Inferring upper mantle rheology from numerical modeling of experimental and geodetic observations

As a major mantle mineral with volume fraction >60%, olivine is assumed to control the rheology of the upper mantle. While the deformation of olivine and olivine bearing rocks has been extensively researched over the past half century, not all aspects of olivine rheology are well understood. In t...

Full description

Saved in:
Bibliographic Details
Main Author: Masuti, Sagar
Other Authors: Fidel Costa Rodriguez
Format: Thesis-Doctor of Philosophy
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/138899
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:As a major mantle mineral with volume fraction >60%, olivine is assumed to control the rheology of the upper mantle. While the deformation of olivine and olivine bearing rocks has been extensively researched over the past half century, not all aspects of olivine rheology are well understood. In this thesis, I investigate the rheology of olivine using two types of data, viz., geodetic data following the 2012 Mw 8.6 Indian Ocean earthquake and data from experiments on olivine under various thermodynamic conditions. I determine the flow law parameters for the soft- and hard-slip systems, and propose a constitutive relation for the rheology of olivine which consistently captures the transient and steady-state creeps. Using the proposed rheology of olivine, I explain the geodetic observations following the Indian Ocean earthquake with the wet dislocation creep, indicating a minimum water content of 0.01 wt% or 1,600 H/106 Si in the asthenosphere. However, the earthquake broke the entire oceanic lithosphere down to great depths, indicating dry olivine in the lithosphere. I explain the steep rheological contrast by dehydration across the lithosphere asthenosphere boundary, presumably by buoyant melt migration while forming the oceanic crust. Also, using the same proposed rheology, I determine the transient creep flow law parameters from mechanical data of experiments performed under anhydrous conditions on natural dunites. Using a Markov Chain Monte Carlo method, I find different properties for the transient and steady-state components of dunites, with activation energies of EK=430+/-20 kJ/mol and EM=535+/-15 kJ/mol, and stress exponents nK=2.0+/-0.1 and nM=3.6+/-0.1, respectively. These differences suggest that a physical mechanism controlling the transient creep differs from the steady-state creep of olivine aggregates. The lower activation energy of the transient creep could be due to a higher jog density of the corresponding soft-slip system. Furthermore, in order to quantify the effect of water on the transient creep of olivine, I gathered experimental data on the deformation of hydrated single-crystal olivine, where experiments were performed at different pressures and with varying water content. I determine water and pressure effects in olivine are anisotropic: while the water effect on the soft-slip system (i.e., [110]c orientation) is small (i.e., r[100]=0.35+/-0.08), it is large (i.e., r[001]=1.3+/-0.30) for the hard-slip system (i.e., [011]c orientation). Assuming a model of soft-slip controlling the transient creep and hard-slip controlling the steady-state creep proposed for ice is true for olivine, I propose that short-term time-dependent small strain deformation associated with post-seismic period is strongly dependent on the water content.