Fundamental-mode Group Velocities Research Articles

Shear wave crustal velocity structure in the Garhwal-Kumaon Himalaya based on noise cross-correlation of Rayleigh wave

The tectonics of the Garwal-Kumaon Himalaya is characterized by thrusts, tectonic windows, and klippen. We investigate crustal shear wave velocity variations using ambient noise cross-correlation tomography. The studied region encompasses the Kali River valley in the east to Satluj valley in the west, with the adjoining Indo-Gangetic Plain in the south covering the northern part of the Delhi-Haridwar ridge. The fundamental mode group velocities of Rayleigh waves are extracted from cross-correlation data of 33 broadband seismological stations from a regional seismic network. A total of 374 dispersion curves with a period range of 4–29 s show a group velocity variation between ∼2.3 and ∼ 3.4 km/s. The shear wave velocity structure of the uppermost lithosphere down to ∼50 km obtained by non-linear inversion of the Rayleigh wave dispersion data provides new insight into the geometry of the crust. A large variation in Vs in the range of ∼2.8 to ∼4.7 km/s corresponds to a variety of changes in the tectonic deformation, structure, and crustal thickness of the Himalayan wedge. Thick low-velocity sedimentary formations are identified beneath the Indo-Gangetic Plain and the frontal Himalaya. Anomalous low Vs zones are also observed in the mid-crust beneath the higher Himalaya and southern Tibet, indicating partial melting or the presence of aqueous fluid zones. The high-velocity anomalies may be correlated with duplex structures beneath the Lesser Himalaya and with lithospheric flexure.

One-dimensional regional shear velocity structure from joint inversion of fundamental mode group velocity dispersion measurements of Love and Rayleigh waves: application to the Uttarakhand Himalaya

One-dimensional regional shear velocity structure from joint inversion of fundamental mode group velocity dispersion measurements of Love and Rayleigh waves: application to the Uttarakhand Himalaya

Controls on surface wave overtone interference

SUMMARYMeasurements of fundamental-mode (FM) surface waves along the minor arc are impacted by overtone interference. This interference is primarily due to major-arc overtones for Rayleigh waves and minor-arc overtones for Love waves. In both cases, interference contaminates measurements of phase and amplitude and can introduce bias in seismic images. Here, we use synthetic seismograms computed via normal mode summation to probe how interference can vary as a function of the surface wave group velocities, source mechanism and depth, which control the relative excitation of the FM and overtones, and period. By comparing seismograms that include all overtones to those that include only the FM, we can quantify the interference, i.e., how the presence of the overtones perturbs the FM phase and amplitude. We compare the strength of this interference to calculations of excitation of the overtones and FM. We show that these calculations explain well the varying strength of interference for different source mechanisms and depths. Notably, these calculations illuminate source depths where Love wave overtone excitation is quite low and therefore interference is unusually weak, and depths where Rayleigh wave FM excitation is low and therefore interference is unusually strong. Our analysis also reinforces the dependence of the interference on the FM and overtone group velocities. For Love waves, this results in weak minor-arc overtone interference at long periods and, for continental paths, short periods. For Rayleigh waves, the differing overtone and FM group velocities and the relative excitation of the overtones and FM explain rapid variations in the strength of Rayleigh wave major-arc overtone interference as a function of epicentral distance. We then show that real data are affected by the relative excitation of the FM and overtones. We find that errors in Rayleigh wave phase velocities determined at the EarthScope USArray stations are larger when the ratios of overtone to FM excitation are larger. We also find a dependence of phase velocity error on excitation ratio for Love waves, and we identify the presence of major-arc overtone interference in Love wave measurements. Our results highlight opportunities for more nuanced quality control of surface wave measurements. The relative excitation ratio of the overtones and FM may be a better criterion for event selection than source depth, allowing deeper events that well excite FMs to be included and shallower events with large overtone excitation to be excluded. This would allow the collection of more accurate measurements that will increase the precision of seismic images.

