Magnetic Properties of Mantle Xenoliths and Implications for Long Wavelength Magnetic Anomalies

Author

Sarah Alyson Friedman, Southern Illinois University CarbondaleFollow

Date of Award

8-1-2011

Degree Name

Master of Science

Department

Geology

First Advisor

Ferre, Eric

Abstract

Unaltered peridotite xenoliths are broadly representative of the lithospheric mantle in both oceanic and continental domains. These peridotites are mainly lherzolites and harzburgites. Other rock types such as dunites, wehrlites and pyroxenites are generally not volumetrically significant. The respective contributions of rock-forming minerals to induced and remanent magnetization in these rocks are currently poorly constrained. This information can be used to assess the significance of long-wavelength magnetic anomalies. It can also provide insights, as an alternate approach to the spinel-olivine-pyroxene oxybarometer, into several important petrologic parameters of the lithospheric mantle including fO₂. Forty-nine representative, uncontaminated and non-serpentinized xenoliths have been magnetically investigated. These specimens display contrasting remanent magnetic properties (NRM, M_r, M_s) depending on their tectonic settings, specifically oceanic hot-spot, continental mantle plume, island arc, and craton. The main paramagnetic silicates (olivine, clinopyroxene, orthopyroxene, etc...) typically account for most of the peridotite magnetic properties. The low-field bulk magnetic susceptibility of pristine, unaltered mantle xenoliths is ≈ 500 ± 60 x 10-6 [SI] and displays limited variability. The total contribution of paramagnetic silicates to magnetic susceptibility (K_{para-silicates}) can be determined from the high-field slope of a saturated hysteresis experiment. K_{para-silicates} can also be calculated by adding the respective contributions of individual silicates based on their modes, chemical composition, and the Bohr magneton numbers of individual cations. Silicates account for between 56 and 97% (average ≈ 85%) of the magnetic susceptibility depending on rock composition. When present, the contribution of chrome spinel, which is paramagnetic in the absence of late-stage exsolution products, remains around 1%. Plagioclase-, spinel- and garnet-lherzolites share similar low-field magnetic properties. The remaining contribution to magnetic susceptibility arises from variable amounts of primary magnetite (and pyrrhotite to a minor extent). These mineral phases, although present in tens to hundreds of ppm only, contribute significantly to the rock magnetic properties because they have large intrinsic magnetic susceptibilities (≈ 1 to 4 [SI] for magnetite). Stoichiometric magnetite has been identified as microscopic exsolutions in the lattice of olivine and accounts for 2 to 43% (average ≈ 8%) of the magnetic susceptibility. Whether these pseudo-single domain magnetite grains are in equilibrium with other rock-forming minerals or not is still being investigated. Pyrrhotite (up to 600 ppm in some rare specimens), although detectable in low-temperature magnetic experiments, does not significantly contribute to magnetic susceptibility. The contribution of ferromagnetic minerals, such as magnetite and pyrrhotite, to remanent magnetization (M_r) is significant and varies greatly (over 250x between specimens) with tectonic setting. The fact that all specimens contain primary magnetite suggests that these assemblages equilibrated at least at or above the wüstite-magnetite (WM) oxygen buffer and near the fayalite-magnetite-quartz (FMQ) oxygen buffer. The amount of magnetite present in the mantle peridotite assemblage seems to correlate with tectonic setting and may be linked to fO₂ in the mantle. The timing of magnetite exsolution in olivine is still poorly understood and may depend on degree of partial melting, rate of cooling to ambient lithospheric temperature, or mantle metasomatic processes due to introduction of hydrous fluids.