Geoscience Reference
In-Depth Information
between pv and fp was adopted along the entire
pressure profile, and the linear relationship for
the variation of elastic properties between Mg
and Fe end-members was assumed, although
there is very limited data on how Fe content and
its spin transition affect the shear modulus of
pv. CaSiO
3
pervskite is believed to be a minor
constituent mineral phase in the lower mantle.
There are no experimental data in the literature
on the sound velocity of CaSiO
3
pervskite at
high pressure. The sound velocity of CaSiO
3
perovskite has been believed to be relatively high
according to the previous computational works
(e.g. Karki & Crain, 1998). However, recent
theoretical reassessment, including the effect of
tetragonal distortion of CaSiO
3
perovskite under
pressure that was recently experimentally identi-
fied (Shim
et al
., 2002), revealed that the effect of
CaSiO
3
perovskite on the seismic velocity profile
in lower mantle is almost negligible (Li
et al
.,
2006). For that reason, CaSiO
3
perovskite is not
included in the present modeling.
3000
Models of lower mantle geotherm
2800
Anderson, 1982
Brown & Shankland, 1981
2600
2400
2200
2000
1800
Best Fit for pyrolite model
1600
30
40
50
60
70 80
Pressure (GPa)
90 100 110 120 130
Fig. 6.11
Representative lower mantle geotherms for
whole mantle convection (Brown & Shankland, 1981)
and layered mantle convection (Anderson, 1982)
models. The gray line represents the best-fit
temperature profile for pyrolitic lower mantle model.
effect of temperature on the lower mantle min-
eralogical modeling is relatively milder than that
of pressure. However, the geothermal structure
strongly depends on the convection styles of the
mantle and distribution of heat sources in the
Earth, the choice of the lower mantle geothermal
models should provide an important key criterion
for the mineralogical model of the lower mantle.
In that context, two extreme geothermal models
were adopted in this modeling, one for whole
mantle convection (Brown & Shankland, 1981)
and
6.4.2 Lower mantle geotherms
Thermal structure of the lower mantle is still
poorly constrained. The uncertainties include the
geothermal gradient and the possible temperature
jump near the 660 km discontinuity (Brown &
Shanland, 1981; da Silva
et al
., 2000; Anderson,
1982; Stacey, 1992; Hofmeister, 1999). Jackson
et al
. (1983) systematically investigated the
tradeoff between composition, temperature and
assumed physical properties such as thermal
expansion of the simple lower mantle assemblage
of silicate perovskite and ferropericlase miner-
alogy, which also leads to the preferable lower
mantle geotherms respectively for peridotitic
mantle and perovskite mantle to satisfy the
seismological data. The result demonstrated that
the peridotitic mantle requires lower goetherm
than that for perovskite mantle. Since the
variation of the density and elastic moduli with
temperature is in fact not so much sensitive
relative to that with pressure (Mattern
et al
.,
2005, Stixrude & Lithgow-Bertelloni, 2005), the
the
other
for
layered
mantle
convection
model (Anderson, 1982) (Figure 6.11).
6.4.3 Elasticity data set
The elasticity data set used in the modeling is
given in Table 6.3. Note that the listed parameters
given in Stixrude & Lithgow-Bertelloni (2005) are
the representative or averaged values compiled
from previous experimental or theoretical data
(Smyth & McCormick, 1995; Wentzcovitch
et al
.,
2004; Fiquet
et al
., 2000; Jeanloz & Thompson,
1983; Kiefer
et al
., 2002; Robie
et al
., 1995; Fiquet
et al
., 1999; Anderson & Isaak, 1995; Fei, 1995;
Jackson
et al
., 1990; Jacobsen
et al
., 2002; Stolen
et al
., 1996). In this modeling, the shear modulus