Variations in crustal and lithospheric structure across the Eastern Indian Shield from passive seismic source imaging: Implications to changes in the tectonic regimes and crustal accretion through the Precambrian

Changes in the dominant tectonic regimes and the modes of formation and maturation of the continental crust during the Precambrian are a subject of active debate. Here, we discuss these issues in terms of new information on the crustal and lithospheric architecture of the different terranes in the Eastern part of the Indian Shield (EIS). We present well-constrained estimates of the crustal and lithospheric thickness characterizing different terranes of the EIS through the joint inversion of P-receiver functions (PRF) and fundamental mode group velocity dispersion data of Rayleigh waves (10–100 s). The lithospheric thicknesses are also computed independently using the S-receiver function (SRF) modeling. The study involved 2167 PRFs and 100 SRFs from 15 broadband stations. Our modeling detects a marked crustal (~4–8 km) and lithospheric (~15–20 km) thinning beneath the Singhbhum-Odisha-Craton (SOC), spatially correlating with the central part of the craton comprising the gneissic terrain contrasted by a crustal thickening (~2–6 km) beneath the horseshoe-shaped Iron Ore Group. We propose that such a crustal configuration beneath the Paleoarchean craton may suggest the dominance of vertical tectonic processes such as thickening of an oceanic mafic plateau and recurrent melting producing the TTGs and the Singbhum Granite suite. Infra-crustal reworking involving Rayleigh-Taylor instabilities and gravitational processes could explain the crustal architecture of the SOC much similar to a few other Paleoarchean cratons such as Pilbara (Australia), Barberton and Kappvaal (South Africa). Our modeling has also revealed a relatively smaller degree of crustal (2–4 km) and lithospheric thinning (4–10 km) beneath the Eastern Ghat Mobile Belt, south of the SOC. We observe a relatively thicker crust (~42 km) characterized by a flat Moho over a large area (~40,000 sq. km) encompassing the Chota Nagpur Granite Gneiss Terrane (CGGT). Such thickening is consistent with the convergence and collision tectonics involving the SOC and CGGT. However, we note a 15–20 km lithospheric thinning associated with the CGGT. Such thinning may mainly relate to Phanerozoic rifting along the Damodar graben and Rajmahal magmatism. Broadly, our model of the crustal and lithospheric architecture of the EIS reflects the imprints of northward subduction of SOC beneath the CGGT.

Use of peaks and troughs in the horizontal-to-vertical spectral ratio of ambient noise for Rayleigh-wave dispersion curve picking

To assist in the identification of fundamental-mode dispersion curves of Rayleigh waves in dispersion diagrams, we explore the relation between the shape of the horizontal-to-vertical spectral-ratio (HVSR) of ambient seismic noise and the shape of the dispersion curves for phase and group velocities in a stratified medium. We propose to use the information coming from the HVSR to identify the osculation zones and multi-mode effects and to locate inflection points and critical points in the observed phase and group dispersion diagrams of Rayleigh waves. The relationship between these curves has been numerically investigated for some models consisting of one and two homogeneous layers overlying a half-space, with velocities increasing downwards. It is primarily found that the first minimum in the HVSR appears close to the frequency of the inflection point of the fundamental mode of phase velocity. In addition, the osculation and multimode effects occur between frequencies of the fundamental peak and the first minimum of the HVSR. On the other hand, the frequencies of the minima in HVSR closely approximate the critical points of the fundamental-mode group-velocity dispersion curve, even better than the inflection points of the fundamental-mode phase-velocity curve. Finally, we show an example of experimental identification of fundamental-mode phase and group dispersion curves supported by the shape of the HVSR, obtaining a reliable velocity profile through the simultaneous inversion of these three curves.

