We present tomographic evidence for the existence of deep-mantle thermal convection plumes. P-wave velocity images show at least six well-resolved plumes that extend into the lowermost mantle: Ascension, Azores, Canary, Easter, Samoa, and Tahiti. Other less well-resolved plumes, including Hawaii, may also reach the lowermost mantle. We also see several plumes that are mostly confined to the upper mantle, suggesting that convection may be partially separated into two depth regimes. All of the observed plumes have diameters of several hundred kilometers, indicating that plumes convey a substantial fraction of the internal heat escaping from Earth.

Summary We use body wave ray theory in conjunction with the Born approximation tocompute 3‐D Fréchet kernels for finite‐frequency seismic traveltimes, measured by cross‐correlation of a broad‐band waveform with a spherical earth synthetic seismogram. Destructive interference among adjacent frequencies in the broad‐band pulse renders a cross‐correlation traveltime measurement sensitive only to the wave speed in a hollow banana‐shaped region surrounding the unperturbed geometrical ray. The Fréchet kernel expressing this sensitivity is expressed as a double sum over all forward‐propagating body waves from the source and backward‐propagating body waves from the receiver to every single scatterer in the vicinity of this central ray. The kernel for a differential traveltime, measured by cross‐correlation of two phases at the same receiver, is simply the difference of the respective single‐phase kernels. In the paraxial approximation, an absolute or differential traveltime kernel can be computed extremely economically by implementing a single kinematic and dynamic ray trace along each source‐to‐receiver ray.

Geochemistry, Geophysics, Geosystems, v. 7, p. Q11007, 2006. http://dx.doi.org/10.1029/2006GC001248International audienceNew finite-frequency tomographic images of S-wave velocity confirm the existence of deep mantle plumes below a large number of known hot spots. We compare S-anomaly images with an updated P-anomaly model. Deep mantle plumes are present beneath Ascension, Azores, Canary, Cape Verde, Cook Island, Crozet, Easter, Kerguelen, Hawaii, Samoa, and Tahiti. Afar, Atlantic Ridge, Bouvet(Shona), Cocos/Keeling, Louisville, and Reunion are shown to originate at least below the upper mantle if not much deeper. Plumes that reach only to midmantle are present beneath Bowie, Hainan, Eastern Australia, and Juan Fernandez; these plumes may have tails too thin to observe in the lowermost mantle, but the images are also consistent with an interpretation as “dying plumes” that have exhausted their source region. In the tomographic images, only the Eifel and Seychelles plumes are unambiguously confined to the upper mantle. Starting plumes are visible in the lowermost mantle beneath South of Java, East of Solomon, and in the Coral Sea. All imaged plumes are wide and fail to show plumeheads, suggesting a very weakly temperature-dependent viscosity for lower mantle minerals, and/or compositional variations. The S-wave velocity images show several minor differences with respect to the earlier P-wave results, including plume conduits that extend down to the core-mantle boundary beneath Cape Verde, Cook Island, and Kerguelen. A more substantial disagreement between P-wave and S-wave images reopens the question on the depth extent of the Iceland plume. We suggest that a pulsating behavior of the plume may explain the shape of the conduit beneath Iceland

S U M M A R YWe calculate three dimensional (3-D) sensitivity kernels for fundamental-mode surface wave observables based on the single-scattering (Born) approximation. The sensitivity kernels for measured phases, amplitudes and arrival angles are formulated in the framework of surface wave mode summation. We derive kernels for cross-spectral multitaper measurements; as a special case, the results are applicable to single-taper measurements. Cross-branch modecoupling effects are fully accounted for in the kernels; however, these effects can probably be ignored at the present level of spatial resolution in global phase-delay tomography. The narrowly concentrated spectra of the windows and tapers commonly used in global surface wave studies enable the kernels to be computed extremely efficiently.Surface wave tomography based upon great-circle ray theory has been used with great success during the past three decades to constrain the large-scale 3-D heterogeneity of the lithosphere and upper mantle. While the growing abundance of seismic data promotes progress in retrieving better-resolved images with smaller-scale details, ray theory, upon which most surface wave tomography is based, has its theoretical limitations. Ray theory assumes that the frequency of seismic waves is infinite; thereby, it breaks down whenever the length scale of the heterogeneity is comparable to the characteristic wavelength of the seismic waves. Due to their finite frequency, surface waves are sensitive to 3-D structure off of the source-receiver great-circle ray. An approach beyond ray theory aiming at resolving small-scale structures is required to take into account the finite-frequency effects of surface waves. Recent studies have shown a growing appreciation of the finite-frequency properties of seismic body waves (

