The spatial distribution of seismic velocity revealed by global tomography has been a powerful tool in understanding the internal structure of the Earth. In the uppermost mantle, the higher than average velocity under continents is commonly interpreted as the signature of continental roots, as they are presumably colder than their surrounding. In general, the extension of this signature in depth correlates to the lithospheric thickness. However, this seismically defined thickness of continental roots shows diverse features among the recent global tomographic models. Unlike the well documented anisotropic feature under Centre Pacific (e.g. Montagner and Tanimoto, 1991; Ekström and Dziewonski, 1998) , these dissimilar features below lithospheric depth are less explored, and are generally accepted as a result of various data coverage and different methodology.
Here we present results from a global tomographic model derived by using joint anisotropic inversion of and . Our results demonstrate that these diverse characteristics can be explained by the ubiquitous existence of transverse anisotropy beneath major continental shields.
Separate isotropic inversion of Love and Rayleigh waves is a convenient approach to measure the anisotropy in the uppermost mantle based on the fact that the fundamental modes of Love/Rayleigh waves are mainly sensitive to /. However, the condition is not generally true for overtone phases (Figure 32.1), and it limits the resolution of the anisotropic features derived from the pseudo-isotropic inversion(e.g. Regan and Anderson, 1984).
In this study, we implement a joint inversion on three component seismic waveform data. The sensitivity kernels of , , , (isotropic ), , and are calculated based on the assumption of weak transverse anisotropy. Starting from our most recent tomographic models, SAW24B16(Mégnin and Romanowicz, 2001) for and SAW16BV (Gung and Romanowicz, 2002; Romanowicz and Gung, 2002 ) for , we invert for and up to degree 16. The scaling relations among anisotropic parameters , and based on Montagner and Anderson(1989) are applied. This allows us to incorporate the effects of on Love waves, on Rayleigh waves and the effects of anisotropy. Since component data have less sensitivity in the lowermost mantle than component data, to avoid the bias from anisotropic features from the deep mantle, we have restricted our inversion to the top 1500km of the mantle in this study.
Three component waveform data composed of surface waves(85,000 wavepackets) and body waves ( 60,000 wavepackets ) are used in this study. In addition to fundamental surface waves, we also include overtone phases, which greatly enhance the resolution in the upper mantle transition zone.
Figure 32.2 shows our preliminary results of depths 175km and 250km. It is observed that the major features in and from the separate inversions are preserved in this depth range. Probably because it is mainly constrained by fundamental surface waves.
At the depth of 175km, the Pacific region shows similar anisotropic features as previous studies (e.g. Montagner and Tanimoto, 1991; Ekström and Dziewonski, 1998) most areas at this depth are characterized by . There are two prominent exceptions in our model: East Pacific Rise and regions around Hawaii islands, where we observe strong signal of , presumably due to vertical flow.
The anisotropy under oceanic regions at larger depth (250km) becomes weaker, and the contrast between and is highlighted in the continental regions. We observe that most continental shields are characterized by , such as Canadian shield, Siberia platform, Baltic shields, West Africa, Amazonian and Australian craton.
We propose that this anisotropy, characterized by , is due to asthenospheric horizontal flow underneath the continental lithosphere. Similar to what happens under the Pacific plate, albeit at shallower depth (because the lithosphere is thiner). We also propose that this strong anisotropy under continents marks the Lehmann discontinuity, and the anisotropy under oceanic plates marks the G discontinuity. This explains the geographic preferential detections of these two discontinuities. (Gu, Dziewonski and Ekström, 2001; Revenaugh J. and T. H. Jordan, 1996)
We thank the National Science Foundation for support of this research.
Ekström, G. and A. M. Dziewonski, The unique anisotropy of the Pacific upper mantle, Nature, 394, 168-171, 1998.
Gu, Yu. J., Dziewonski Adam, and Ekström G., Preferential detection of Lehmann discontinuity beneath continents,Geophys. Res. Lett. 28, ,4655-4658, 2001
Gung, Y., and B. Romanowicz, tomography of the upper mantle using three component long period waveforms, submitted to Geophys. J. Res., 2002
Li, X.D. and B. Romanowicz, Global mantle shear-velocity model developed using nonlinear asymptotic coupling theory, Geophys. J. R. Astr. Soc., 101, 22,245-22,272, 1996.
Mégnin, C., and B. Romanowicz, The 3-D shear velocity structure of the mantle from the inversion of body, surface, and higher mode waveforms, Geophys. J. Inter., 143 709-728, 2001.
Montagner J. P. and Don L. Anderson, Petrological constraints on seismic anisotropy, Phys. Earth and Plan. Int., 54, 82-105, 1989.
Montagner J. P. and T. Tanimoto, Global upper mantle tomography of seismic velocities and anisotropies, Geophys. J. Res. 96,, 20,337-20,351, 1991.
Regan, Janice and Don L. Anderson, Anisotropic model of the upper mantle, Phys. Earth and Plan. Int., 35, 227-263, 1984.
Revenaugh J. and T. H. Jordan, Mantle layering from reverberations 3. the upper mantle Geophys. J. Res. 96,, 19,781-19,810, 1991.
Romanowicz, B., and Y. Gung, Superplumes from the core-mantle boundary to the lithosphere: implications for heat flux, Science, 296, 513-516, 2002.