Goals of the
project:
With Ludovic
Margerin (IRIGM, Grenoble, FRance), I am investigating the amplitude variations
of precursors to PKP, which are indicative of small scale heterogeneity
in the mantle. These scatterers arrive before any other wave and
are clearly visible in the shadow zone, as shown in Figure 1.
Figure 1.
Evidence that such scatterers originate
in the mantle is shown in Figure 2, where we plot the signal/noise ratio
of the observations (given by the size of the circle - small dots have
S/N=1) in our data set in a plane of epicentral distance vs. the onset
time with respect to the PKP arrival. The solid line in this plot denotes
the theoretical location for scattered waves originating at the core-mantle
boundary itself. No scattered waves are observed outside this theoretical
boundary, clearly indicating no scattered waves originate in the core.
Figure 2
This heterogeneity
was long thought to be located in or near the D" layer itself, i.e.
in the first 100 or 200 km above the core mantle boundary, but recently
Michael Hedlin and Peter Shearer have cast doubt on this interpretation.
Our analysis uses a multiple scattering approach, in contrast to the single
scattering used by previous investigators.
Figure 3
We have focused
on four corridors in the Earth (Figure 3) that generate strong precursive
energy preceding PKP at epicentral distances between 124 and 142 degrees,
and have assembled a data set of strong precursors.
Figure 4
Averaging these
over all regions, we find we can match the observed envelopes over all
epicentral distances considered with just one model that has weak (about
0.1%) scatterers throughout the mantle with a scale length that is ill
resolved (Figure 5). This differs from Hedlin's result of 1%, but
the low scattering strength does not seem to indicate that Hedlin's single
scattering formalism is the cause of this discrepancy.