TY - JOUR
AU - Trindade, Ricardo I, F
AB - SUMMARY Anisotropy of magnetic susceptibility is a petrofabric tool used to estimate the alignment of minerals at the site-scale, the imbrication between the magnetic foliation and the emplacement surface being an indicator of flow direction. However, despite numerous studies examining the flow direction in pyroclastic deposits and lava flows, the effect of magnetic mineralogy and the domain state of ferromagnetic phases on the magnetic fabric remains poorly understood. This paper describes the magnetic mineralogy and its influence on the magnetic fabric of Plio-Pleistocene lava flows and ignimbrites of the Caviahue-Copahue Volcanic Complex in the Andean Southern Volcanic Zone, Argentina. Rock magnetism, anisotropy of magnetic susceptibility and anhysteretic remanent magnetization and petrographic observations were performed on 30 sites of the volcanic complex. Results revealed the extrusive and pyroclastic rocks present varied magnetic mineralogy, formed in different stages of the magmatic evolution. Magnetic mineralogy variations strongly affect the anisotropy of magnetic susceptibility data in volcanic rocks and associated ignimbrites, providing ‘scattered’ fabrics when late Ti-rich titanomagnetite phases dominate the fabric, and ‘inverse’ or ‘intermediate’ fabrics when single-domain grains are present. ‘Normal’ fabrics are typically found when early crystallized pure magnetite is present. Our results highlight the complexity in the interpretation of magnetic anisotropy data in volcanic rocks and ignimbrites. AMS, AARM, Magnetic Mineralogy, Caviahue-Copahue Volcanic Complex 1 INTRODUCTION Volcanic environments, including the distribution of the caldera and associated volcanoes, has long been a focus of several studies (e.g. Geshi et al.2002; Cole et al.2005; Riel et al.2015). The study of volcanism and, especially large collapsing caldera events, provide insights into the mantle thermal flux and tectonomagmatic process (Ort et al.2013). The evolution of volcanic calderas usually involves the collapse of a volcanic edifice and the accumulation of extensive lava and ignimbrite deposits (Gudmundsson et al.2016). The relatively short timescale of this process often results in explosive volcanism and the development of Pyroclastic Density Currents (PDCs), which can be used in the identification of volcanic centres, assisting the reconstruction of volcanic environments and their geologic evolution. Located in the Andean Southern Volcanic Zone, the Caviahue-Copahue Volcanic Complex (CCVC), Argentina, represents a natural laboratory to the study of volcano-caldera complexes, given its structural complexity (Folguera et al.2016) and lithological diversity (Melnick et al.2006). Anisotropy of magnetic susceptibility (AMS) has been used as a petrofabric approach in a wide range of geological applications (e.g. Graham 1954; Hrouda 1982; MacDonald & Ellwood 1987; Rochette et al.1992; Kodama 1995; Borradaile & Henry 1997; Martín-Hernández et al.2004; Borradaile & Jackson 2010). In particular, AMS is used to evaluate flow direction in lava and pyroclastic flow deposits especially in volcanic systems with restricted outcrops (e.g. Cañón-Tapia 2004) and undefined stratigraphic relationships, in order to determine vent location. In lava flows, the AMS results from the preferential crystallographic or shape preferred orientation at the final stage of the flux and by crystal shear during lava emplacement (Cañón-Tapia et al.1996). However, several factors, such as viscosity and flux velocity control mineral alignment and distribution and, as a consequence, can directly influence the AMS signal. In ignimbrites, the AMS is strongly influenced by the transport, deposition and concentration of fragments. In the interior of PDCs, these fragments are preferably aligned along with the flux by interaction with other grains and the substrate, acquiring an imbrication that indicates the flow direction (Elston & Smith 1970). Several authors studied the control exerted by flow dynamics on the AMS signal of lavas and ignimbrites (e.g. Cañón-Tapia et al.1995; Cañón-Tapia 2004; LaBerge et al.2009; Ort et al.2013). However, comparatively little attention has been given to the role of magnetic mineralogy on the magnetic fabrics (e.g. Khan 1962; Wing-Fatt & Stacey 1966; Rochette et al.1999). Recent studies have shown that in addition to macroscopic factors related to the flow of lava and pyroclastic material, the AMS fabrics also show strong dependence on microscopic factors, such as composition, mineral magnetic interactions and domain structure of the Fe-Ti oxides, which most often control the AMS of volcanic rocks (Borradaile 1987; Cañón-Tapia 2004; Chadima et al.2009; Cañón-Tapia & Mendoza-Borunda 2014; Magee et al.2016; Wiegand et al.2017). For example, the ‘SD effect’, which is commonly associated to the inverse magnetic fabric of single-domain titanomagnetite grains (Stephenson et al.1986; Potter & Stephenson 1988), can be unveiled by correlating AMS data, magnetic mineralogy and measurements of the anisotropy of anhysteretic remanent magnetization (AARM). The rarity of observations of the ‘SD-effect’ may be caused by the fact that SD grains have a much smaller volume and thus contribute to the overall susceptibility (and its anisotropy) much less than MD grains. Also, correlation of AMS and AARM can indicate different carrier minerals, e.g. AMS may be dominated by MD magnetite while AARM by SD magnetite. In this paper, we present an AMS study including different volcanic units from CCVC. This work consists of an analysis of magnetic mineralogy control on the magnetic fabric and proposes some ways to reduce the ambiguity of interpretation on lava and pyroclastic flow directions. To attain this objective, we present the results of geology fieldwork, petrography and scanning electron microscopy observations, AMS and AARM, and a detailed examination of the magnetic mineralogy of the different volcanic units along with the volcanic complex. The CVCC shows a wide compositional range, from basalts to rhyolites and displays both explosive (ignimbrites) and effusive (lava flows) deposits, making the site an excellent environment to evaluate petrofabric studies from AMS data. 2 GEOLOGICAL SETTING AND SAMPLING 2.1 Local geology and stratigraphy The Caviahue-Copahue Volcanic Complex (CCVC) is located in the Andean southern volcanic zone (SVZ: 33.3°S–46°S) along the Chile-Argentina border (Fig. 1). It is positioned about 30 km from the current volcanic arc and, in these latitudes, volcanism is associated with the oblique subduction of the Nazca Plate under the South American Plate. Several studies suggest compositional variations for the evolution of the CCVC, from basaltic andesites to rhyolites, with characteristics of medium to high-K calc-alkaline and metaluminous series (Mazzoni & Licitra 2000; Varekamp et al.2006). Figure 1. Open in new tabDownload slide (a) Location map and main regional structures of the Andean intra-arc zone with a detail of southern volcanic zone (SVZ); (b) Geological map and stratigraphic chart for CCVC (Modified from Melnick et al.2006), where the black dots represent all 30 sampling sites. Figure 1. Open in new tabDownload slide (a) Location map and main regional structures of the Andean intra-arc zone with a detail of southern volcanic zone (SVZ); (b) Geological map and stratigraphic chart for CCVC (Modified from Melnick et al.2006), where the black dots represent all 30 sampling sites. Stratigraphy of the CCVC is composed of nine volcanic units (Fig. 1b). The eruptive