Passive source seismic imaging of the crust and upper mantle underlying the Archean Singhbhum Craton, Eastern India

An improved image of crust-mantle structure has been obtained through the joint inversion of P-to-s receiver functions and fundamental mode group velocity data (7–70 s) of Rayleigh and Love waves at fifteen broadband stations in the Eastern Indian Craton. Our modelling reveals a marked crustal thinning of 5–10 km below the Singhbhum-Odisha-Craton (SOC) while the modelled Moho depths show a flat crust with a thickness of 42 km below the Chotanagpur Granitic Gneissic Terrain (CGGT). The modelled lithosphere is thin (∼90 km) beneath the SOC and deepens on both north and south sides to ∼100 km with a sharp drop of ∼4.4% in shear velocity across the lithosphere-asthenosphere boundary. The amount of melt required in the asthenosphere to account for this drop in shear velocity is ∼1%. The isotopic constituents of the Dalma volcanic rocks contains asthenospheric component. We suggest that the modeled crustal and lithospheric thinning might have been resulted from the earlier magmatism or plume episodes that led to lithospheric delamination below the SOC. Our modeling also detects a common elongated feature (extending between latitude 21–23°N and longitude 85–86.4°E), which is characterized by a marked crustal (5–10 km) as well as lithospheric (10–15 km) thinning and a significant drop in Vs (−2.1 to −4.4%) in the upper mantle, below the SOC. This mapped area coincides with the horse-shoe shaped broad Singhbhum synclinorium known as Iron Ore Group (IOG) basin that hosts huge iron ore and to some extent manganese deposits of Precambrian age. This lithospheric feature is inferred to have played a key role in forming the passage for melts to carry various iron rich minerals and metals to the uppermost oceanic crust during the Archean plume associated with magmatism episode.

A possible origin of intraplate earthquakes in the Kachchh rift zone, India, since the 2001 Mw7.7 Bhuj earthquake

We herein use the joint inversion of P-receiver functions and fundamental mode group velocity dispersion data (9–70 s) of Rayleigh and Love waves to estimate crustal and lithospheric thicknesses at twenty three-component mobile broadband stations in Kachchh, Gujarat. Modeled Moho depths range from 35 to 43 km, while lithospheric thicknesses vary from 64 to 106 km. The main result of our modelling is the delineation of a marked crustal (∼2–4 km) as well as lithospheric (∼10–20 km) thinning and a 2–6% drop in Vs across the lithosphere-asthenosphere-boundary (LAB), within the Samkhiali graben (associated with gravity high) underlying the central Kachchh rift zone (KRZ), where 95% of the continued aftershock activity took place since 2001. Such a large drop in Vs could be attributed to the presence of carbonatite melts in the upper mantle. Our modeling reveals a 4 km crustal thinning below the central KRZ and a 4 km crustal thickening below the surrounding riftless regions. This kind of crustal structure is inferred to induce large flexural deviatoric stresses (∼50–100 MPa) in the upper crust, thereby, these stresses in the presence of regional plate tectonic stresses can bring the Samkhiali graben below the central KRZ near to the critical stress level. While stress-transfer, meteoric water (in the upper crust), and metamorphic fluid as well volatile CO2 flows (in the lower crust) provide the required triggering effect to the critically stressed graben structure (down to 35 km depth) for generating continued aftershock activity in the Kachchh rift zone, since 2001. We also propose that the deeper circulation of volatile CO2 through the inferred conduit (related to the 65 Ma Deccan Plume activity) extending from lower crust down to asthenosphere plays a key role in the generation of uninterrupted occurrence of earthquakes in the Kachchh rift zone, Gujarat, India.

S-wave velocity model for several regions of the Kamchatka Peninsula from the cross correlations of ambient seismic noise

The data from the seismic networks of the Kamchatka Branch of the Geophysical Survey of the Russian Academy of Sciences are used for calculating the cross correlations of seismic noise for the stationary digital stations over 2013 and for radio telemetric stations (RTS) in the region of the Klyuchevskoy volcano over the period from January 1, 2009 to May 31, 2013. Four hundred and two correlations overall are calculated. The fundamental-mode group velocities of the Rayleigh waves are calculated in the periods ranging from 5 to 50 s. The calculations for the region of the Klyuchevskaya group of volcanoes are based on the RTS data and cover the periods from 2 to 8 s. The two-dimensional (2D) maps of group velocity distributions in different periods are constructed with the use of the algorithm of surface wave tomography (Barmin, 2001). The velocity sections for the selected Kamchatka regions are reconstructed by the dispersion curve inversion technique (Mordret, 2014). For each region, the structure of the Earth’s crust and upper mantle down to a depth of 50 km was obtained.

