Magnetism at Depth: A View from an Ancient Continental Subduction and Collision Zone

Abstract

Recent sophisticated global data compilations and magnetic surveys have been used to investigate the nature of magnetization in the lower crust and upper mantle. Two approaches to constraining magnetizations are developed, providing minimum (0.01 SI) and maximum (0.04 SI) susceptibility estimates, given some assumed thickness (15+ km here). These values are higher than are found in many continental rocks. Are there rocks deeper in the crust or upper mantle that are more magnetic than expected, or are the model assumptions incomplete? What is the magnetic behavior of deep‐crustal and upper mantle rocks, when slightly cooler than the Curie or Néel temperatures of their magnetic minerals, after being exhumed from locations of high‐grade metamorphism at greater depth? Different sets of equilibrium metamorphic minerals can be considered that would form under different conditions. Results on 1,501 samples from the Western Gneiss Region (WGR) Norway, mainly from mafic and ultramafic bodies subducted to depths of 60–200 km and temperatures of 750 up to 950°C at the very highest pressures, show that rocks did not fully equilibrate to the dominant metamorphic‐facies conditions. There is a large variation in petrophysical properties, oxide minerals, and mineral assemblages in WGR samples, though they cannot explain the broad high‐amplitude (deep‐seated) anomalies measured in this region. The presence of magnetite, and exsolved titanohematite and hemoilmenite in samples, shows those magnetic phases are preserved even at eclogite‐facies conditions, in part because complete eclogite‐facies equilibrium was rarely achieved.

Origin of Long‐Wavelength Magnetic Anomalies at Subduction Zones

“…For example, an extensive study of rocks from the Western Gneiss Region of Norway (McEnroe et al, ), which has a large satellite magnetic anomaly, revealed a wide range of metamorphism and magnetic properties that are mostly weaker than is required to explain the observed anomalies. McEnroe et al () placed bounds on susceptibilities for the lower crust and upper mantle of 0.01 and 0.04 SI assuming a 15‐km‐thick layer, which translates to a VIM of 6 to 24 kA. These are comparable with our values but too small for a magnetized dipping slab.…”

Magnetic Mineralogy and Petrophysical Properties of Ultramafic Rocks: Consequences for Crustal Magnetism

“…Early studies by Warner and Wasilewski (), Wasilewski and Warner (), and Wasilewski and Mayhew () report the lithospheric mantle to be too hot and too weakly magnetic to contribute to long‐wavelength anomalies. However, more recent studies propose a potential magnetic contribution from parts of the upper mantle and/or the lower crust (Blakely et al, ; Ferré et al, , ; Friedman et al, ; McEnroe et al, ).…”

“…Magnetite is usually considered the predominant magnetic mineral in the deep crust (e.g., Frost & Shive, ; Pilkington & Percival, ; Schlinger, ). However, the hematite‐ilmenite solid solution should also be considered as stable sources of remanent magnetization (Robinson et al, ) and as a source for short‐ and long‐wavelength anomalies (Brown & McEnroe, ; Kasama et al, ; Kletetschka & Stout, ; McEnroe & Brown, ; McEnroe et al, , , , , , , , ). The high stability and unblocking temperatures of exsolved phases in the hematite‐ilmenite system increase the remanent contribution to the total magnetization (McCammon et al, ; McEnroe & Brown, ; McEnroe et al, , , , , , ).…”

“…Commonly, samples from exhumed deep‐crustal rocks show evidence of a later lower P&T metamorphic overprint due to retrograde metamorphism (Strada et al, ). Identifying the magnetic phases that are stable at high P&T is challenging because many show disequilibrium microtextures that developed during the exhumation process (McEnroe et al, ). Equally challenging is obtaining a large suite of pristine deep‐crustal samples due to the ease with which these (commonly ultramafic) rocks alter in the presence of fluid.…”

“…However, the hematite-ilmenite solid solution should also be considered as stable sources of remanent magnetization and as a source for short-and long-wavelength anomalies (Brown & McEnroe, 2008;Kasama et al, 2004;Kletetschka & Stout, 1998;McEnroe & Brown, 2000;McEnroe et al, 2001aMcEnroe et al, , 2001bMcEnroe et al, , 2009McEnroe et al, , 2002McEnroe et al, , 2004McEnroe et al, , 2007McEnroe et al, , 2016McEnroe et al, , 2018. The high stability and unblocking temperatures of exsolved phases in the hematite-ilmenite system increase the remanent contribution to the total magnetization (McCammon et al, 2009;McEnroe & Brown, 2000;McEnroe et al, 2001aMcEnroe et al, , 2001cMcEnroe et al, , 2002McEnroe et al, , 2004McEnroe et al, , 2007McEnroe et al, , 2018. Chrome-spinel can also be stable at lower-crustal and upper-mantle levels (Barnes & Roeder, 2001;Wasilewski & Mayhew, 1992;Wasilewski et al, 1979); however, whether this phase is magnetic is controlled by the composition (Francombe, 1957;Ziemniak & Castelli, 2003).…”