Olivine [100] normal to foliation: lattice preferred orientation in prograde garnet peridotite formed at high H2O activity, Cima di Gagnone (Central Alps)

Abstract

Automated electron backscattered diffraction (EBSD) was applied using a scanning electron microscope to obtain lattice preferred orientation (LPO) data for olivine in garnet peridotites of the Central Alps. As a reference frame, the LPOs of enstatite were also investigated. In the garnet peridotite at Cima di Gagnone (CDG), a weak foliation carrying a distinct lineation is present. The lineation is characterized by elongated enstatite, olivine and poikiloblastic garnet. Olivine shows a very unusual LPO with [100] normal to foliation and [001] parallel to lineation. Achsenverteilungsanalyse (AVA) maps demonstrate that [001] of olivine grains corresponds quite well to their maximum length axes which are preferentially parallel to the lineation. Numerous planar hydrous defects within (001) planes of olivine are marked by palisades of ilmenite rods and show a preferred orientation normal to lineation. Calculated P-wave velocities for CDG are fastest (8.32 km s−1) normal to foliation with a relatively low anisotropy (2.9%). Compared to mantle peridotites with the usual (010)[100] LPO where the fastest Vp direction is towards the lineation, the relationship between flow geometry and seismic anisotropy is significantly different at CDG. Several mechanisms for the formation of the LPO type at CDG are considered, with glide possible on (100)[001] of olivine. On the basis of field data as well as petrographic and petrologic evidence, it has been demonstrated that the CDG garnet peridotite formed by prograde metamorphism from a hydrous protolith at pressures and temperatures of about 3.0 GPa and 750 °C, respectively. The CDG LPO is interpreted to have formed during hydrous subduction zone metamorphism. The same interpretation may hold for the previously investigated olivine LPO at Alpe Arami, which is similar to that at the nearby CDG. The observed anomalous LPO is no proof for ultradeep (>3.0 GPa) conditions.