Single-station and single-event marsquake location and inversion for structure using synthetic Martian waveforms

In anticipation of the upcoming InSight mission, which is expected to deploy a single seismic station on the Martian surface in November 2018, we describe a methodology that enables locating marsquakes and obtaining information on the interior structure of Mars. The method works sequentially and is illustrated using single representative 3-component seismograms from two separate events: a relatively large teleseismic event (Mw5.1) and a small-to-moderate-sized regional event (Mw3.8). Location and origin time of the event is determined probabilistically from observations of Rayleigh waves and body-wave arrivals. From the recording of surface waves, averaged fundamental-mode group velocity dispersion data can be extracted and, in combination with body-wave arrival picks, inverted for crust and mantle structure. In the absence of Martian seismic data, we performed full waveform computations using a spectral element method (AxiSEM) to compute seismograms down to a period of 1s. The model (radial profiles of density, P- and S-wave-speed, and attenuation) used for this purpose is constructed on the basis of an average Martian mantle composition and model areotherm using thermodynamic principles, mineral physics data, and viscoelastic modeling. Noise was added to the synthetic seismic data using an up-to-date noise model that considers a whole series of possible noise sources generated in instrument and lander, including wind-, thermal-, and pressure-induced effects and electromagnetic noise. The examples studied here, which are based on the assumption of spherical symmetry, show that we are able to determine epicentral distance and origin time to accuracies of ∼0.5–1° and ±3–6s, respectively. For the events and the particular noise level chosen, information on Rayleigh-wave group velocity dispersion in the period range ∼14–48s (Mw5.1) and ∼14–34s (Mw3.8) could be determined. Stochastic inversion of dispersion data in combination with body-wave travel time information for interior structure, allows us to constrain mantle velocity structure to an uncertainty of ∼5%. Employing the travel times obtained with the initially inverted models, we are able to locate additional body-wave arrivals including depth phases, surface and Moho (multiple) reflections that may otherwise elude visual identification. This expanded data set is reinverted to refine interior structure models and source parameters (epicentral distance and origin time).

Rayleigh-Wave, Group-Velocity Tomography of the Borborema Province, NE Brazil, from Ambient Seismic Noise

Ambient seismic noise has traditionally been regarded as an unwanted perturbation that “contaminates” earthquake data. Over the last decade, however, it has been shown that consistent information about subsurface structure can be extracted from ambient seismic noise. By cross-correlation of noise simultaneously recorded at two seismic stations, the empirical Green’s function for the propagating medium between them can be reconstructed. Moreover, for periods less than 30 s the seismic spectrum of ambient noise is dominated by microseismic energy and, because microseismic energy travels mostly as surface-waves, the reconstruction of the empirical Green’s function is usually proportional to the surface-wave portion of the seismic wavefield. In this paper, we present 333 empirical Green’s functions obtained from stacked cross-correlations of one month of vertical component ambient seismic noise for different pairs of seismic stations in the Borborema Province of NE Brazil. The empirical Green’s functions show that the signal obtained is dominated by Rayleigh waves and that dispersion velocities can be measured reliably for periods between 5 and 20 s. The study includes permanent stations from a monitoring seismic network and temporary stations from past passive experiments in the region, resulting in a combined network of 34 stations separated by distances between approximately 40 and 1,287 km. Fundamental-mode group velocities were obtained for all station pairs and then tomographically inverted to produce maps of group velocity variation. For short periods (5–10 s) the tomographic maps correlate well with surface geology, with slow velocities delineating the main rift basins (Potiguar, Tucano, and Reconcavo) and fast velocities delineating the location of the Precambrian Sao Francisco craton and the Rio Grande do Norte domain. For longer periods (15–20 s) most of the velocity anomalies fade away, and only those associated with the deep Tucano basin and the Sao Francisco craton remain. The fading of the Rio Grande do Norte domain fast-velocity anomaly suggests this is a supracrustal structure rather than a lithospheric terrain, and places new constraints on the Precambrian evolution of the Borborema Province.