International audienceEastward subduction of oceanic tectonic plates has shaped the geologic history of western North America over the past 150 million years. The mountain-building and volcanism that brought forth the spectacular landscapes of the West are credited to the vast ancient Farallon plate, which interacted mechanically and chemically with the overlying continent as it plunged back into the mantle. Here, we use finite-frequency travel-time and amplitude measurements of teleseismic P-waves in seven frequency bands to obtain a high-resolution tomographic image to approx1,800 km depth. We discover several large, previously unknown pieces of the plate which show that two distinct stages of whole-mantle subduction are present under North America. The currently active one descends from the Pacific northwest coast to 1,500 km depth beneath the Great Plains, whereas its stalled predecessor occupies the transition zone and lower mantle beneath the eastern half of the continent. We argue that the separation between them is linked to the Laramide era 70–50 Myr ago, a time of unusual volcanism and mountain-building far inland generally explained by an episode of extremely flat subduction

No abstract

[1] We present a three-dimensional, S velocity model of the SE Asian-western Pacific upper mantle with 400-km lateral resolution. Using the novel Automated Multimode Inversion technique, we processed 4038 vertical-component seismograms and extracted 22,708 linear equations with uncorrelated uncertainties that constrain upper mantle structure. We used time-frequency windows to select signal with negligible proportion of scattered energy. The windows included the fundamental Rayleigh mode and S and multiple S waves. The observed range of S velocity variations is the widest (17-18%) in the upper 150 km of the mantle. High-velocity continental roots can reach beyond the present extent of the overlying Archean-Proterozoic crust by 500 km. Beneath some Precambrian units the roots are absent, which can be attributed to deformation and gradual destruction of the ancient lithosphere. At 120-150 km, S velocity beneath some cratons reaches 4.8 km/s; this can be accounted for by thermal and compositional effects. Beneath the Hainan Island area a low-velocity anomaly is observed from near the surface to the bottom of our model; the hot spot-type volcanism here may be caused by the deepmantle Hainan plume. A low-velocity mantle domain underlies the south central Sea of Japan, surrounded on the surface by intraplate volcanoes. A deep-seismicity gap is present near 40°N in the Pacific slab subducting below and may result from a plume-slab interaction. A high-velocity anomaly is present in the transition zone beneath the northern boundary of the Yangtze Craton. We propose that the anomaly corresponds to subducted continental lithosphere, stagnant atop the 660-km discontinuity.

Summary We use a coupled surface wave version of the Born approximation to compute the3‐D sensitivity kernel KT(r) of a seismic body wave traveltime T measured by cross‐correlation of a broad‐band waveform with a spherical earth synthetic seismogram. The geometry of a teleseismic S wave kernel is, at first sight, extremely paradoxical: the sensitivity is zero everywhere along the geometrical ray! The shape of the kernel resembles that of a hollow banana; in a cross‐section perpendicular to the ray, the shape resembles a doughnut. The cross‐path extent of such a banana–doughnut kernel depends upon the frequency content of the wave. The kernel for a very high‐frequency wave is a very skinny hollow banana; wave‐speed heterogeneity wider than this banana affects the traveltime, in accordance with ray theory. We also use the Born approximation to compute the sensitivity kernel KΔT(r) of a differential traveltime ΔT measured by cross‐correlation of two phases, such as SS and S, at the same receiver. The geometries of both an absolute SS wave kernel and a differential SS–S kernel are extremely complicated, particularly in the vicinity of the surface reflection point and the source‐to‐receiver and receiver‐to‐source caustics, because of the minimax character of the SS wave. Heterogeneity in the vicinity of the source and receiver exerts a negligible influence upon an SS–S differential traveltime ΔT only if it is smooth.