sequence begins with effusive and pyroclastic events under an extensional tectonic regime (Folguera et al.2003), whose deposits are grouped in the Cola de Zorro Formation, also described in the literature as the Hualcupén Formation (Pesce 1989). This unit is characterized by sub-horizontal basalt and andesite lava flow deposits intercalated with non-welded ignimbrites of andesitic composition (Fig. 2a) which covers a wide segment of the studied area (Fig. 1b). Figure 2. Open in new tabDownload slide Field aspects of CCVC units. (a) scoriaceous basal portion of the Cola de Zorro basaltic lava flows; (b) subvertical joints in the lower lavas of the Las Mellizas Volcanic Sequence; (c) welded ignimbrite of the Las Mellizas Volcanic Sequence; (d) rheomorphic ignimbrite of the Las Mellizas Volcanic sequence showing parataxitic texture and lithic fragments; (e) geometry and subvertical joints in the upper lavas of the Las Mellizas Volcanic Sequence; (f) no welded andesitic ignimbrite of the Riscos Bayos unit. Figure 2. Open in new tabDownload slide Field aspects of CCVC units. (a) scoriaceous basal portion of the Cola de Zorro basaltic lava flows; (b) subvertical joints in the lower lavas of the Las Mellizas Volcanic Sequence; (c) welded ignimbrite of the Las Mellizas Volcanic Sequence; (d) rheomorphic ignimbrite of the Las Mellizas Volcanic sequence showing parataxitic texture and lithic fragments; (e) geometry and subvertical joints in the upper lavas of the Las Mellizas Volcanic Sequence; (f) no welded andesitic ignimbrite of the Riscos Bayos unit. A volcano-tectonic depression of approximately 20 × 15 km and 0.5 km deep (Fig. 1b) was developed during the Upper Pleistocene Pliocene (ca. 2 Ma; Linares et al.1999; Folguera & Ramos 2000) and was named Caviahue Depression (Groeber 1925) or Caldera del Agrio (Pesce 1989). Based on the structural analysis and previous works (e.g. Folguera & Ramos 2000; Melnick et al.2006), Sruoga & Consoli (2011) recommended the abandonment of the term caldera, advocating for a structural origin for the Caviahue Depression. In contrast, other works support the volcanic origin based on the ignimbrite presence both inside and outside of the Caviahue Depression (e.g. Pesce 1989; Mazzoni & Licitra 2000), favouring, therefore, the caldera hypothesis. In this depression, two main units were recognized: Las Mellizas Formation and Trolope Lavas (Pesce 1989; Melnick et al.2006). The Las Mellizas Formation is characterized by two effusive events (Lower and Upper Lavas) intercalated by an explosive one (Las Mellizas Ignimbrites). The Lower Lavas unit is represented by remarkably columnar-jointed lavas of basaltic to andesitic composition (Fig. 2b) interbedded with some sedimentary deposits (Melnick et al.2006). Las Mellizas Ignimbrites are characterized by andesitic to dacitic compositions presenting a widespread distribution inside the Caviahue Depression (Mazzoni & Licitra 2000). This pyroclastic sequence is composed of high-grade ignimbrites with vertical lithofacies variation, including basal vitrophyres with gradations from lithic clast-bearing deposits to rheomorphic units with eutaxitic (Fig. 2c) and parataxitic structures (Fig. 2d) (Mazzoni & Licitra 2000). The Upper Lavas sequence (Fig. 2e) overlie the ignimbrites and is partially covered by flows from the Copahue Volcano. This unit is essentially composed of andesitic lava flows with sub-horizontal foliation and homogenous thicknesses and textures (Melnick et al.2006). Located northeast of the studied area and filling the northeastern entrance canyon of the Caviahue Depression, the Trolope Lavas form a 200 m thick succession of homogeneous trachyandesitic lava flows and breccias (Melnick et al.2006). Almost 15 km outside the Caviahue Depression border, a thin cover of dense pyroclastic current deposit forms the Riscos Bayos Formation. This irregularly distributed unit has been often interpreted as the outflow ignimbrites associated with the putative caldera collapse (Mazzoni & Licitra 2000). This deposit is composed of non-welded ignimbrites (Fig. 2f) with andesitic (lower and intermediate units) and rhyolitic (upper unit) compositions with observable internal structures marked by the concentration of lithics in the base and of pumices in the top of the deposits (Mazzoni & Licitra 2000). The Copahue Volcano, an active volcano of basaltic andesitic composition, occupies the southwestern corner of the Caviahue Depression. Copahue evolution is marked by successive effusive and pyroclastic (subordinate) events from a central conduit and related lateral fissures, with the units divided into pre-, sin- and postglacial phases (Melnick et al.2006; Sruoga P 2011). The historical deposits of the eruptive cycles of 1992, 1995, 2000 and 2012 mainly involved ash, bombs and lithic fragments, besides rare lahars (Melnick et al.2006). The Pucón Mahuida and Cerro Bayo domes present rhyolitic to dacitic compositions and are composed of lava flows and subvolcanic bodies, with the first one located on the southeastern slope of the Copahue volcano and the second located in the northern edge of the Caviahue (Folguera & Ramos 2000; Melnick et al.2006). K-Ar ages are available for all the studied units (Muñoz & Stern 1988; Linares et al.1999), although some authors point out possible errors in the analyses (e.g. Melnick et al.2006; Folguera et al.2016) The volcanic stage of the Cola de Zorro Formation is Early Pliocene (5.67 ± 0.14 Ma to 4.00 ± 0.10 Ma) according to available K-Ar ages obtained from the inner wall of the Caviahue Depression. A Late Pliocene age for Las Mellizas sequence is given by three K-Ar ages (2.68 ± 0.14 Ma to 2.60 ± 0.10 Ma) (Linares et al.1999). The age of Riscos Bayos ignimbrite is less well constrained, with contrasting K-Ar results obtained by Linares et al. (1999) (2.05 ± 0.10 Ma) and Muñoz & Stern (1988) (1.1 ± 0.5 Ma). A Pleistocene age for the Trolope Lavas is provided by four K-Ar ages between 1.63 ± 0.10 and 0.82 ± 0.16 Ma with the following development of the Copahue volcano beginning about 1.2 Ma (Linares et al.1999). K-Ar ages of 1.23 to 0.6 Ma were obtained from the Pucón Mahuida and Cerro Bayo domes (Muñoz & Stern 1988; Linares et al.1999). 2.2 Sampling In the fieldwork, a total of 239 oriented cylindrical cores were collected from 30 sites (Fig. 1b) using a portable, gasoline-powered drill. All samples were oriented using a magnetic compass, and when possible sun compass. The magnetic declination was systematically 03°W. Sampled sites with location are reported in Table 1. Sites COP01-COP05 and COP20 were collected from Cola de Zorro Formation, and correspond to sub-horizontal lava flow at the base of the stratigraphy. Sites COP23-COP27 and COP29-COP30 were collected on the Riscos Bayos ignimbrite (non-welded ignimbrites). From Las Mellizas Volcanic Sequence, we collected from Lower lavas (sites COP06, COP19, COP28 and COP33), welded ignimbrites (sites COP07-COP08, COP11-COP14, COP22 and COP32) and Upper lavas (sites COP09-COP10, COP16-COP17 and COP21). At the laboratory, all cylindrical samples were cut into specimens of 2.5 cm diameter and 2.2 cm height for AMS and AARM measurements, totalizing 430 specimens. From each site, we also separated some material for magnetic mineralogy experiments, petrography and scanning electron microscopy analyses. Table 1. AMS data from the studied sites. N = number of specimens measured; Km = mean susceptibility; L = lineation; F = foliation; P′ = corrected degree of anisotropy; T = Jelineck’s parameter (Jelinek 1981); D = declination; I = inclination; E1 − 2, E2 − 3, E1 − 3: semi-angles of the 95 per cent confidence ellipses around the principal susceptibility axes, values in parentheses are standard deviation. *sites with AARM measurements. Tc is the estimated Curie temperature. Site ID . UTM . n . Km . L . F . P’ . T . K1 . K3 . E1 − 2 . E2 − 3 . E1 − 3 . Tc (°C) . . E . N . . (10−2 SI) . . . . . D, I . D, I . (°) . (°) . (°) . . Las Mellizas - Upper Lavas COP09* 319698 5803947 21 1.18 (0.10) 1.004 (0.001) 1.006 (0.001) 1.010 (0.002) 0.170 (0.161) 234.80 93.8 9 6 6 533 COP10 319005 5803952 16 2.37 (0.22) 1.009 (0.002) 1.006 (0.003) 1.016 (0.004) -0.193 (0.147) 7.12 220. 