Crustal velocity structure of the Deccan Volcanic Province, Indian Peninsula, from observed surface wave dispersion

<p>Through inversion of fundamental mode group velocities of Love and Rayleigh waves, we study the crustal and subcrustal structure across the central Deccan Volcanic Province (DVP), which is one of the world’s largest terrestrial flood basalts. Our analysis is based on broadband seismograms recorded at seismological station Bhopal (BHPL) in the central India from earthquakes located near west coast of India, with an average epicentral distance about 768 km. The recording station and epicentral zone are situated respectively on the northern and southern edges of DVP with wave paths across central DVP. The period of group velocity data ranges from 5 to 60 s for Rayleigh waves and 5 to 45 s for Love waves. Using the genetic algorithm, the observed data have been inverted to obtain the crust and subcrustal velocity structure along the wavepaths. Using this procedure, a similar velocity structure was also obtained earlier for the northwestern DVP, which is in the west of the present study region. Comparison of results show that the crustal thickness decreases westward from central DVP (39.6 km) to northwestern DVP (37.8 km) along with the decrease of thickness of upper crust; while the thickness of lower crust remains nearly same. From east to west S-wave velocity in the upper crust decreases by 2 to 3 per cent, while P-wave velocity in the whole crust and subcrust decreases by 3 to 6 per cent. The P- and S-wave velocities are positively correlated with crustal thickness and negatively correlated with earth’s heat flow. It appears that the elevated crustal and subcrustal temperature in the western side is the main factor for low velocities on this side.</p>

Upper mantle structure of South America from joint inversion of waveforms and fundamental mode group velocities of Rayleigh waves

A new tomographic S wave velocity model for the upper mantle beneath South America is presented. We developed and applied a new method of simultaneously inverting regional S and Rayleigh waveforms and fundamental mode Rayleigh wave group velocities, to better constrain upper mantle S velocity structure and Moho depth. We used ∼5700 Rayleigh wave group velocity dispersion curves and 1537 regional wave trains with paths principally passing through the South American continent. The joint inversion of this data set provided a new three‐dimensional (3‐D) upper mantle S velocity model and a Moho depth model for South America, which fits both the group velocity and regional waveform data sets well. New features of the final three‐dimensional (3‐D) S velocity and Moho depth model correlate well with known geotectonic units on a regional scale. The Moho depth ranges from 30 km in the central Chaco basin to 42 km beneath the Amazonian craton and 45–70 km beneath the orogenic Andean belt. The imaged S velocity indicates an average lithosphere thickness of around 160 km for the Amazonian craton. High velocities are imaged beneath the Amazon and part of the Paraná and Parnaíba basins down to about 150 km. Low to very low velocities are imaged beneath the central Andes and the Chaco, Pantanal, and northwestern Paraná basins.

Test of the upper mantle low velocity layer in Siberia with surface waves

The existence of the upper mantle low velocity layer (LVL) below 100 km depth in cratonic areas is tested with surface waves dispersion curves. Given the ambient noise we find that a pronounced LVL (80 km thick and 2% velocity reduction or 40 km thick and 5% velocity reduction) can be distinguished from a constant velocity model by comparison of the fundamental mode group velocities, whereas a thin LVL (less than 40 km thick) with small velocity contrast (less than 2%) cannot be resolved. The fundamental modes of Love and Rayleigh waves have similar properties and, in general, the phase velocity differences are smaller than the standard error. Phase velocity alone cannot discriminate between the models, and the group velocity is in general more sensitive to the velocity structure than the phase velocity. The higher modes at short periods could potentially determine a LVL but in reality it is difficult to obtain sufficiently accurate measurements. We invert the synthetic dispersion curves by the non-linear Hedgehog inversion method. A pronounced LVL (more than 40 km thick and with a strong velocity contrast of about 5%) is detectable by the non-linear inversion but for a thin LVL with a strong velocity contrast it is not possible to resolve both velocity and thickness. In the inversions all solutions include a LVL for models with a pronounced LVL, whereas the solution space includes models with and without a LVL for models with a zero or positive gradient velocity–depth structure. We invert also real data with travel path across the Siberian craton with the Hedgehog method. Almost all solutions include a LVL in the depth range of 80–150 km with a velocity contrast up to 2% to the surrounding intervals. Hence, the LVL appears to be a common feature of the Siberian upper mantle, although a constant velocity at the same depth range cannot be totally excluded. Despite low resolution at large depth, a pronounced asthenospheric LVL below a depth of about 225 km is a constant characteristic of the set of solutions.

