Toward the reconciliation of seismological and petrological perspectives on oceanic lithosphere heterogeneity

Abstract

The character of the high‐frequency seismic phases Po and So, observed after propagation for long distances in the oceanic lithosphere, requires the presence of scattering from complex structure in 3‐D. Current models use stochastic representations of seismic structure in the oceanic lithosphere. The observations are compatible with quasi‐laminate features with horizontal correlation length around 10 km and vertical correlation length 0.5 km, with a uniform level of about 2% variation through the full thickness of the lithosphere. Such structures are difficult to explain with petrological models, which would favor stronger heterogeneity at the base of the lithosphere associated with underplating from frozen melts. Petrological evidence mostly points to smaller‐scale features than suggested by seismology. The models from the different fields have been derived independently, with various levels of simplification. Fortunately, it is possible to gently modify the seismological model toward stronger basal heterogeneity, but there remains a need for some quasi‐laminate structure throughout the mantle component of the oceanic lithosphere. The new models help to bridge the gulf between the different viewpoints, but ambiguities remain.

Thermal structure, radial anisotropy, and dynamics of oceanic boundary layers

“…Kennett and Furumura [] show that pervasive stochastic heterogeneities with horizontal and vertical correlation lengths of ∼10 km and ∼0.5 km, respectively, provide a good explanation for their P o/ S o observations. Since such quasi‐laminated petrological fabrics would manifest themselves in the form of radial anisotropy, when seen by longer‐period surface waves, Kennett and Furumura [] suggest the level of heterogeneity as a potential ξ proxy.…”

Pn Tomography of the Juan de Fuca and Gorda Plates: Implications for Mantle Deformation and Hydration in the Oceanic Lithosphere

“…The Pn phase is distinguishable by its high frequency content (3–30 Hz), typically emergent onset, long coda, and a move‐out velocity around 8 km/s. Scattering, possibly from elongate lithospheric heterogeneities, in combination with sediment and water column reverberations generates an emergent waveform with a coda that decays slowly over many tens of seconds (Kennett & Furumura, , ; Sereno & Orcutt, ). Scattering near the source region may obscure the polarity of the Pn onset adding further complexity to the waveform.…”

Enhanced waveguide effect for deep‐focus earthquakes in the subducting Pacific slab produced by a metastable olivine wedge

“…The model of small‐scale heterogeneity in the subducting slab also builds on Furumura and Kennett [], using a von Karmán distribution function with horizontal correlation length of a h = 10 km along the plate dip direction and much shorter correlation distance of a z = 0.5 km across the plate thickness. Based on the work of Kennett and Furumura [], we introduce a sinusoidally increasing strength of heterogeneities with standard deviation ε ranging from 0.5% at the top to 2.5% at the bottom of the slab, superimposed on the background P and S wave velocities and densities of the oceanic mantle (80% of the standard deviation for P and S wave speeds). The heterogeneity model in the continental and oceanic crust are modeled by a h = 0.25 km and a z = 0.5 km with ε = 2%, and much larger scale but weaker heterogeneity ( a h = 20 km, a z = 5 km, ε = 1%) is introduced in the mantle.…”