Magnetic interactions at the origin of abnormal magnetic fabrics in lava flows: a case study from Kerguelen flood basalts

Abstract

International audienceAnisotropy of low-field magnetic susceptibility (AMS) of basaltic lava flows can give some clues about post-emplacement tilting occurring in volcanic sequences. Such a study has been carried out on a sequence of 19 successive lava flows from Kerguelen Archipelago. Surprisingly, two different patterns were observed. The first one - herein called normal fabric - is consistent with the lava-flow direction inferred from geological observations, whereas the second one observed for about 70 per cent of the analysed samples - herein called abnormal fabric - appears to be unrelated to the shear history of the lava flows during their emplacements. These abnormal fabrics are not strictly inverse fabric but seem to have a more complex origin. The unusual nature of these abnormal fabrics has been confirmed by lattice preferred orientation of plagioclases obtained from high-resolution electron backscattered diffraction measurements. The aim of this study is to propose a physical interpretation of these fabrics. A first step in our understanding was to clearly identify the nature and the magnetic properties of the Fe-Ti oxides assemblage. Thus, we present a comprehensive rock magnetic analysis relying on low- and high-temperature thermomagnetic curves (K-T), crossover diagrams, first-order reversal curves, ore-microscopy, Mössbauer measurements and electron backscattered images. We found that in the present case the abnormal fabrics are probably linked to a mixture of prevailing single domain (SD) strongly interacting grains population and a subsidiary multidomain grains population. SD grains are identified as small magnetite crystallized within the interstitial glass in globular aggregates along the silicate framework. Then a question arises: to what extent such SD interacting grains in globular aggregates can affect the magnetic susceptibility and its anisotropy? To answer this question, we developed a physical model in which the total energy density of an SD grain with an uniaxial anisotropy is minimized. In this model, the distribution function of the direction of the magnetic interaction field is comprised between two boundary states. It is either isotropic, which formally corresponds to the thermal demagnetized state, or it is ordered, which formally corresponds to the Alternative Field demagnetized state. We show that in both cases the degree of anisotropy decreases as the interaction field increases. Thus, we conclude that the abnormal fabric encountered in this study can be simply explained by strong magnetostatic interactions that would explain the large scattering of AMS often observed in basaltic lava flows

Magnetic fingerprint of southern Portuguese speleothems and implications for paleomagnetism and environmental magnetism

“…For all rock types, R values less than 0.5 indicate strong interaction of SD particles. However, it has been suggested that the R ratio is unable to differentiate between MD and interacting SD particles [ Fanjat et al ., ]. The values of R and Hcr are very similar in all SPA and SPAII samples.…”

“…The point of intersection of the two curves is close to the remanent coercive force (Hcr) and its projection on the abscissa represents the degree of interaction, R. For all rock types, R values less than 0.5 indicate strong interaction of SD particles. However, it has been suggested that the R ratio is unable to differentiate between MD and interacting SD particles [Fanjat et al, 2012]. The values of R and Hcr are very similar in all SPA and SPAII samples.…”

Crystal Mush Flow in Small Concentrically Expanded Pluton (Castle Crags Pluton; Klamath Mountains, CA, USA)

“…The studied samples are characterized by high magnetic susceptibility caused by magnetite content, which could lead to the occurrence of magnetostatic interactions that significantly alter the AMS (Cañón‐Tapia, ; Fanjat et al, ; Gaillot et al, ; Gregoire et al, ; Hargraves et al, ). Using the remanence‐based technique described by Petrovský et al (), we infer that in the CCG granites, the magnetostatic interaction has negligible influence on the magnetic rock properties (Figure S3c and Text S1, Henkel, ; Mayo, Bradbury, et al, ; Mayo, Erkkila, et al, ; Mayo, Ogrady, et al, ; Mayo, Ogrady, et al, ; Wohlfarth, ).…”

“…These high‐values are not typical for magmatic fabrics defined by alignment of noninteracting magnetite particles during the magmatic flow in ferromagnetic granites (Georgiev et al, ; Ježek et al, ; Tarling & Hrouda, ). Such anomalously high values have been previously explained either by (1) hematite content (Tarling & Hrouda, ), (2) magnetostatic interactions between adjacent magnetite grains (Cañón‐Tapia, ; Fanjat et al, ; Gaillot et al, ; Gregoire et al, ; Hargraves et al, ), or (3) subsolidus overprint of magmatic fabric (Georgiev et al, ; Henry et al, ; Hrouda & Lanza, ; Nouar et al, ). The detailed magnetic mineralogy study of the CCG rock types revealed that neither hematite nor magnetostatic interactions are responsible for the high P values (Text S1 and Figure S2).…”

Distinguishing diapirs from inflated plutons: an integrated rock magnetic fabric and structural study on the Roundstone Pluton, western Ireland

“…Several representative specimens from each facies were selected for detailed rock magnetic experiments to determine the dominant magnetic mineralogy of both facies and to identify or eliminate the possible presence of complicating features (e.g. Potter & Stephenson 1988;Rochette 1988;Wolff et al 1989;Hargraves et al 1991;Stephenson 1994;Cañón-Tapia 1996, 2001Aubourg & Robion 2002;Bouchez et al 2006;Gaillot et al 2006;Chadima et al 2009;Almqvist et al 2010;Fanjat et al 2012). …”