Seismological Tests (and Implications) of Post-Perovskite Presence in the Deep Mantle In collaboration with Ed Garnero, Alex Hutko John Hernlund In collaboration.

Seismological Tests (and Implications) of Post-Perovskite Presence in the Deep Mantle

Seismological Tests (and Implications) of Post-Perovskite Presence in the Deep Mantle

In collaboration withEd Garnero, Alex HutkoJohn Hernlund

In collaboration withEd Garnero, Alex HutkoJohn Hernlund

Thorne LayUniversity of California, Santa CruzDepartment of Earth and PlanetarySciences

University of California, Santa CruzDepartment of Earth and PlanetarySciences

Lay, Garnero and Williams, PEPI, 2004

Is the D” S-wave velocitydiscontinuity first observedby Lay and Helmberger (1983) caused by the Perovskite-to-Post-perovskite phase changediscovered in 2004?

Lay et al., EOS, 2005

A Diversity of P- and S-Velocity Structures in D” is Observed

Lay and Garnero, AGU Monograph, in press, 2007

MgSiO3

Pv-pPv

Lay and Garnero, AGU Monograph, 2007

Synthetic predictions for P-wave models

Model PWDK Model WOOKWeber and Davis [1993] Wookey et al. [2005]

Pv-pPvphase change

model

PdP ?

Pv-pPv predicts littleP wave reflectivityWookey et al. 2005 Wentzcovich et al. 2006

P-wave models arenot easily reconciledwith predictions of thePv-pPv transition, but the predicted effects are small.

How to explain large P-discontinuties in some regions? Anisotropy?Effect of Fe, Al?

Lay and Garnero, AGU Monograph, in press, 2007

P-Wave Predictions andModels

Pv-pPv

Some S-wave models can be reconciledwith predictions of the Pv-pPv transition, and the predicted effects are observable. But some models arenot consistent.

How to explain S-velocity decreases?Fe, Al, anisotropy?

Lay and Garnero, AGU Monograph, in press, 2007

S-Wave Predictions and Models

Pv-pPv

Most Studied Region:Caribbean, CentralAmerica, Cocos Plate

High S velocity, slightlyHigh P velocity

A dozen Swave studiesindicate wide-spread Sdiscontinuity200-300 km above the CMB

Lay, Garnero, Russell, GRL, 2004

Complexity of D” reflector observed beneath the Cocos Plate.

Thomas, Garnero, Lay, JGR, 2004

Complexity of reflectors may imply thermal complexityaffecting the phase boundary. Seismic migration isUsed to image 3D structure.

Hutko, Lay, Garnero and Revenaugh, Nature, 2006

Rapid changes in depth of reflector appear to exist; needsspecial circumstances to account for by phase boundary.

Kito, Rost, Thomas, Garnero, GJI, 2007

Step in S velocity supported.P velocity decreases (?), maybe two reflectors for P (?),near 340 km and 190 km above the CMB.

Hutko, Lay, Revenaugh, Garnero, in prep., 2007

P velocity structure investigation with very large data set.

Hutko, Lay, Revenaugh, Garnero, in prep., 2007

Can we determine sharpness of the reflectors?

S increase

Hutko, Lay, Revenaugh, Garnero, in prep., 2007

Are the P velocity decreases really there? Maybe, but <0.5%.

S increase

Lay, Garnero, Williams, PEPI, 2004

LLSVPs haveDistinctive Vs/Vp Behavior

Shear velocitiesanti-correlatedwith bulk soundvelocities.

Lay, Hernlund, Garnero and Thorne, Science, 2006

Large low shear velocity provinces (LLSVPs) are usually thought to be warm; hence no post-perovskite, or only in a thin layer. But, this appears to be wrong. S-wave discontinuities are seen where they “shouldn’t be”.

Lay, Hernlund, Garnero and Thorne, Science, 2006

Lay, Hernlund, Garnero and Thorne, Science, 2006

Paired velocitydiscontinuitiesmay be due todouble-crossingof the phaseboundary in lowermost thermalboundary layer,as suggested by Hernlund,Thomas, andTackley, Nature, 2005

Lay, Hernlund, Garnero and Thorne, Science, 2006

The local CMB heat flux can be estimated; depends on the Clapeyron slope of Pv-pPv and on excess temperature above the phase boundary at the CMB, along with thermal conductivity (poorly constrained).

My preference~85 mW/m2

Lay, Hernlund, Garnero and Thorne, Science, 2006

So, we think pPv could exist both in cold down-welling regionsand in cool down-wellings inside large chemical piles.

Lay et al., Science, 2006

Lay and Garnero, AGU Monograph, in press, 2007

So, is there post-perovskiteIn the lower mantle?

It seems like the best working hypothesis we have to explain shear velocity discontinuities.

It may account for a significant portion of large-scale heterogeneity patterns, reducing inferred thermal and chemical variations

But, further testing is warranted. All interpretations subject to future revision.

Lay and Garnero, AGU Monograph, 2007

Large-scaleLowermost mantle heterogeneity

Large-scaleLowermost mantle heterogeneity

Lay, Garnero, Williams, PEPI, 2004

Mapping Tomographic Shear Velocity VariationsInto a Post-perovskite Layer

Helmberger, Lay, Ni and Gurnis, PNAS, 2005

Garnero, Lay and McNamara, GSA Monograph, in press, 2007

Cool down-wellings onedge of LLSVP maybe where a post-perovskitelens can form. It will not be found in everyregion then.

Phase Change Could Occur Twice in Thermal Boundary Layer

But, if is Hard to Observe

Flores and Lay, GRL, 2005