Group-velocity tomography and lithospheric S-velocity structure of the South American continent

The lithosphere of the South American continent has been studied little, especially in northern Brazil (the Amazonian region). A 3D lithospheric S-velocity model of South America was obtained by first carrying out Rayleigh and Love wave group-velocity tomography, and then inverting the regionalized dispersion curves. Fundamental mode group velocities were measured using a Multiple Filtering Technique. More than 12,000 paths were examined and about 6000 Rayleigh- and 3500 Love-wave dispersion curves with good quality were retrieved. Checkerboard tests showed that our dataset permits the resolution of features 400–800km across laterally in the central part of the continent from crustal to upper mantle depths. Our results confirm previous tomographic results and correlate well with the major geological provinces of South America. The 3D S-velocity model confirms both regional features of SE Brazil from P-wave travel-time tomography and continental-scale features of central and western South America from waveform inversion, e.g., lowest velocities in the Andean upper mantle; three parts of the Nazca plate with flat subduction; strong low-velocity anomalies in the upper-mantle depth beneath the Chaco basin. Furthermore, our 3D model revealed new features in the South American continent: (1) high velocities in the lower crust were consistently found in regions with high Bouguer or free-air anomalies; (2) the NE–SW trending TransBrasiliano shear zone was delineated by a NE–SW low-velocity belt at lithospheric depths; (3) the eastern Amazonian craton appears to have thicker lithosphere than the western craton; (4) in areas of Archean nuclei located in the northeastern Guaporé shield and southeastern São Francisco craton, high velocity anomalies were found down to 150km.

Lithospheric structure of the Arabian Shield from the joint inversion of receiver functions and surface-wave group velocities

We estimate lithospheric velocity structure for the Arabian Shield by jointly modeling receiver functions and fundamental-mode group velocities from events recorded by the 1995–1997 Saudi Arabian Portable Broadband Deployment. Receiver functions are primarily sensitive to shear-wave velocity contrasts and vertical travel times, and surface-wave dispersion measurements are sensitive to vertical shear-wave velocity averages, so that their combination bridge resolution gaps associated with each individual data set. Our resulting models correlate well with the observed surface geology; the Asir terrane to the West consists of a 10-km-thick upper crust of 3.3 km/s overlying a lower crust of 3.7–3.8 km/s; in the Afif terrane to the East, the upper crust is 20 km thick and has an average velocity of 3.6 km/s, and the lower crust is about 3.8 km/s; separating the terranes, the Nabitah mobile belt is made of a gradational upper crust up to 3.6 km/s at 15 km overlying an also gradational lower crust up to 4.0 km/s. The crust–mantle transition is found to be sharp in terranes of continental affinity (east) and gradual in terranes of oceanic affinity (west). The upper mantle shear velocities range from 4.3 to 4.6 km/s. Temperatures around 1000 °C are obtained from our velocity models for a thin upper mantle lid observed beneath station TAIF, and suggest that the lithosphere could be as thin as 50–60 km under this station.

