mineral physics and seismic constraints on earth’s structure and dynamics earth stucture,...
Post on 28-Dec-2015
219 Views
Preview:
TRANSCRIPT
Mineral physics and seismic constraints on Earth’s structure and dynamicsEarth stucture, mineralogy, elasticity
Primary source of information about the deep interior structure:
seismological data
Other sources of information, Earth structure and dynamicsGravitational field
Magnetic field
Heat flow, dynamic topography and geoid
Plate movements
Cosmochemistry and geochemistry
High pressure mineralogy and mineral physics: experimental – computational
Other planets: No seismology (except rudimentary Moon seismology)
Planetary mass distribution: determined via simple flyby
Moment of inertia factor (MIF)
MIF = I / R2 M
Homogeneous massive sphere: MIF = 0.4
Planet with high-density core: MIF < 0.4
Lay & Garnero (2011, Ann Rev Earth Plan Sci)
Seismic phases generated by a 562 km deep source in PREM 1D Reference Earth Model of Dziewonski and Anderson (1981)
Travel time curves. Dashed and dotted curves are upward-radiated P- and S- wave surface reflections (e.g. pPdiff and sPdiff)
G/r = vs2
K/r = vp2 – 4/3vs
2 = vf2 = F
vf: bulk sound velocity,
F : seismic parameter
Seismic velocities physical properties
Bulk modulus: K (= incompressibility / stiffness)
Shear modulus: G
vs2 = G/r
vp2 = (K + 4/3*G)/r
Bragg's lawpositive interference when nl = 2d sinq
Unit cell V and r as a function of p In-situ high-pT XRD, using high-intensity synchrotron radiation
Angle-dispersive XRDMonochromatic beam, fixed , l variable q
Energy-dispersive XRDPolychromatic (”white”) beam, fixed qd-spacings from the energy peaks
E = hf = hc/l
= l hc/E 2d sinq = nl = nhc/E
gasket
Dziewonski & Anderson (1981, PEPI) 4082 citations, May 28, 2013
ol ga px
wd rw ga
bm fp Ca-pv
pbm fp Ca-pv
liquid FeNi0.1
+ minor Si, O, S
solid FeNi0.1
First-order Earth structurePREM: Preliminary Earth Reference Model
- from seismology (normal modes) and gravity- includes r and p
Mg-pv
Ferro-periclase
garnet
Ca-pvga
garnetfp
FeNiS-metal
Mg-perovskite
BSE-image of subsolidus phase relations, 24 GPa
Simple system Mg2SiO4
Best one-component analogue to peridotite
Phase relations UM, TZ, LM:
Modified from Fei & Bertka (1999, Geochem. Soc. Spec. Publ. 6)
Pyroxene: Mg[6] Si[4] O3
Garnet: Mg3[8] MgSi[6] Si3
[4] O12
Akomotoite (ilmenite): Mg[6] Si[6] O3
Perovskite: Mg[8] Si[6] O3
System MgSiO3
Modified from Fei & Bertka (1999, Geochem. Soc. Spec. Publ. 6)
High-p crystal chemistry- without coordination number (CN) increase: high-p (or low-T) phase transitions: often decreasing symmetry
- CN-increase is common for high-p phase transitions Explanation: large anions are more compressible than small cations → reduced ranion/rcation-ratio
Stixrude and Lithgow-Bertelloni (2011, GJI)pv: Mg-perovskitefp: ferropericlasemw: magnesiowustiteol: olivinewd: wadsleyiterwd: ringwooditest: stishovite
op: orthopyroxenecp: clinopyroxeneak: akimotoitega: garnetcor: corundum
First-order constraints on temperature
Inner-outer core boundary at 330 GPa / 5150 km: melting temperature of FeNi
660 km discontinuity:Reaction rwd = pv+fp at 24 GPa (endothermic transition - small drop in adiabat)
Location of mantle adiabat:below solidi of peridotite and basalt
Location of outer core adiabat:above solidus of FeNi (+ Si, O, S)
CMB: extreme thermal boundary layer ! 2500 - 3800 K ! (DT: 1300K)
Why such a large thermal boundary layer at CMB ?
Density contrast 5500 - 9900 kg/m3
precludes mantle - core mixing
Viscocity of solid rock is quite high, even at very high T near the CMB
peridotite liquid FeNi
T-dependent viscosity models Steinberger and Calderwood (2006, GJI)
CMB
Grand model, Masters and Laske, website
Seismic tomography models Large vS-amplitudes at the top and bottom of the mantle
Montelli et al. (2006, GGG)(finite frequency tomography)
S-wave models 6 depth sections: 900-2800 km
Two large anti-podal, slow provinces - LLSVP Africa – Pacific (near equator - 180º apart)
S-wave models, lowermost mantle (D”-zone)
The degree-2 velocity anomalies, recognized
>30 years ago, coincide with the residual geoid
e.g. Dziewonsky et al. (1977, JGR), Dziewonski & Anderson (1984, Am Sci)
Dziewonski et al. (2010, EPSL)
Dziewonski et al. (2010, EPSL)
L2-norm L1-norm
Cluster analysis of 5 tomographic modelsLekic et al. (2012, EPSL)
Seismic tomography, D"
Note the plume locations: many/most along the LLSVP-margins
Paleogeographic relocation Clustering near periphery of LLSVPs
- long-term stability- dense and hot (thermo-chemical piles)
Large igneous provinces (LIPs) - age span: 16-297 Ma
SC
–1%
slow
+2.5% fast
–3%slow
AfricaPacific
Burke & Torsvik, 2004, EPSLTorsvik et al., 2006, GIJBurke et al. 2007, EPSLTorsvik et al. 2008, EPSL
SC
Comparison of seismic tomography (LLSVPs)and slab-sinking model at 2800 km depth Dziewonski et al. (2010, EPSL)
Lithgow-Bertelloni & Richards(1998, Rev. Geophys.)
Degree 2 Degree 2
Spherical harmonics modeling
Power spectra Cumulative power spectra
Slab m
odel
Tomog
r.
mod
elsTomographic models
Slab model
Tentative conclusions 1. The observed degree-2 pattern is only partly reproduced by calculated slab-accumulation
2. The LM-structure may thus be old ( > 300-500 Ma)
S-wave model
NE part of Pacific LLSVPSamoa quakes, recorded in N-America
S-wave model
Double crossing of thepv-ppv-transition
Large lateral variation Horizontal flow
Lay et al (2006)
Bin 1-3
Mantle flow model
top related