(1) Models of the inner core anisotropy

Xiaodong Song

Lamont-Doherty Earth Observatory, Palisades, NY 10964
and Dept. of Earth and Environmental Sciences, Columbia Univ., New York, NY 10027,
xsong@ldeo.columbia.edu

poster/oral:

The presence of significant anisotropy in the inner core is now well established from both travel-time and normal-mode data. Current seismological researches focus on the resolution of 3-D distribution of the inner core anisotropy, which is not only critical in understanding the cause of the anisotropy but also essential in estimating elastic properties of the inner core crystals and in resolving the rotation of the inner core.

We recently summarizes the most recent models of seismic anisotropy of the inner core and the elastic parameters of iron crystal aggregate inferred from these models. All the models assume that inner core anisotropy is transversely isotropic with five independent elastic constants (A, C, F, N, and L), which appears to be adequate for the observations currently available. Simple derivation of wave propagation in transversely isotropic media suggest that four of the five elastic constants can be determined from the average P velocity, the average S velocity, and the two parameters from P-wave anisotropy. The fifth elastic constant is completely decoupled from the P wave and can only be directly determined from the observation of the shear wave polarized along the plane perpendicular to the symmetry axis. These estimates of elastic parameters can be compared directly with those of theoretical calculations (Stixrude and Cohen, 1995) and laboratory measurements (Mao et al., 1996).

In particular, anisotropy of one type of shear waves in the inner core (polarized in the plane defined by the symmetry axis and the propagation direction) is estimated to be from a few percent to as much as 13%. The estimates of shear wave anisotropy from normal modes (Tromp, 1993, 1995) vary markedly with depth, but the estimates from body-wave travel times and theoretical calculations (Stixrude and Cohen, 1995) do not differ much (from 8.5 to 12.9\%). The typical estimates of the shear-wave anisotropy from differential PKP times are around 12%.

We also show better determination of these elastic parameters from seismology relies not only on polar paths (nearly parallel to the spin axis) but also on the paths between polar and equatorial directions.


(2) PKP differential travel times:
Implications for three-dimensional lower mantle structure

Xiaodong Song

Xiaodong Song(1) and Don V Helmberger (2)
(1) Lamont-Doherty Earth Observatory, Palisades, NY 10964, xsong@ldeo.columbia.edu
(2) Caltech 252-21, Pasadena, CA 91125, helm@gps.caltech.edu

poster/oral:

Differential PKP(AB)-PKP(DF) differential travel-time measurements are compared directly with predicted residuals for mantle tomographic models to investigate the lateral variation of the lowermost mantle. Much of the AB-DF time is accumulated in the lowermost portions of the mantle, resulting in heavy weighting of 3-D lowermost mantle structure on the differential travel-time perturbations. Comparisons of the observations with tomographic models show strong model dependence. The best variance reductions are achieved using the recent P-wave model of Van der Hilst et al. [1997] (up to 20%). The predictions for present S-wave tomographic models correlate poorly with the observations if we assume P-velocity anomalies are proportional to S-velocity anomalies. We argue that S-velocity anomalies in the lowermost mantle do not always track P-velocity anomalies. Our study shows no obvious evidence of slab effect on the observed AB-DF anomalies and, thus, the use of PKP differential travel times is likely to provide a powerful tool in mapping the P-velocity structure in the lowermost mantle.


Xiaodong Song ( xsong@ldeo.columbia.edu)

Go back to List of Participants
Go to main page