and a less pronounced higher attenuation region in the northeastern United States. This distribution pattern is similar to that reported by Solomon & Toksoz for long period S waves, but differs from it sufficiently to indicate lateral variations in the frequency dependence of the average crust-upper mantle attenuation across the United States.

SUMMARY Moho underside reflection pMP is a weak precursor to the surface reflection pP. It is best observed in long-period records from deep-focus earthquakes, provided that the P coda has sufficiently low amplitudes. pMP reflection in continental or shelf regions produces a clear arrival, whereas pMP reflection at the oceanic Moho produces a smooth transition from pMP to pP. These properties are illustrated by WWSSN seismogram examples for events below the Sea of Japan, the Sea of Okhotsk, Argentina and Tonga-Fiji. Synthetic seismograms are calculated with a simple reflectivity method, and comparison with observed seismograms, corresponding to continental pMP reflection, gives estimates of crustal thickness and Moho P-velocity change; they are averages over reflection zones with a diameter of about 300 km. Application to the crust of North Korea indicates an increase of crustal thickness from about 27 to 32 km over about 300 km from N to S, accompanied by a decrease in the velocity change at the Moho from about 1.3 to 0.9 km s-'.

An earthquake sequence can be considered to be a complex extension of a shear crack in a viscoelastic medium under the influence of non-uniform stresses. Because of these inhomogeneous stresses, the crack grows rapidly at rates comparable to seismic body-wave velocities during some intervals of time; during other intervals, which punctuate the episodes of rapid expansion, the major mode of slip and growth is one of creep. Under suitable pre-stress and frictional conditions, the post-seismic creep phase of one earthquake may become the pre-seismic creep phase of a succeeding earthquake. Thus inhomogeneity of the pre-stress and/or the static and dynamic frictional stresses combined with viscoelasticity of the medium provides a mechanism that accounts for not only the pre-seismic and the post-seismic creep and the stopping of the crack, but also the various types of earthquake sequences that occur in nature. By increasing the amplitude of the fluctuations in the spatial distribution of these stresses, the type of earthquake sequence can be varied progressively from an ‘isolated earthquake', to a sequence with foreshocks, main shock and aftershocks, and finally, to an earthquake swarm. As the wavelengths of the fluctuations in the stresses decrease, the frequency of earthquake occurrence increases. The type of earthquake sequence is also controlled by the general level of the stresses. A silent earthquake or aseismic creep event will occur if the pre-stress is sufficiently high.

summary. Approximately 1000 PKP observations in the range 110° < Δ < 170° have been statistically studied to examine the existence of Bolt's GH branch of the PKP travel-time curve. Data give firm evidence for the existence of this branch from 130 to 153° but the possibility of its extending further cannot be ruled out. Observations for this branch fit the form: Where a=(1150.3254±0.5328)s, b= (16.4907±0.3803)s, c= (−0.1316±0.0630)s. Times for this branch are always greater by 2 s than the corresponding Gogna times from Pacific earthquakes.