76 11 8 10 575 COP16* 316909 5810063 12 1.10 (0.18) 1.001 (0.003) 1.003 (0.002) 1.005 (0.004) 0.410 (0.326) 185.27 309.47 50 29 30 178 COP17 316909 5810063 20 1.48 (0.12) 1.004 (0.001) 1.007 (0.002) 1.011 (0.002) 0.304 (0.202) 259.21 68.69 7 4 5 587 COP21 321530 5811343 7 1.12 (0.10) 1.008 (0.002) 1.010 (0.001) 1.018 (0.002) 0.127 (0.108) 101.82 240. 6 13 6 8 400 Las Mellizas - Ignimbrites COP07* 318723 5804683 12 1.18 (0.12) 1.002 (0.001) 1.009 (0.004) 1.012 (0.004) 0.598 (0.305) 247.18 62.72 22 4 6 565 COP08* 321804 5803438 16 1.32 (0.14) 1.001 (0.001) 1.004 (0.001) 1.005 (0.001) 0.739 (0.250) 112.5 223.77 34 6 6 564 COP11* 319460 5803932 15 1.20 (0.15) 1.005 (0.002) 1.004 (0.003) 1.009 (0.002) -0.111 (0.452) 179.36 341.52 21 12 6 562 COP12* 323911 5804243 17 2.67 (0.33) 1.001 (0.001) 1.009 (0.002) 1.011 (0.002) 0.840 (0.224) 119.3 17.74 39 7 7 537 COP13* 315966 5812495 14 1.36 (0.17) 1.001 (0.002) 1.010 (0.004) 1.012 (0.004) 0.822 (0.293) 184.21 22.68 56 7 12 444 COP14* 316149 5810346 7 1.52 (0.17) 1.005 (0.002) 1.011 (0.002) 1.016 (0.003) 0.372 (0.138) 236.59 327.1 18 6 13 430 COP22* 318068 5804710 15 1.49 (0.14) 1.002 (0.001) 1.006 (0.001) 1.009 (0.002) 0.500 (0.227) 288.6 174.77 5 19 6 571 COP32* 325305 5804519 19 1.50 (0.07) 1.001 (0.001) 1.014 (0.003) 1.017 (0.003) 0.839 (0.137) 329.14 94.67 38 5 6 568 Las Mellizas - Lower Lavas COP06 318396 5805231 13 1.44 (0.12) 1.001 (0.004) 1.003 (0.001) 1.004 (0.004) 0.333 (0.369) 82.13 177.21 53 15 15 158 COP19 326776 5811393 10 1.58 (0.05) 1.005 (0.003) 1.002 (0.006) 1.007 (0.007) -0.433 (0.324) 64.45 313.19 51 27 44 189 COP28* 320900 5807231 23 2.57 (0.22) 1.017 (0.003) 1.004 (0.002) 1.023 (0.003) -0.609 (0.154) 249.3 152.65 17 7 15 230 COP33 318230 5807530 24 1.24 (0.09) 1.021 (0.003) 1.005 (0.003) 1.028 (0.003) -0.640 (0.188) 276.11 124.78 9 4 9 191 Riscos Bayos - Ignimbrites COP23* 341855 5797063 18 0.70 (0.04) 1.003 (0.002) 1.010 (0.006) 1.013 (0.007) 0.581 (0.148) 245.3 145.75 23 7 7 597 COP24 341855 5797063 8 0.35 (0.01) 1.004 (0.002) 1.002 (0.003) 1.006 (0.003) -0.305 (0.412) 330.61 89.15 35 24 32 585 COP25 343248 5797194 10 0.33 (0.05) 1.009 (0.005) 1.018 (0.005) 1.027 (0.006) 0.353 (0.280) 301.18 98.71 21 9 11 432 COP26 342200 5797020 13 0.25 (0.07) 1.002 (0.003) 1.003 (0.004) 1.005 (0.005) 0.302 (0.459) 312.49 122.41 58 27 27 280 COP27 342191 5796983 8 0.65 (0.32) 1.004 (0.005) 1.002 (0.005) 1.006 (0.009) -0.191 (0.506) 298.33 95.54 43 25 34 584 COP29* 339110 5797170 12 0.42 (0.02) 1.009 (0.003) 1.010 (0.002) 1.019 (0.003) 0.065 (0.183) 34.32 123.58 11 7 10 584 COP30 397107 5797178 14 0.34 (0.06) 1.003 (0.001) 1.004 (0.002) 1.007 (0.002) 0.164 (0.316) 311.22 146.68 16 9 12 530 Cola de Zorro - Lavas COP01 338971 5797984 19 1.79 (0.13) 1.000 (0.001) 1.018 (0.002) 1.021 (0.002) 0.971 (0.095) 128.10 21.59 79 5 10 585 COP02* 338663 5797976 12 1.80 (0.11) 1.016 (0.007) 1.009 (0.004) 1.026 (0.009) -0.261 (0.183) 77.28 202.48 9 4 7 574 COP03* 338201 5798069 15 1.25 (0.14) 1.010 (0.002) 1.014 (0.004) 1.025 (0.004) 0.169 (0.167) 212.5 4.85 15 9 12 586 COP04* 338158 5798078 12 1.03 (0.14) 1.002 (0.003) 1.002 (0.003) 1.004 (0.003) -0.157 (0.398) 359.21 96.16 49 36 47 190 COP05 337627 5798182 17 1.99 (0.31) 1.012 (0.007) 1.002 (0.005) 1.015 (0.005) -0.742 (0.462) 248.14 60.76 44 14 40 313 COP20 330723 5812160 11 3.52 (0.53) 1.017 (0.004) 1.018 (0.004) 1.035 (0.005) 0.035 (0.184) 268.12 160.55 9 7 9 587 Site ID . UTM . n . Km . L . F . P’ . T . K1 . K3 . E1 − 2 . E2 − 3 . E1 − 3 . Tc (°C) . . E . N . . (10−2 SI) . . . . . D, I . D, I . (°) . (°) . (°) . . Las Mellizas - Upper Lavas COP09* 319698 5803947 21 1.18 (0.10) 1.004 (0.001) 1.006 (0.001) 1.010 (0.002) 0.170 (0.161) 234.80 93.8 9 6 6 533 COP10 319005 5803952 16 2.37 (0.22) 1.009 (0.002) 1.006 (0.003) 1.016 (0.004) -0.193 (0.147) 7.12 220. 76 11 8 10 575 COP16* 316909 5810063 12 1.10 (0.18) 1.001 (0.003) 1.003 (0.002) 1.005 (0.004) 0.410 (0.326) 185.27 309.47 50 29 30 178 COP17 316909 5810063 20 1.48 (0.12) 1.004 (0.001) 1.007 (0.002) 1.011 (0.002) 0.304 (0.202) 259.21 68.69 7 4 5 587 COP21 321530 5811343 7 1.12 (0.10) 1.008 (0.002) 1.010 (0.001) 1.018 (0.002) 0.127 (0.108) 101.82 240. 6 13 6 8 400 Las Mellizas - Ignimbrites COP07* 318723 5804683 12 1.18 (0.12) 1.002 (0.001) 1.009 (0.004) 1.012 (0.004) 0.598 (0.305) 247.18 62.72 22 4 6 565 COP08* 321804 5803438 16 1.32 (0.14) 1.001 (0.001) 1.004 (0.001) 1.005 (0.001) 0.739 (0.250) 112.5 223.77 34 6 6 564 COP11* 319460 5803932 15 1.20 (0.15) 1.005 (0.002) 1.004 (0.003) 1.009 (0.002) -0.111 (0.452) 179.36 341.52 21 12 6 562 COP12* 323911 5804243 17 2.67 (0.33) 1.001 (0.001) 1.009 (0.002) 1.011 (0.002) 0.840 (0.224) 119.3 17.74 39 7 7 537 COP13* 315966 5812495 14 1.36 (0.17) 1.001 (0.002) 1.010 (0.004) 1.012 (0.004) 0.822 (0.293) 184.21 22.68 56 7 12 444 COP14* 316149 5810346 7 1.52 (0.17) 1.005 (0.002) 1.011 (0.002) 1.016 (0.003) 0.372 (0.138) 236.59 327.1 18 6 13 430 COP22* 318068 5804710 15 1.49 (0.14) 1.002 (0.001) 1.006 (0.001) 1.009 (0.002) 0.500 (0.227) 288.6 174.77 5 19 6 571 COP32* 325305 5804519 19 1.50 (0.07) 1.001 (0.001) 1.014 (0.003) 1.017 (0.003) 0.839 (0.137) 329.14 94.67 38 5 6 568 Las Mellizas - Lower Lavas COP06 318396 5805231 13 1.44 (0.12) 1.001 (0.004) 1.003 (0.001) 1.004 (0.004) 0.333 (0.369) 82.13 177.21 53 15 15 158 COP19 326776 5811393 10 1.58 (0.05) 1.005 (0.003) 1.002 (0.006) 1.007 (0.007) -0.433 (0.324) 64.45 313.19 51 27 44 189 COP28* 320900 5807231 23 2.57 (0.22) 1.017 (0.003) 1.004 (0.002) 1.023 (0.003) -0.609 (0.154) 249.3 152.65 17 7 15 230 COP33 318230 5807530 24 1.24 (0.09) 1.021 (0.003) 1.005 (0.003) 1.028 (0.003) -0.640 (0.188) 276.11 124.78 9 4 9 191 Riscos Bayos - Ignimbrites COP23* 341855 5797063 18 0.70 (0.04) 1.003 (0.002) 1.010 (0.006) 1.013 (0.007) 0.581 (0.148) 245.3 145.75 23 7 7 597 COP24 341855 5797063 8 0.35 (0.01) 1.004 (0.002) 1.002 (0.003) 1.006 (0.003) -0.305 (0.412) 330.61 89.15 35 24 32 585 COP25 343248 5797194 10 0.33 (0.05) 1.009 (0.005) 1.018 (0.005) 1.027 (0.006) 0.353 (0.280) 301.18 98.71 21 9 11 432 COP26 342200 5797020 13 0.25 (0.07) 1.002 (0.003) 1.003 (0.004) 1.005 (0.005) 0.302 (0.459) 312.49 122.41 58 27 27 280 COP27 342191 5796983 8 0.65 (0.32) 1.004 (0.005) 1.002 (0.005) 1.006 (0.009) -0.191 (0.506) 298.33 95.54 43 25 34 584 COP29* 339110 5797170 12 0.42 (0.02) 1.009 (0.003) 1.010 (0.002) 1.019 (0.003) 0.065 (0.183) 34.32 123.58 11 7 10 584 COP30 397107 5797178 14 0.34 (0.06) 1.003 (0.001) 1.004 (0.002) 1.007 (0.002) 0.164 (0.316) 311.22 146.68 16 9 12 530 Cola de Zorro - Lavas COP01 338971 5797984 19 1.79 (0.13) 1.000 (0.001) 1.018 (0.002) 1.021 (0.002) 0.971 (0.095) 128.10 21.59 79 5 10 585 COP02* 338663 5797976 12 1.80 (0.11) 1.016 (0.007) 1.009 (0.004) 1.026 (0.009) -0.261 (0.183) 77.28 202.48 9 4 7 574 COP03* 338201 5798069 15 1.25 (0.14) 1.010 (0.002) 1.014 (0.004) 1.025 (0.004) 0.169 (0.167) 212.5 4.85 15 9 12 586 COP04* 338158 5798078 12 1.03 (0.14) 1.002 (0.003) 1.002 (0.003) 1.004 (0.003) -0.157 (0.398) 359.21 96.16 49 36 47 190 COP05 337627 5798182 17 1.99 (0.31) 1.012 (0.007) 1.002 (0.005) 1.015 (0.005) -0.742 (0.462) 248.14 60.76 44 14 40 313 COP20 330723 5812160 11 3.52 (0.53) 1.017 (0.004) 1.018 (0.004) 1.035 (0.005) 0.035 (0.184) 268.12 160.55 9 7 9 587 Open in new tab Table 1. AMS data from the studied sites. N = number of specimens measured; Km = mean susceptibility; L = lineation; F = foliation; P′ = corrected degree of anisotropy; T = Jelineck’s parameter (Jelinek 1981); D = declination; I = inclination; E1 − 2, E2 − 3, E1 − 3: semi-angles of the 95 per cent confidence ellipses around the principal susceptibility axes, values in parentheses are standard deviation. *sites with AARM measurements. Tc is the estimated Curie temperature. Site ID . UTM . n . Km . L . F . P’ . T . K1 . K3 . E1 − 2 . E2 − 3 . E1 − 3 . Tc (°C) . . E . N . . (10−2 SI) . . . . . D, I . D, I . (°) . (°) . (°) . . Las Mellizas - Upper Lavas COP09* 319698 5803947 21 1.18 (0.10) 1.004 (0.001) 1.006 (0.001) 1.010 (0.002) 0.170 (0.161) 234.80 93.8 9 6 6 533 COP10 319005 5803952 16 2.37 (0.22) 1.009 (0.002) 1.006 (0.003) 1.016 (0.004) -0.193 (0.147) 7.12 220. 