Dados da Submissão

Título

APPLYING PRINCIPAL COMPONENT ANALYSIS AND MAGNETIC COERCIVITY DISTRIBUTION TO ASSESS DIFFERENT DOMAIN STATES IN INTRUSIVE ROCKS

Texto do resumo

Understanding the domain magnetic state associated with the distribution of ferromagnetic particles in a mineral assembly is essential for paleomagnetic studies. The increase in size of magnetite crystals results in the progression from single domain (SD) to pseudo-single domain (PSD), or vortex, and multi-domain (MD) structures. Particles with PSD/SD magnetic domain structures are frequently found in volcanic and subvolcanic rocks and less frequent in plutonic rocks. They are the main subject of paleomagnetic investigations due to their stability in recording the primary components of magnetism. The higher stability of these crystals is primarily given by their higher coercivity values compared to MD-structured crystals. The considerable textural variety of plutonic rocks makes it difficult to define the relative contribution of a mixture of magnetic domain configurations. Discriminating stable primary magnetisation components in these rocks is often difficult due to the predominance of MD magnetite. This study focuses on characterising the magnetic mineralogy of plutonic and hypabyssal units to assess the potential influence of different domain states and their correlation with characteristic remanent magnetisation components. Multiple magnetic mineralogy experiments in both studied units lead to determining magnetite as the magnetic signal's main carrier. We used MAX UnMix software to model IRM acquisition curves and compared the findings to those obtained from a principal component analysis (PCA) based on diagrams produced with the FORCinel software. Both approaches were used to model 12 sites, 6 from plutonic and 6 from subvolcanic units. In the PCA method, the relative proportion is based on defining two end members (EMs) for each data set. The EMs show characteristic MD and PSD/SD behaviours for both rock types, with a greater contribution of SD magnetite particles associated with the subvolcanic unit. The models generated for all samples are best described with two components and exhibit coherent relative proportions in both methods. Based on that, we established a mean coercivity for each component using MAX UnMix. The plutonic rocks' lower coercivity component (C1) has a mean value of 15.35 mT, while a mean value of 46.06 mT defines the higher coercivity component (C2). In the subvolcanic unit, C1 is characterised by a mean coercivity of 15.05 mT and C2 by a mean value of 56.76 mT. These coercivity values agree with the proposed values in the literature for MD and PSD/SD magnetite particles, thus corroborating the qualitative EMs defined through the FORC diagrams. Our results demonstrate that an integrative approach can estimate the relative contribution of different domain states. Furthermore, by combining methods, we could determine the relative proportions of components with higher magnetic stability (PSD/SD). This suggests that the primary remanent magnetization in the analysed samples is likely recorded within the 30 mT to 80 mT coercivity range.

Palavras Chave

paleomagnetism; magnetic mineralogy; magnetic domain states; characteristic remanent magnetisation component.