A compilation of elastic anisotropy measurements from metamorphic rocks

Abstract

Preamble

Nikolas Christensen, a pioneer in the study of the elasticity of crust and upper mantle lithologies, passed away on May 19, 2022 (see https://rock.geosociety.org/net/documents/gsa/memorials/v51/Christensen-NI.pdf). Prior to his passing, Nik had been working on a manuscript that described and interpreted his extensive data base of seismic anisotropy measurements. He provided the manuscript materials to one of us (M.G.B.) for comment with the intent of eventual submission. Although the manuscript was not completed, the initial sections, which provide context, describe the methodology, and summarize the measurements, were reasonably self contained. More importantly, the extensive tables of painstakingly taken measurements that form the basis of the work had been prepared in publication-ready form and as spreadsheets. Owing to the profound and invaluable contributions of Nik's previous compilation papers on crustal composition (Christensen & Mooney, 1995) and the isotropic elasticity of common lithologies Christensen, 1996) to the seismological and geological communities, it is our opinion that the present work should be published for the benefit of future scientific investigations of lithospheric anisotropy. We have elected to submit the manuscript on Nik's behalf as a research note to Seismica. The title has been changed from the original ``Metamorphism and crustal seismic anisotropy: A global perspective'' to the present one, but aside from this and minor editorial revisions, it is a faithful representation of the original draft. Note that this manuscript is accompanied by an independent commentary by Douglas Schmitt in this issue of Seismica.
-- Michael G. Bostock, Simon M. Peacock, Matthew S. Tarling. The University of British Columbia.

Abstract

An increasing number of seismic investigations have reported convincing evidence for the widespread existence of crustal anisotropy in a variety of tectonic regions. Interpretations of these observations, as well as future seismic studies designed specifically to investigate crustal anisotropy, require detailed knowledge of anisotropic wave propagation in rocks which have undergone deformation and accompanying recrystallization. Of particular importance are the symmetries and magnitudes of P- and S-wave anisotropies and S-wave splitting. A detailed experimental investigation of the anisotropic properties of metamorphic rocks has been carried out to hydrostatic pressures of 1GPa. Each measurement averages the orientations and correct elastic properties of hundreds of thousands of grains, as well as takes into account the important effects of grain shape and grain boundaries on velocities. Common metamorphic rocks, especially those with pelagic protoliths, often have axial symmetries with slow P-wave velocities normal to cleavage, schistosity, and banding. For slates, phyllites, and quartz mica schists, S-wave singularities occur at angles averaging 42° from their symmetry axes, as well as parallel to symmetry axes. Many axial symmetry amphibolites also have slow P velocities and elastic properties similar to crystals with hexagonal symmetry, but unlike metapelitic rocks do not possess off axis S-wave singularities. Rocks with fast axis P-waves and axial symmetry include blueschists, marbles, and dunites. S-wave singularities for these rocks appear to be limited to propagation parallel to symmetry axes. Of importance, maximum S-wave splitting does not always coincide with propagation normal to symmetry axes, and fast vibration directions can be normal as well as parallel to the strike of foliation. Rocks with well-developed foliations and lineations have, as expected, seismic properties similar to those of orthorhombic single crystals. P-wave velocities are fast parallel to lineations originating from foliation crenulations and mineral elongations. Orthorhombic rock S-wave singularities are rare for propagation in mirror planes, but, when present, occur in symmetry planes defined by the maximum and minimum P-wave velocities. Crustal regions most likely to show strong seismic anisotropy include accretionary prisms containing abundant slate and phyllite and crustal regions rich in quartz mica schist and amphibolite.