76 11 8 10 575 COP16* 316909 5810063 12 1.10 (0.18) 1.001 (0.003) 1.003 (0.002) 1.005 (0.004) 0.410 (0.326) 185.27 309.47 50 29 30 178 COP17 316909 5810063 20 1.48 (0.12) 1.004 (0.001) 1.007 (0.002) 1.011 (0.002) 0.304 (0.202) 259.21 68.69 7 4 5 587 COP21 321530 5811343 7 1.12 (0.10) 1.008 (0.002) 1.010 (0.001) 1.018 (0.002) 0.127 (0.108) 101.82 240. 6 13 6 8 400 Las Mellizas - Ignimbrites COP07* 318723 5804683 12 1.18 (0.12) 1.002 (0.001) 1.009 (0.004) 1.012 (0.004) 0.598 (0.305) 247.18 62.72 22 4 6 565 COP08* 321804 5803438 16 1.32 (0.14) 1.001 (0.001) 1.004 (0.001) 1.005 (0.001) 0.739 (0.250) 112.5 223.77 34 6 6 564 COP11* 319460 5803932 15 1.20 (0.15) 1.005 (0.002) 1.004 (0.003) 1.009 (0.002) -0.111 (0.452) 179.36 341.52 21 12 6 562 COP12* 323911 5804243 17 2.67 (0.33) 1.001 (0.001) 1.009 (0.002) 1.011 (0.002) 0.840 (0.224) 119.3 17.74 39 7 7 537 COP13* 315966 5812495 14 1.36 (0.17) 1.001 (0.002) 1.010 (0.004) 1.012 (0.004) 0.822 (0.293) 184.21 22.68 56 7 12 444 COP14* 316149 5810346 7 1.52 (0.17) 1.005 (0.002) 1.011 (0.002) 1.016 (0.003) 0.372 (0.138) 236.59 327.1 18 6 13 430 COP22* 318068 5804710 15 1.49 (0.14) 1.002 (0.001) 1.006 (0.001) 1.009 (0.002) 0.500 (0.227) 288.6 174.77 5 19 6 571 COP32* 325305 5804519 19 1.50 (0.07) 1.001 (0.001) 1.014 (0.003) 1.017 (0.003) 0.839 (0.137) 329.14 94.67 38 5 6 568 Las Mellizas - Lower Lavas COP06 318396 5805231 13 1.44 (0.12) 1.001 (0.004) 1.003 (0.001) 1.004 (0.004) 0.333 (0.369) 82.13 177.21 53 15 15 158 COP19 326776 5811393 10 1.58 (0.05) 1.005 (0.003) 1.002 (0.006) 1.007 (0.007) -0.433 (0.324) 64.45 313.19 51 27 44 189 COP28* 320900 5807231 23 2.57 (0.22) 1.017 (0.003) 1.004 (0.002) 1.023 (0.003) -0.609 (0.154) 249.3 152.65 17 7 15 230 COP33 318230 5807530 24 1.24 (0.09) 1.021 (0.003) 1.005 (0.003) 1.028 (0.003) -0.640 (0.188) 276.11 124.78 9 4 9 191 Riscos Bayos - Ignimbrites COP23* 341855 5797063 18 0.70 (0.04) 1.003 (0.002) 1.010 (0.006) 1.013 (0.007) 0.581 (0.148) 245.3 145.75 23 7 7 597 COP24 341855 5797063 8 0.35 (0.01) 1.004 (0.002) 1.002 (0.003) 1.006 (0.003) -0.305 (0.412) 330.61 89.15 35 24 32 585 COP25 343248 5797194 10 0.33 (0.05) 1.009 (0.005) 1.018 (0.005) 1.027 (0.006) 0.353 (0.280) 301.18 98.71 21 9 11 432 COP26 342200 5797020 13 0.25 (0.07) 1.002 (0.003) 1.003 (0.004) 1.005 (0.005) 0.302 (0.459) 312.49 122.41 58 27 27 280 COP27 342191 5796983 8 0.65 (0.32) 1.004 (0.005) 1.002 (0.005) 1.006 (0.009) -0.191 (0.506) 298.33 95.54 43 25 34 584 COP29* 339110 5797170 12 0.42 (0.02) 1.009 (0.003) 1.010 (0.002) 1.019 (0.003) 0.065 (0.183) 34.32 123.58 11 7 10 584 COP30 397107 5797178 14 0.34 (0.06) 1.003 (0.001) 1.004 (0.002) 1.007 (0.002) 0.164 (0.316) 311.22 146.68 16 9 12 530 Cola de Zorro - Lavas COP01 338971 5797984 19 1.79 (0.13) 1.000 (0.001) 1.018 (0.002) 1.021 (0.002) 0.971 (0.095) 128.10 21.59 79 5 10 585 COP02* 338663 5797976 12 1.80 (0.11) 1.016 (0.007) 1.009 (0.004) 1.026 (0.009) -0.261 (0.183) 77.28 202.48 9 4 7 574 COP03* 338201 5798069 15 1.25 (0.14) 1.010 (0.002) 1.014 (0.004) 1.025 (0.004) 0.169 (0.167) 212.5 4.85 15 9 12 586 COP04* 338158 5798078 12 1.03 (0.14) 1.002 (0.003) 1.002 (0.003) 1.004 (0.003) -0.157 (0.398) 359.21 96.16 49 36 47 190 COP05 337627 5798182 17 1.99 (0.31) 1.012 (0.007) 1.002 (0.005) 1.015 (0.005) -0.742 (0.462) 248.14 60.76 44 14 40 313 COP20 330723 5812160 11 3.52 (0.53) 1.017 (0.004) 1.018 (0.004) 1.035 (0.005) 0.035 (0.184) 268.12 160.55 9 7 9 587 Site ID . UTM . n . Km . L . F . P’ . T . K1 . K3 . E1 − 2 . E2 − 3 . E1 − 3 . Tc (°C) . . E . N . . (10−2 SI) . . . . . D, I . D, I . (°) . (°) . (°) . . Las Mellizas - Upper Lavas COP09* 319698 5803947 21 1.18 (0.10) 1.004 (0.001) 1.006 (0.001) 1.010 (0.002) 0.170 (0.161) 234.80 93.8 9 6 6 533 COP10 319005 5803952 16 2.37 (0.22) 1.009 (0.002) 1.006 (0.003) 1.016 (0.004) -0.193 (0.147) 7.12 220. 76 11 8 10 575 COP16* 316909 5810063 12 1.10 (0.18) 1.001 (0.003) 1.003 (0.002) 1.005 (0.004) 0.410 (0.326) 185.27 309.47 50 29 30 178 COP17 316909 5810063 20 1.48 (0.12) 1.004 (0.001) 1.007 (0.002) 1.011 (0.002) 0.304 (0.202) 259.21 68.69 7 4 5 587 COP21 321530 5811343 7 1.12 (0.10) 1.008 (0.002) 1.010 (0.001) 1.018 (0.002) 0.127 (0.108) 101.82 240. 6 13 6 8 400 Las Mellizas - Ignimbrites COP07* 318723 5804683 12 1.18 (0.12) 1.002 (0.001) 1.009 (0.004) 1.012 (0.004) 0.598 (0.305) 247.18 62.72 22 4 6 565 COP08* 321804 5803438 16 1.32 (0.14) 1.001 (0.001) 1.004 (0.001) 1.005 (0.001) 0.739 (0.250) 112.5 223.77 34 6 6 564 COP11* 319460 5803932 15 1.20 (0.15) 1.005 (0.002) 1.004 (0.003) 1.009 (0.002) -0.111 (0.452) 179.36 341.52 21 12 6 562 COP12* 323911 5804243 17 2.67 (0.33) 1.001 (0.001) 1.009 (0.002) 1.011 (0.002) 0.840 (0.224) 119.3 17.74 39 7 7 537 COP13* 315966 5812495 14 1.36 (0.17) 1.001 (0.002) 1.010 (0.004) 1.012 (0.004) 0.822 (0.293) 184.21 22.68 56 7 12 444 COP14* 316149 5810346 7 1.52 (0.17) 1.005 (0.002) 1.011 (0.002) 1.016 (0.003) 0.372 (0.138) 236.59 327.1 18 6 13 430 COP22* 318068 5804710 15 1.49 (0.14) 1.002 (0.001) 1.006 (0.001) 1.009 (0.002) 0.500 (0.227) 288.6 174.77 5 19 6 571 COP32* 325305 5804519 19 1.50 (0.07) 1.001 (0.001) 1.014 (0.003) 1.017 (0.003) 0.839 (0.137) 329.14 94.67 38 5 6 568 Las Mellizas - Lower Lavas COP06 318396 5805231 13 1.44 (0.12) 1.001 (0.004) 1.003 (0.001) 1.004 (0.004) 0.333 (0.369) 82.13 177.21 53 15 15 158 COP19 326776 5811393 10 1.58 (0.05) 1.005 (0.003) 1.002 (0.006) 1.007 (0.007) -0.433 (0.324) 64.45 313.19 51 27 44 189 COP28* 320900 5807231 23 2.57 (0.22) 1.017 (0.003) 1.004 (0.002) 1.023 (0.003) -0.609 (0.154) 249.3 152.65 17 7 15 230 COP33 318230 5807530 24 1.24 (0.09) 1.021 (0.003) 1.005 (0.003) 1.028 (0.003) -0.640 (0.188) 276.11 124.78 9 4 9 191 Riscos Bayos - Ignimbrites COP23* 341855 5797063 18 0.70 (0.04) 1.003 (0.002) 1.010 (0.006) 1.013 (0.007) 0.581 (0.148) 245.3 145.75 23 7 7 597 COP24 341855 5797063 8 0.35 (0.01) 1.004 (0.002) 1.002 (0.003) 1.006 (0.003) -0.305 (0.412) 330.61 89.15 35 24 32 585 COP25 343248 5797194 10 0.33 (0.05) 1.009 (0.005) 1.018 (0.005) 1.027 (0.006) 0.353 (0.280) 301.18 98.71 21 9 11 432 COP26 342200 5797020 13 0.25 (0.07) 1.002 (0.003) 1.003 (0.004) 1.005 (0.005) 0.302 (0.459) 312.49 122.41 58 27 27 280 COP27 342191 5796983 8 0.65 (0.32) 1.004 (0.005) 1.002 (0.005) 1.006 (0.009) -0.191 (0.506) 298.33 95.54 43 25 34 584 COP29* 339110 5797170 12 0.42 (0.02) 1.009 (0.003) 1.010 (0.002) 1.019 (0.003) 0.065 (0.183) 34.32 123.58 11 7 10 584 COP30 397107 5797178 14 0.34 (0.06) 1.003 (0.001) 1.004 (0.002) 1.007 (0.002) 0.164 (0.316) 311.22 146.68 16 9 12 530 Cola de Zorro - Lavas COP01 338971 5797984 19 1.79 (0.13) 1.000 (0.001) 1.018 (0.002) 1.021 (0.002) 0.971 (0.095) 128.10 21.59 79 5 10 585 COP02* 338663 5797976 12 1.80 (0.11) 1.016 (0.007) 1.009 (0.004) 1.026 (0.009) -0.261 (0.183) 77.28 202.48 9 4 7 574 COP03* 338201 5798069 15 1.25 (0.14) 1.010 (0.002) 1.014 (0.004) 1.025 (0.004) 0.169 (0.167) 212.5 4.85 15 9 12 586 COP04* 338158 5798078 12 1.03 (0.14) 1.002 (0.003) 1.002 (0.003) 1.004 (0.003) -0.157 (0.398) 359.21 96.16 49 36 47 190 COP05 337627 5798182 17 1.99 (0.31) 1.012 (0.007) 1.002 (0.005) 1.015 (0.005) -0.742 (0.462) 248.14 60.76 44 14 40 313 COP20 330723 5812160 11 3.52 (0.53) 1.017 (0.004) 1.018 (0.004) 1.035 (0.005) 0.035 (0.184) 268.12 160.55 9 7 9 587 Open in new tab 3 METHODS 3.1 Petrography and scanning electron microscopy (SEM) observations We performed petrographic, directional and scanning electron microscope (SEM) analysis in six representative samples of the studied sites. Petrographic analysis was performed at the Laboratório de Paleomagnetismo e Magnetismo de Rochas, Universidade de São Paulo (USPMag), using a petrographic transmitted/reflected light microscope (Olympus BX41). In addition to conventional petrographic analyses, we used ImageJ (Schindelin et al.2012) to perform supervised directional analyses of the major axis of both silicate and oxide crystals. SEM observations were carried out at the Laboratório de Caracterização Tecnológica of the Departamento de Engenharia de Petróleo, Universidade de São Paulo (LCT/USP). Quantitative observations of SEM were performed on six carbon-coated samples using a Stereoscan S440 (LEO) SEM with resolution about 10 nm. Compositional maps were performed using a Quanta 600 FEG SEM (FEI Company). Both microscopes were operated at 20 kV, with a magnification of 7000–7500 X. Backscattered Energy Dispersive Spectra (EDS) images (ESPRIT software) and chemical maps (EDS X-ray maps) of major elements abundances were acquired on representative parts of each sample. Note that SEM observations were performed on bulk rock samples. 3.2 Magnetic mineralogy characterization To complement the interpretation of magnetic fabric, the magnetic properties and proportions of grain size were analysed. Magnetic mineralogy was investigated through temperature-dependent low-field magnetic susceptibility curves (χ-T diagrams), isothermal remanent magnetization (IRM) acquisition curves, hysteresis loops and first-order reversal curves (FORC) diagrams. Temperature-dependent low-field magnetic susceptibility measurements (χ-T) were performed in one representative sample for each sampling site. Samples were manually crushed in an agate mortar and the magnetic susceptibility of the powder was measured during continuous heating and cooling curves between room temperature and ∼600 °C using a Kappabridge KLY4 coupled with a CS3 furnace (Agico Ltd.). The degree of reversibility of heating and cooling cycles reflect phase transformations during the thermomagnetic experiment. It was estimated using the alteration indexes defined in Hrouda (2003) that estimate the maximum difference in susceptibility between curves (Amax) and the difference in susceptibility at 40 °C (A40). Curves were considered reversible when both Amax and A40 values were below 20 (absolute value). The determination of phase transformations of magnetic minerals and the reversibility of the heating-cooling cycles provide information on the nature of the original magnetic grains and the stability of the magnetic carriers upon temperature (e.g. Hrouda 1994, 2003). Transition temperatures of magnetic minerals (Curie temperatures, Tc) were obtained by the second-derivative of the heating curve (Tauxe 1998). Hysteresis loops and isothermal remanent magnetization (IRM) acquisition curves were measured at room temperature in small rock chips cut from one specimen per site. Measurements were performed by applying fields of up to 1 T using a Princeton Measurements Corporation Micromag vibrating sample magnetometer (VSM). The saturation magnetization (Ms), saturation remanent magnetization (Mrs), coercivity (Bc) and coercivity of remanence (Bcr) are all determined from hysteresis loops and backfield measurements. The ratios of Mrs/Ms and Bcr/Bc reflect relative proportions of the single domain (SD), multidomain (MD) and pseudo-single domain (PSD) structures of the magnetic minerals (e.g. Day et al.1977; Dunlop & Özdemir 1997; Dunlop 2002). These standard hysteresis parameters, however, provide only a measure of the bulk magnetic properties and therefore are not suitable for discriminating the different magnetic components contributing to the magnetization in samples with mixed magnetic assemblage, e.g. samples plotting in the PSD region in the Day et al. (1977) diagram could be mixtures of SD and MD grains rather than PSD grains and this distinction cannot be made from a Day plot alone. Given the complex magnetic mineralogy in our samples, we used first-order reversals curves (FORCs) to identify and discriminate the different magnetic mineral assemblages (Roberts et al.2014). In addition, FORC diagrams also indicate the presence of magnetostatic interactions among magnetic grains (Roberts et al.2000, 2014). FORC measurements were performed at room temperature after 300 reversal curves with an averaging time of 200 ms. FORC diagrams were calculated using the FORCinel software package (Harrison & Feinberg 2008) with a smoothing factor of 4 for all samples. 3.3 Anisotropies of magnetic susceptibility (AMS) and anhysteretic remanent magnetization (AARM) The anisotropy of low-field magnetic susceptibility (AMS) was used to infer the petrofabric of our rock collection. AMS is a superposition of dia-, paramagnetic and ferromagnetic minerals in the rock and depends on their intrinsic anisotropy and spatial distribution within a rock sample (Tarling & Hrouda 1993). AMS measurements consist of the acquisition of magnetic susceptibility measurements in different directions to resolve the magnetic susceptibility (K) tensor. In practice, the magnetic susceptibility tensor is represented by an ellipsoid of magnetic susceptibility, defined by the length and orientation of its three principal axes K1, K2 and K3 (K1 ≥ K2 ≥ K3). The mean magnetic susceptibility (Km) is the arithmetic mean of K1, K2 and K3. Several parameters have been proposed to express the shape and degree of anisotropy of the susceptibility ellipsoid. The magnetic lineation is defined as the maximum susceptibility axis (K1) and the magnetic foliation is defined as the plane containing maximum and intermediate axes (K1 and K2) and perpendicular to the minimum axis (K3). The magnitude of these parameters can be computed as a ratio between the main axes: magnetic lineation (L = K1/K2) and magnetic foliation (F = K2/K3). The shape of the anisotropy tensors is usually evaluated through the T parameter (T = 2.ln(K2/K3)/ln(K1/K3) − 1), which is used to distinguish among prolate (T < 0), oblate (T > 0) or triaxial ellipsoids (T = 0) (Jelinek 1981). The degree of anisotropy is given by P = K1/K3 or, in its corrected form, by |$P^{\prime } = \mathrm{ exp}[\sqrt{2((\eta _1 - \eta _m)^2+(\eta _2 - \eta _m)^2+(\eta _3 - \eta _m)^2)}]$| (Jelinek 1981), where ηi is the natural logarithm of Ki. AMS measurements were performed with an automatic Kappabridge MFK1-A apparatus (Agico Ltd.), using the rotation mode and operating in a low alternating field (300 A m−1) at 976 Hz. Measurements were performed on all 30 sampled sites (430 specimens), comprising 16 sites of lava flows and 14 of both welded and non-welded ignimbrites. All the stereonets presented in this study are in the bedding coordinate system (bedding