Summary. Analysis of seismograms from 70 deep-focus earthquakes recorded at nearby stations in the Fiji Islands has yielded estimates of the waveforms of P waves radiated from deep sources and of the seismic moments of the events. The earthquake sample consists of all well-recorded events in 1965–66 located in a limited hypocentral region within 4° of the stations and at depths of 500 to 700 km. The spatial distribution restricts the angle of incidence for the incoming direct P waves to less than about 25°, thereby reducing the extent to which the early portion of the observed P wave is contaminated by crustal effects. A study of nuclear explosions recorded by these stations indicates that the effects of the mantle, crust and seismograph on the shape of the P wave are reasonably well-accounted for in the seismogram synthesis employed in this study. The magnitudes (mb) of the events studied ranged from 3.5 to 5.9. The character of the seismograms for these events ranged from small-amplitude, high-frequency, short-duration P waves for the smallest events to large-amplitude and low-frequency or multiple events for some of the largest events. However, 60 of the 70 events, spanning nearly the entire magnitude range, have P waveforms which, during the first 1.2s, are quite similar to one another. Synthetic seismograms were computed for these events, and the best fit is obtained with a triangular farfield displacement pulse, with a 0.6 s rise time and a 0.4 s decay tune. To describe the seismograms of these 60 events, only the pulse height (proportional to seismic moment) need vary; the duration of the triangular pulse form remains nearly constant and implies a spectral ‘corner frequency’ of 0.7 Hz. Thus, although for the entire sample of earthquakes the duration of the source is found to be somewhat dependent on seismic moment, there is a range of two orders of magnitude in moment (1022 to 1024 dyne cm) over which the source duration does not significantly vary. This cannot be accounted for by variations of focal mechanism orientations but indicates large variations in stress drop among the deep events. Published studies of shallow-focus earthquakes in California have indicated that a wide range of seismic moments is associated with a small range of source durations or corner frequencies, particularly for samples of earthquake populations within limited areas. This study shows a similar variation of moment and source duration within a localized region of deep earthquakes. The similarity is in both the range and location on a moment versus source duration plot, indicating similar ranges of stress drops for the shallow California and deep Fiji events.

Smylie & Mansinha (1971) have attempted to utilize static elastic dislocation theory in order to compute the change in the inertia tensor of a realistic spherically symmetric Earth model produced by a point tangential displacement dislocation. The present author has performed similar computations and has combined this theory together with an empirical earthquake moment-magnitude relation in order to estimate the total excitation of the Chandler wobble by all observed large earthquakes since 1904 (Dahlen 1971). The purpose of this note is to point out that there is significant disagreement between the final numerical results of Smylie & Mansinha (1971) and those of Dahlen (1971) and to suggest a possible explanation for this disagreement. In order to investigate the effect of the earthquake focal depth on the shift of the Earth's mean pole of rotation, Smylie & Mansinha (1971) have considered two point tangential displacement dislocation models for the faulting associated with the 1964 Alaskan earthquake. One model is a realistic shallow focus (depth h = lskm), shallow dip angle, thrust faulting mechanism; the other model has the same mechanism, but the focal depth h is taken to be 637 km. Table 1 shows a direct comparison of the results obtained for these two models by Smylie & Mansinha (1971) and the results obtained from my own computations (Dahlen 1971). The fault parameters used in all cases were those considered by Smylie and Mansinha (epicentral colatitude O,, = 29.0°, epicentral longitude 4,, = 213-0°, fault strike N42"E, fault dip 8"W, mechanism pure dip slip thrust, fault slip area 1-54 x 1OI2 m3). These fault parameters for the 1964 Alaskan earthquake differ slightly from those considered by Dahlen (1971). It can be seen that the computations of Smylie & Mansinha (1971) predict that the change in the inertia tensor produced by a deep focus earthquake will be nearly an order of magnitude larger than that produced by a shallow focus earthquake with the same fault parameters; my own computations predict that a deep focus earthquake will have a slightly smaller effect than a similar shallow focus earthquake. The major differences in the theoretical treatments of Smylie & Mansinha (1971) and Dahlen (1971) arise in the method used to deal with the fluid core of the Earth model. The reason why the treatment of a fluid core is not straightforward is that in the absence of both inertia and viscosity, individual fluid particles in a non-rotating fluid core encounter no resistance to a tangential (non-radial) displacement, and because of this the particle displacement field within the fluid core produced by a static tangential displacement dislocation (or any other static force system) located within the solid elastic mantle becomes indeterminate. In discussing this problem I

Summary Distance and time corrections for PcP, PKP and P phases with small angles of emergence are determined. The distance corrections permit extension of a ray path from the observing station through the source to the surface. This results in surface focus distances for the phases which permits more consistent comparison of amplitudes for phases from earthquakes at different depths. Application of the corresponding time corrections results in a surface focus travel-time curve.