Frequency‐dependent seismic attenuation in the inner core 2. A scattering and fabric interpretation

Abstract

[1] Broadband velocity waveforms of PKIKP in the distance range 150-180°are inverted for a model of inner core attenuation due to forward scattering by a threedimensional heterogeneous fabric. A mean velocity perturbation of 8.4 ± 1.8% and a scale length of heterogeneity of 9.8 ± 2.4 km are determined from 262 available PKIKP ray paths. The velocity perturbations are larger for polar than for equatorial paths, decrease with depth, and show anisotropy in both global and regional data. For paths beneath North America, the smallest scale lengths (1-5 km) tend to lie in either the upper 200 km of the inner core or along paths close to the rotational axis. The depth dependence of attenuation is roughly similar to that obtained assuming a viscoelastic origin, except a more abrupt transition is seen between higher attenuation in the upper inner core and lower attenuation in the lower inner core. This transition may be sharp enough to produce either a first-or second-order discontinuity with depth in the long-wavelength (composite) elastic moduli. A fabric that satisfies the observed depth dependence and anisotropy of attenuation requires solidification of iron crystals having high (>10%) intrinsic anisotropy, which are preferentially aligned in time and depth. Since weak velocity dispersion, elastic anisotropy, attenuation anisotropy, and their depth dependence agree with that predicted by such a fabric, we suggest that scattering attenuation is not a small fraction but, rather, the predominant mechanism of attenuation in the inner core in the 0.02-2-Hz frequency band. Citation: Cormier, V. F., and X. Li, Frequency-dependent seismic attenuation in the inner core, 2, A scattering and fabric interpretation,

Sharp and seismically transparent inner core boundary region revealed by an entire network observation of near-vertical PKiKP

“…As this is hard to observe in our data, their model does not seem to be inconsistent with our data. On the other hand, the model suggested by Cormier and Li (2002) predicts much larger scattered energy (similar to the input PKiKP amplitude) that should be observed in the present data. As their model is the maximum one for attributing all the cause of the attenuation to the scatterers, absence of such a large scattered energy indicates that there exists substantial intrinsic attenuation in the inner core.…”

Deep Earth Structure: The Earth’s Cores

“…An innermost inner core with P-wave anisotropy rate e ¼ 0 is obtained by Ishii et al (2002a) (Figure 11(c), middle). There is however some agreement about the mean radius of the innermost inner core, in the range 300-600 km for the most recent studies (Beghein and Trampert, 2003;Calvet et al, 2006;Cormier and Li, 2002;Sun and Song, 2008). There is however some agreement about the mean radius of the innermost inner core, in the range 300-600 km for the most recent studies (Beghein and Trampert, 2003;Calvet et al, 2006;Cormier and Li, 2002;Sun and Song, 2008).…”

Deep Earth Structure: Seismic Scattering in the Deep Earth

“…examined LASA data for two nuclear explosions 3 years apart and separated by <1 km. Cormier and Li (2002) argued that the large discrepancy in RMS velocity perturbation and scale length between their study (RMS ¼ 8%, scale length ¼ 10 km) and the model (RMS ¼ 1.2%, scale length ¼ 2 km) of Vidale and Earle (2002) may be due to the significant intrinsic attenuation assumed by Vidale and Earle and differences in depth sensitivity between the studies. In contrast, much smaller time differences are observed in PKKP and PKPPKP arrivals, supporting the idea that the time dependence originates in the inner core.…”

“…They computed a fractional energy loss of 10% from scattering in the top 300 km of the inner core, helping to justify the use of the Born singlescattering approximation. Cormier and Li (2002) inverted 262 broadband PKP(DF) waveforms for a model of inner-core scattering attenuation based on the dynamic composite elastic medium theory of Kaelin and Johnson (1998) for a random distribution of spherical inclusions. They identified systematic time shifts in PKiKP coda, which they explained as resulting from differential inner-core rotation, as previously proposed by Song and Richards (1996).…”