being rotated to the horizontal). The AARM was determined to isolate the contribution of remanence-carrying minerals from that of the paramagnetic and diamagnetic matrix as well as to exclude the presence of inverse fabrics (Jackson 1991). The anhysteretic remanent magnetization (ARM) is acquired when a ferromagnetic grain is subjected simultaneously to an alternating magnetic field (AF) and a small bias direct field. Similarly to the AMS, for low bias fields the AARM can be described by an ellipsoid of anhysteretic remanence (MA), with three orthogonal axis M1, M2 and M3 (M1 ≥ M2 ≥ M3) (Jackson 1991). Schemes with 6, 12 and 15 positions were tested and the 15 positions scheme was the only one that resolved for all samples the AARM tensor with confidence following the F-test of Jelínek (1977). Before AARM analysis, samples were AF demagnetized in 100 mT to establish the base level. ARM was acquired with an AF peak field of 100 mT and a bias field of 0.50 mT to cover most of the grain-size populations. ARM acquisitions were performed with an LDA3-AMU1 (AGICO, Inc) and remanence measurements after each inducing step were performed with a JR6A magnetometer (sensitivity of less than 100 A m−1) (AGICO, Inc). AARM was measured on samples from 16 sites (Table 1). 4 RESULTS 4.1 Petrography and SEM observations Overall, petrographic analysis shows large silicate crystals comprising essentially plagioclase with different sizes involved by a matrix with different degrees of crystallinity and textures (Fig. 3). Site COP01 (lava flow from Cola de Zorro Fm.; Figs 2a and 3a) presents a preferential alignment of plagioclase crystals, both of phenocrysts and of smaller crystals in the matrix (trachytic texture - S° = 200°/05°NW). Site COP05 (also a lava flow from Cola de Zorro Fm.; Fig. 3b) shows plagioclase phenocrysts roughly oriented within a fine matrix (S° = 200°/04°NW). Site COP33 (Las Mellizas Fm., Lower Lavas; Fig. 3c) exhibit slightly oriented phenocrysts of plagioclase and olivine involved by a glassy matrix (S° = 030°/03°SE). Site COP09 (Las Mellizas Fm., Upper Lavas; Fig. 3d) presents oriented plagioclase phenocrysts and a fine matrix mainly composed of olivine, pyroxene, plagioclase and glass. All lava samples contain plagioclase with fine to coarse crystal reabsorption (sieve texture) suggesting rapid decompression during the magmatic events. The crystal alignment observed in the samples evidences the effective mechanical orientation of the silicate fabric inside the lava flows, which was observed and measured in the field. Sites COP23 and COP24 (Riscos Bayos ignimbrite; Figs 3e and f) present lithic and crystal fragments of pyroxene surrounded by a fine ash matrix and non-welded pumice. These samples also display a mean orientation defined by pumices porosity elongation and the imbrication of crystals and lithic fragments that were measured in the field (S° = 020°/05°SE). Figure 3. Open in new tabDownload slide Photographs of thin sections of some sampling sites. (a, b) Cola de Zorro Fm. lava flows with trachytic texture showing clear matrix and plagioclase crystals orientation. (c) Las Mellizas Fm. Lower Lavas showing slightly oriented plagioclase phenocrystals involved by glassy matrix. (d) Las Mellizas Fm. Upper Lavas sample displaying oriented plagioclase phenocrystals and fine matrix. (e, f) Riscos Bayos Ignimbrites samples with non-welded pumice, lithics and crystal fragments involved by a fine ash matrix. Figure 3. Open in new tabDownload slide Photographs of thin sections of some sampling sites. (a, b) Cola de Zorro Fm. lava flows with trachytic texture showing clear matrix and plagioclase crystals orientation. (c) Las Mellizas Fm. Lower Lavas showing slightly oriented plagioclase phenocrystals involved by glassy matrix. (d) Las Mellizas Fm. Upper Lavas sample displaying oriented plagioclase phenocrystals and fine matrix. (e, f) Riscos Bayos Ignimbrites samples with non-welded pumice, lithics and crystal fragments involved by a fine ash matrix. Considering that most of our samples present plagioclase crystals as an important mineral phase, we used the major axis of these crystals to determine the orientation of the silicate fabrics in the studied samples, as proposed by Archanjo & Launeau (2004) and Bascou et al. (2005). Due to the abundance of pumice and crystal fragments in the ignimbrites, we used these elements to define the silicate fabrics of these rocks. Rose diagrams based on the directional analysis performed using ImageJ suggest an effective orientation of both silicate and magnetite crystals (Fig. 4), with variable degrees of scattering. In special, Riscos Bayos Ignimbrites samples display a minor amount of magnetite crystals (Fig. 4; COP23 and COP24), resulting in a less defined magnetic visual orientation. Figure 4. Open in new tabDownload slide Rose diagrams showing the directions of major axis of opaque and silicate crystals. Figure 4. Open in new tabDownload slide Rose diagrams showing the directions of major axis of opaque and silicate crystals. SEM observations show a wide variation of Fe-Ti oxide constituents, from pure magnetite to titanomagnetite crystals (Ti-mag) with variable amounts of Ti (from 0.14 to 51.37 per cent), as revealed by composition mapping of the selected samples (Fig. 5). In plagioclase and pyroxenes (large grain; Fig. 5), some Ti-magnetite (Ti-mag) crystals seems to have exsolved from the host mineral. This exsolved Ti-mag may carry remanence with a distinct orientation, depending upon its position along the edge of the lattice of the host mineral, which can ultimately influence or dominate the AMS signal (Biedermann et al.2013). Samples also display wide variations concerning the crystal sizes, their spatial distribution, and shape, as presented by BSE images shown in Fig. 5. Site COP01 (Cola de Zorro Fm.) presents a uniform Ti-mag population with crystal sizes ranging from 10 to 50 |$\mu$|m and low-Ti content (Fig. 5a), while site COP05 (Fig. 5b), from the same geological unit, exhibits a bimodal distribution of Ti-mag, one marked by abundant small crystals (∼20 |$\mu$|m) and another characterized by larger, more scarce, rounded crystals (∼300 |$\mu$|m), also with low-Ti content. Site COP33 (Las Mellizas Fm. - Lower Lavas) also displays two populations of Ti-mag with distinct crystal sizes (∼20 and ∼100 |$\mu$|m), both marked by idiomorphic crystals with straight and well-defined borders (Fig. 5c). Also from Las Mellizas Fm., site COP09 (Upper Lavas) shows a single low-Ti fine-grained (∼10 |$\mu$|m) Ti-mag population well distributed throughout the entire fabric (Fig. 5d). In this site, we also observe a considerable amount of iron-oxides crystals smaller than ∼2 |$\mu$|m. Sites COP23 and COP24 (Riscos Bayos Fm.) display lower concentrations of Ti-mag and both are marked by idiomorphic crystals ranging from 40 to 200 |$\mu$|m and presenting low-Ti content, although some high-Ti crystals may be visible in site COP23 (Figs 5e and f). Figure 5. Open in new tabDownload slide Mosaic of representative X-ray compositional maps (Fe and Ti) and BSE images showing major silicates (grey) and oxide (bright white) distribution. (a, b) Cola de Zorro Fm. (c) Las Mellizas Fm. - Lower lavas. (d) Las Mellizas Fm. - Upper lava. (e, f) Riscos Bayos Fm. Figure 5. Open in new tabDownload slide Mosaic of representative X-ray compositional maps (Fe and Ti) and BSE images showing major silicates (grey) and oxide (bright white) distribution. (a, b) Cola de Zorro Fm. (c) Las Mellizas Fm. - Lower lavas. (d) Las Mellizas Fm. - Upper lava. (e, f) Riscos Bayos Fm. 4.2 Magnetic mineralogy