Earth and Planetary Science Letters 393 (2014) 60–72 Contents lists available at ScienceDirect Earth and Planetary Science Letters www.elsevier.com/locate/epsl Rheological properties of the mantle lid beneath the Mojave region in southern California Whitney M. Behr a,∗ , Greg Hirth b a Dept. of Geological, Jackson School of Geosciences, Sciences University of Texas at Austin, 2275 Speedway Stop C9000, Austin, TX, USA b Department of Geological Sciences, Brown University, Providence, RI, 02912, USA article info abstract Article history: Received 19 October 2013 Received in revised form 5 February 2014 Accepted 16 February 2014 Available online xxxx Editor: Y. Ricard Keywords: mantle rheology postseismic relaxation lithospheric strength paleopiezometry olivine deformation strain localization Recently deformed mantle peridotite xenoliths derived from Moho depths constrain the geothermal gradient, stress magnitude, effective viscosity, and degree of localization in the uppermost mantle within the tectonically active Mojave region of California. Microstructural observations and water content measurements in the xenoliths indicate that upper mantle deformation is accommodated by dislocation creep of modestly hydrated olivine. Differential stress measured in the xenoliths using olivine paleopiezometry is 13–17 MPa, which is at least one order of magnitude less than peak stresses estimated for the gabbroic lower crust and the brittle upper crust. Similarly, the mean effective viscosity of ∼3 × 10 19 Pa s for the uppermost mantle is one to two orders of magnitude less than the mean viscosity estimated for the lower crust, consistent with recent models of postseismic relaxation following the Landers (1992) and Hector Mine (1999) earthquakes. These results support a rheological model for the Mojave region in which the peak stress resides in the crust, rather than within the lithospheric mantle (consistent with the ‘crème brûlée’ model of lithospheric strength). Temperatures and pressures recorded in the xenoliths indicate a high geothermal gradient of at least ∼30 ◦ C/km, which may explain the regional weakness of the mantle lid. Strain rates calculated for the uppermost mantle using the xenolith data and olivine flow laws for wet dislocation creep are ∼7–70 times faster than bulk strain rates estimated across the central Mojave region from GPS-constrained surface velocities. This suggests that faults at the eastern border of the Eastern California Shear Zone persist through the seismogenic zone and retain their identities as narrow ductile shear zones into the mantle beneath the Moho. The observation that recent deformation is localized into only 10–25% of the rock body may explain why upper mantle seismic anisotropy in the Mojave region is highly oblique to present-day plate boundary motion.  2014 Elsevier B.V. All rights reserved. 1. Introduction Long-standing questions in geodynamics include: What is the integrated strength of the crust and upper mantle, and at what depth does the peak strength reside (Hanks and Raleigh, 1980; Jackson, 2002; Burov and Watts, 2006)? Stress in the lithosphere can be estimated by extrapolating laboratory deformation data to natural conditions using flow laws of the form ˙ ǫ = Ad m σ n f H 2 O exp  −Q + PV RT  (1) where ˙ ǫ is strain rate, A is a material constant, σ is stress, f H 2 O is water fugacity, d is grainsize, Q is the activation energy, R is * Corresponding author. Tel.: +1 512 232 1941. E-mail address: behr@utexas.edu (W.M. Behr). the gas constant P is pressure, V is the activation volume and T is temperature. The stress and grainsize exponents, n and m, vary from ∼1–5 and 0–3, respectively, depending on deforma- tion mechanism. These extrapolations lead to strength profiles (Sibson, 1983), which form the basis for geodynamic models of processes ranging from regional post-seismic relaxation (Freed and Bürgmann, 2004; Bürgmann and Dresen, 2008) to global mantle convection (Landuyt et al., 2008; Bercovici, 2003). Extrapolation of flow laws to the lower crust and upper mantle requires assump- tions about the rock type and deformation mechanism, as well as grainsize, water content, thermal gradient, strain rate and scales of localization—parameters that vary by two to three orders of magnitude regionally within Earth, and by up to ten orders of mag- nitude (for strain rate) in Earth relative to the laboratory. Advances in electron-backscatter diffraction (EBSD) and high-resolution el- emental mapping, however, have great potential for quantitatively characterizing these parameters for individual regions. Quantitative http://dx.doi.org/10.1016/j.epsl.2014.02.039 0012-821X/ 2014 Elsevier B.V. All rights reserved.