Joint inversion for three‐dimensional S velocity mantle structure along the Tethyan margin

Abstract

1] We construct a new three-dimensional S velocity model and Moho map by jointly inverting regional S and Rayleigh waveform fits, teleseismic S and SKS arrival times, fundamental mode Rayleigh wave group velocities, and independent Moho depth estimates for the region that extends from the mid-Atlantic ridge through northern Africa, southern Europe, and western Asia. The joint inversion benefits from both better resolution and wider data coverage than when using only individual data sets. Resolution tests confirm that the joint inversion yields good resolution ranging from the Moho to a depth of 1400 km. The complementary and overlapping nature of the different data sets' resolving power has reduced disparities in resolving power that exist for individual data sets, for example between resolving power for crustal and lower-mantle structure. This increases the utility of the new tomographic model for explaining and predicting a variety of observations and dynamics. The new model derived from the joint inversion assembles a large number of mantle structures known from a wide variety of previous studies into one model and in some cases reconciles different local studies that previously seemed contradictory. Finally, the model shows that shallow low-velocity anomalies beneath the Pannonian basin and the Iranian plateau are connected to similar anomalies in the transition zone, the latter possibly related to a deep dehydration process in the subducted lithosphere of the Neo-Tethys Ocean. The model shows the Hellenic slab penetrating the lower mantle, the Calabrian slab extending flatly in the transition zone, and discontinuous slabs beneath the Apennines and the Zagros belt. (2010), Joint inversion for three-dimensional S velocity mantle structure along the Tethyan margin,

Crust and upper mantle of the western Mediterranean – Constraints from full-waveform inversion

“…Spakman et al, 1993;Villaseñor et al, 2003;Piromallo and Morelli, 2003;Koulakov et al, 2009;Bezada et al, 2013;Bonnin et al, 2014), or a combination of both (e.g. Chang et al, 2010;Fichtner et al, 2013b;Zhu et al, 2015). However, most of these models have insufficient resolution in the western Mediterranean and northern Africa, either due to a lack of data or as a result of methodological restrictions.…”

“…Lailly, 1983;Tarantola, 1984), full-waveform inversion has only recently developed into a viable tool for the assimilation of seismic waveform data into 3-D Earth models (e.g. Chen et al, 2007;Tape et al, 2010;Rickers et al, 2013). Using complete seismograms instead of few selected phases generally improves tomographic resolution.…”

Receiver function images of the Hellenic subduction zone and comparison to microseismicity

“…At greater depths, global and regional P-wave tomographies show that the Hellenic subduction zone coincides with high seismic velocity anomalies that extend through the upper mantle into the lower mantle (e.g. Spakman et al, 1988Spakman et al, , 1993Papazachos and Nolet, 1997;Bijwaard et al, 1998;Piromallo and Morelli, 2003;Schmid et al, 2004;Chang et al, 2010;Biryol et al, 2011). Furthermore, they imaged a not completely amphitheatrical geometry for the subducted African slab and showed a gradual increase in the slope of the Hellenic slab from west to east and toward the larger depths of the slab.…”

“…In general, seismic tomography provided relatively unsharp images of the Hellenic subduction zone (e.g. Spakman et al, 1988Spakman et al, , 1993Papazachos and Nolet, 1997;Bijwaard et al, 1998;Piromallo and Morelli, 2003;Schmid et al, 2004;Chang et al, 2010;Biryol et al, 2011). Studies of converted or scattered waves in the Aegean were mainly obtained from small networks mostly covering the western part of the Hellenic arc (e.g.…”

Foundations for a multiscale collaborative Earth model

“…Giardini et al 1987;Resovsky & Ritzwoller 1999;Ishii & Tromp 2001, or combinations of these (e.g. Nolet 1978;Ritsema et al 2004;Mégnin & Romanowicz 2000;Lebedev & van der Hilst 2008;Chang et al 2010;Ritsema et al 2011). Data used in a particular model may simply be insensitive to parameters which are well sampled by data used in another, thereby contributing to differences between models.…”