Representative thermomagnetic curves are presented in Fig. 6. Estimated Curie temperatures (Table 1) suggest that the main magnetic mineral present in the rocks is titanomagnetite with variable Ti content from pure magnetite (0 per cent Ti) to Ti-rich titanomagnetite (40 per cent) (Dunlop & Özdemir 1997). The reversibility of heating–cooling cycles depends mainly on the lithology and stratigraphic unit. Samples from Cola de Zorro Fm. show reversible curves during heating–cooling cycles (Figs 6a–c) and magnetic susceptibility with a high-temperature transition at 580–590 °C, suggesting the presence of magnetite or Ti-mag with low Ti content. From Riscos Bayos Fm., samples also show high Tc around 580–600 °C (Figs 6 o, p, r and s) related to magnetite, however, they commonly present irreversible behaviour and in some cases, a Hopkinson peak in the heating curve (e.g. COP30). From Las Mellizas Lower Lavas, curves are not reversible and samples have the lowest Tc around 200 °C (Figs 6e and m) indicating Ti-mag with high Ti content. Welded ignimbrites from Las Mellizas present the same behaviour as that described for Riscos Bayos (Figs 6f, g, i, j and t). The Upper Lavas exhibit irreversible thermomagnetic curves with high Tc around 580 °C (Figs 6h and k) indicating the presence of magnetite and low Tc (200 °C; Fig. 5k) indicating Ti-mag with high Ti content. Figure 6. Open in new tabDownload slide Thermomagnetic curves for individual rock types. Full heating curve (red) and cooling curve (blue) were obtained heating from room temperature up to 600 °C and cooling again to room temperature. Powdered samples were heated in air atmosphere. Figure 6. Open in new tabDownload slide Thermomagnetic curves for individual rock types. Full heating curve (red) and cooling curve (blue) were obtained heating from room temperature up to 600 °C and cooling again to room temperature. Powdered samples were heated in air atmosphere. High-field experiments of IRM acquisition curves, hysteresis loops and FORC diagrams revealed three different behaviours in our sample collection (Fig. 7). In the first one, seven sites present IRM curves saturating at fields around 0.2 T (e.g. Fig. 7a); hysteresis loops for these samples exhibit coercivities <10 mT (Fig. 7b) and FORC diagrams (Fig. 7c) are typical of MD grains. This behaviour is observed on COP01, COP03, COP05, COP20 (Cola de Zorro Fm.), COP10 (Upper Lavas, Las Mellizas Fm.), COP28 (Lower Lavas, Las Mellizas Fm.) and COP29 (Riscos Bayos Fm.). The second one is observed on 15 sites presenting IRM curves with saturating fields at around 0.25 T (Fig. 7d), coercivities between 10 and 20 mT (Fig. 7e) and FORC diagrams characteristic of magnetite with PSD structure or mixtures of SD and MD magnetite grains. (Fig. 7f). Finally, the third behaviour comprises seven sites where IRM curves show saturating fields ≥0.3 T (Figs 7g and j), coercivities are between 20 and 35 mT (Figs 7h and k) and FORC diagrams are typical of SD grains (Figs 7i and l). This last behaviour is observed on COP12, COP14, COP22, COP32 (Ignimbrites, Las Mellizas Fm.), COP09, and COP16 (Upper Lavas, Las Mellizas Fm.) and COP23 (Riscos Bayos Fm.). Figure 7. Open in new tabDownload slide IRM acquisition curves, hysteresis loops and FORC diagrams. (a, d, g and j) representative IRM curves for sites COP29, COP02, COP09 and COP32. (b, e, h and k) represents three characteristic hysteresis loops with respective coercivity and magnetization parameters, where red loops represent uncorrected para- and diamagnetic data. The hysteresis parameters are inside the graphics. (c, f, i and l) FORC diagrams for the same samples (smoothing factor 3–5), which (c) is typical of MD, (f) PSD, and (i and l) SD grains. Figure 7. Open in new tabDownload slide IRM acquisition curves, hysteresis loops and FORC diagrams. (a, d, g and j) representative IRM curves for sites COP29, COP02, COP09 and COP32. (b, e, h and k) represents three characteristic hysteresis loops with respective coercivity and magnetization parameters, where red loops represent uncorrected para- and diamagnetic data. The hysteresis parameters are inside the graphics. (c, f, i and l) FORC diagrams for the same samples (smoothing factor 3–5), which (c) is typical of MD, (f) PSD, and (i and l) SD grains. 4.3 Magnetic fabric 4.3.1 AMS fabric Histograms of mean magnetic susceptibility (Km) (Fig. 8a) show different distributions for the three studied units. Samples from Las Mellizas Fm. presented a bimodal distribution, with lower values around 9−17 × 10−3 SI and higher values at 20−33 × 10−3 SI. This bimodal distribution is unrelated to the rock type since both lavas and welded ignimbrites have representative samples in the two ranges of susceptibility. Samples from Riscos Bayos ignimbrites are characterized by the lowest susceptibility values among the studied formations, around 3−16 × 10−3 SI (average of ∼4.2 × 10−3 SI). Cola de Zorro Fm. show a wide-range distribution of magnetic susceptibility within 9−45 × 10−3 SI (average of ∼15.5 × 10−3 SI), the highest values (>30 × 10−3) correspond to site COP20, located in the northeast of the Caviahue Depression. The large variation in susceptibility values of the three studied formations may be linked to their magnetic mineral content. Also, most values are larger than 10−2 SI (except in Riscos Bayos Fm.), which can be attributed to the dominance of the ferromagnetic fraction in a bulk arrangement of studied rocks (Tarling & Hrouda 1993). Figure 8. Open in new tabDownload slide AMS scalar results. (a) Histograms with mean magnetic susceptibility (Km); (b) Degree of anisotropy (P’) and the mean magnetic susceptibility (Km); (d) Jelinek’s plot relating the shape parameter (T) and the degree of anisotropy (P’). Figure 8. Open in new tabDownload slide AMS scalar results. (a) Histograms with mean magnetic susceptibility (Km); (b) Degree of anisotropy (P’) and the mean magnetic susceptibility (Km); (d) Jelinek’s plot relating the shape parameter (T) and the degree of anisotropy (P’). The degree of anisotropy is low in all samples, reaching values up to 5 per cent (average of 1.5 per cent), as is commonly found for ignimbrites and volcanic rocks (Tarling & Hrouda 1993; Agrò et al.2015). The studied formations show a clear distinction in their magnetic susceptibilities. But there is no clear correlation between magnetic susceptibility and the degree of anisotropy within each unit (Fig. 8b), except for the Cola de Zorro collection, which presents a slight direct correlation between these two parameters. Of the 430 analysed samples, 300 are in the oblate field (121 with T > 0.5) and 130 in the prolate one (42 with T < 0.5). Therefore, most anisotropy ellipsoids have a strongly triaxial to slightly oblate shape (Fig. 8c). The AMS directional data were analysed through the distribution of the main susceptibility axes within the lower hemisphere of equal-area stereonets (Fig. 9). The spatial orientation and clustering of each axis were statistically evaluated by principal directions of mean tensor and the semi-angles of their 95 per cent confidence ellipses, E1 − 2 and E2 − 3 (Jelínek & Kropáček 1978). For each site, the magnetic lineation (K1) and foliation pole (K3) were not considered, respectively, when E1 − 2 and E2 − 3 were higher than 26° (Jelínek & Kropáček 1978). That is, the directional data were summarized as high scattered and no statistical interpretation was performed neither the specimen nor the site level. Figure 9. Open in new tabDownload slide Stereonets with main axes of AMS (K1 > K2 > K3) and AARM (M1 > M2 > M3) for all studied