EurID: a regionalized 3-D seismological model of Europe

This paper presents a 3-D regionalized model of the European crust and mantle velocity structure down to depths of approximately 300 km. The model is entirely based upon the existing geological and geophysical literature and it incorporates topographic and bathymetric features. The resulting model, which we call EurID (European I-Data set), consists of approximately 6400 1-D vertical structures having an average separation of about 100 km. EurID is compared quantitatively with EUR-S91, a recent S-wave velocity tomographic model obtained from body and surface wave waveform inversion [Zielhuis, A., Nolet, G., 1994. Shear-wave velocity variations in the upper mantle beneath central Europe, Geophys. J. Int., 117: 695–715]. It is found that the S-wave velocities in EurID and in EUR-S91 (Zielhuis and Nolet, 1994) do not differ by more than about 4% over most of the area. To validate EurID, surface wave dispersion analysis is performed along some selected paths. The fundamental mode group–velocity curves obtained using 1-D average models derived from EurID are found in good agreement with those determined from observations. Broadband waveform synthetics are also calculated along the same paths and compared with the observed time series. Overall, we believe that EurID is representative of our current knowledge of the crust/upper mantle structure beneath Europe and it can be used to model realistically ground motion for periods above 10 s within structurally homogeneous regions. The first digital version of this structure database is available for public use.

Resolving a low-velocity zone with surface-wave data

SUMMARY The main purpose of seismic inversion is the retrieval of seismic velocities and densities in the earth. Inversion of body-wave traveltimes cannot uniquely determine the seismic velocities as a function of depth when a low-velocity layer is present. It is generally assumed that surface waves do not suffer from the same non-uniqueness. The issue is addressed whether it is possible to remove the non-uniqueness of traveltime inversions with a realistic set of surface-wave data. This requires the exact determination of the velocity distribution within and around the low-velocity zone. Waveforms, phase velocities as well as group velocities are investigated qualitatively. Synthetic Love-wave phase and group velocities are actually inverted. Waveforms are shown to be sensitive to the exact velocity distribution of a low-velocity layer, but it is concluded that the removal of the non-uniqueness with the use of waveforms is difficult because the differences of the waveforms are generally small and because complications such as lateral heterogeneity and poorly known source parameters reduce the accuracy of waveform inversions. The inversions of Love phase and group velocities indicate that it is difficult to determine the velocity distribution of a low-velocity layer in a statistically significant way with a realistic set of dispersion data. Group velocities are shown to be more sensitive to the low-velocity structure than phase velocities. Unfortunately, group-velocity data suffers from a practical non-uniqueness because in general only fundamental-mode group velocities can be measured. It is concluded that the non-uniqueness in the detailed structure caused by a low-velocity layer cannot readily be resolved by using surface-wave dispersion data.

Rayleigh wave anelastic attenuation across the Bermuda Rise and the volcanic Caribbean Arch

Summary. Average fundamental mode group velocities and attenuation coefficients of Rayleigh waves have been obtained in a region of the Western North Atlantic comprising the Bermuda Rise, the volcanic Caribbean Arch, and normal seafloor greater than 65Myr in age. The average Rayleigh wave attenuation coefficients decrease from a value of 7.78~ 10-4km-' at a period of 15s to a value of 0.21 ~lO-~km-' at a period of 100s. Attenuation values and QR factors obtained are characteristic of a young oceanic region, and they do not correspond to a typical region greater than 65 Myr in age. It is apparent from the data that the high attenuation values reported are due to the effect that the back-arc basin behind the Caribbean Arch has on Rayleigh wave attenuation. Inversion theory applied to the attenuation data yields a Qj' model presenting the same features as those corresponding to extension zones such as the spreading centres of the Atlantic and Pacific Oceans. The low-Q region in the model is situated between depths of around 15 and 100 km, with the maximum Qp' value (- 17 x occurring at about 50 km depth.

Surface wave dispersion and Earth structure in south-eastern China

Summary. A reconnaissance study of crust and mantle structure in southeastern China was made using surface waves confined to that region from recent earthquakes. Data from the WWSSN stations ANP arid SEO, and from the digital stations TATO and MAT, were used to measure fundamental-mode group velocities of Love and Rayleigh waves over nine paths in south-eastern China, an area which has been technically quiet since the early Cenozoic. Crustal structure in this region is typical of stable continents, but shear-wave velocities in the uppermost mantle are low for a continent, 4.45 km s−1 or less. Other seismological data support this observation.