sites. The red squares represent K1, green triangles K2, blue dots K3, black squares M1, triangles M2 and dots M3. The grey and white stereonets correspond respectively to ignimbrite and lava formations. Figure 9. Open in new tabDownload slide Stereonets with main axes of AMS (K1 > K2 > K3) and AARM (M1 > M2 > M3) for all studied sites. The red squares represent K1, green triangles K2, blue dots K3, black squares M1, triangles M2 and dots M3. The grey and white stereonets correspond respectively to ignimbrite and lava formations. In the lava formations (white stereonets), four sites (COP04, COP06, COP16 and COP19) present low grouping of AMS principal directions of individual samples. These sites correspond to those with the lowest Tc values (200 °C; Fig. 6). For sites that passed the selection criteria above, the fabrics were grouped into ‘normal’, ‘inverse’ or ‘intermediate’ (Fig. 10a) according to bedding plane. The normal fabric was observed in six sites (COP03, COP10, COP17, COP21, COP28 and COP33) and the inverse fabric was observed in the site COP09. Intermediate fabric was not observed, and sites COP01, COP02, COP05 and COP20 presented anomalous fabrics that do not fall in any definition given in Fig. 10. Concerning the orientation of the magnetic lineation, each unit presented a different behaviour. In the Cola de Zorro Fm., magnetic lineation presents W–E (COP02, COP20) and SW–NE (COP03) directions. Las Mellizas Upper Lavas exhibit magnetic lineation with variable directions: N–S (COP10), W–E (COP17) and SW–NE (COP21). For Las Mellizas Lower Lavas, the magnetic lineation present SW–NE (COP28) and W–E (COP33) directions. Figure 10. Open in new tabDownload slide Classification of magnetic fabric. (a) different fabrics found in lava flows, defined as normal, intermediate and inverse according to the bedding plane. (b) fabrics found in ignimbrites, defined as parallel, oblique and transverse according to de imbrication direction of the foliation plane (e.g. Agrò et al.2015). Figure 10. Open in new tabDownload slide Classification of magnetic fabric. (a) different fabrics found in lava flows, defined as normal, intermediate and inverse according to the bedding plane. (b) fabrics found in ignimbrites, defined as parallel, oblique and transverse according to de imbrication direction of the foliation plane (e.g. Agrò et al.2015). In the ignimbrites (grey stereonets), poorly defined fabrics were observed in three sites (COP24, COP26 and COP27). The other sites present well-defined magnetic foliations with a sub-vertical pole (K3). These fabrics were grouped in ‘parallel’, ‘transverse’ and ‘oblique’ (Fig. 10b) according to the imbrication direction (e.g. Agrò et al.2015). The parallel fabric was found in six sites (COP07, COP11, COP13, COP25, COP29 and COP30), transverse in four sites (COP08, COP12, COP22 and COP23) and oblique in the site COP32. For the ignimbrites from Las Mellizas Fm., four sites present imbrication on their magnetic foliation indicating mass transport towards NE or N (COP07, COP11, COP12 and COP13) with inclinations varying from 10° to 35° (Fig. 9). Other sites show imbrication towards E (COP32), SW (COP08) and S (COP22). In the ignimbrites from Riscos Bayos Fm., all sites (COP23, COP25, COP29 and COP30) present a well-defined fabric with imbrication towards SE and inclinations between 15° and 30° (Fig. 9). 4.3.2 AARM fabric Coercivity spectra (Fig. 11) shows two different behaviours, one comprising samples with low-coercivity grains (red curves) and other with both low- and high-coercivity grains (black curves). Directional ARM was subsequently imparted using AC field of 100 mT to cover most of the grain-size populations. In the 16 sites analysed by AARM (Table 1 and Fig. 9), the anisotropy significance was evaluated through the confidence angles E1 − 2 (17° to 40°), E2 − 3 (3° to 15°) and E1 − 3 (5° to 27°). Sites COP02, COP03, COP07, COP08, COP11, COP12, COP13, COP22, COP23, COP28, COP29 and COP32 presented parallel AMS and AARM axes, suggesting that multidomain Ti-mag dominates the magnetic anisotropy. Sites COP08, COP09 and COP23 showed anomalous fabrics with an exchange between susceptibility and anhysteretic remanence main axes. Particularly, for site COP09 an inverse correlation is observed, with the maximum remanence axis (M1) being nearly horizontal and parallel to the minimum susceptibility axis (K3). No correlation was observed in sites COP04, COP14 and COP16. Figure 11. Open in new tabDownload slide Coercivity spectra for samples with unimodal (red curves) and bimodal (black curves) distributions of coercivities. Figure 11. Open in new tabDownload slide Coercivity spectra for samples with unimodal (red curves) and bimodal (black curves) distributions of coercivities. 5 DISCUSSION Directional differences and the scatter of the AMS signal in the ignimbrites studied here have been previously associated with geological/geomorphological macroscopic factors, such as flow regime, topographical irregularities and the distance from the volcanic centre (e.g. Ort et al.2015). It has also been shown that even in rocks with high bulk susceptibility, such as the ones studied here, the anisotropy of susceptibility can be carried by minerals other than magnetite (e.g. Rochette & Vialon 1984; Hounslow 1985; Borradaile 1987; Rochette et al.1992; Hirt et al.1995). In our case, we compare the anisotropy of susceptibility and the anisotropy of remanence and showed that they are compatible with almost all sites except for those where a high proportion of SD grains is observed. We then explore the possibility that even when magnetite carries most of the anisotropy in rocks, microscopic factors such as the order of crystallization of magnetic carriers, their grain-size and composition may exert a strong influence on the AMS signal, even for sites with well-defined macroscopic structures (e.g. magmatic foliation). Field and petrographic analyses indicate well-oriented silicate fabrics for the studied units (Fig. 4), with two different textures of titanomagnetite in lavas and ignimbrites. The first one is characterized by straight-edge idiomorphic crystals often included in plagioclase phenocrysts (Fig. 5c) and the second by xenomorphic crystals with rounded edges (Fig. 5b). The thermomagnetic curves for sites showing exclusively idiomorphic, small magnetite crystals exhibit Curie temperatures near to 580–600 °C, which indicates low-Ti titanomagnetites or pure magnetites as the main magnetic carriers in these rocks, not necessarily implying Ti-mag as the main carriers of the magnetic fabric. The AMS in those sites is well-defined, with well-grouped susceptibility axes, with exception to sites with a very low degree of anisotropy (Fig. 12). These results suggest a primary origin for the low-Ti titanomagnetites (to pure magnetites) and that the crystallization of the magnetic carriers occurred in the early stage of magma evolution, resulting in an effective orientation of Ti-mag in concordance with the flow-induced strain (Fig. 13a). Figure 12. Open in new tabDownload slide Comparison among the curie temperatures, shape parameter and the dispersion of the measured sites. Dispersion is expressed through E1 − 2 for prolate sites and E2 − 3 for oblate sites. Figure 12. Open in new tabDownload slide Comparison among the curie temperatures, shape parameter and the dispersion of the measured sites. Dispersion is expressed through E1 − 2 for prolate sites and E2 − 3 for oblate sites. Figure 13. Open in new tabDownload slide A summary of the AMS possibilities resulting from the combination of various scenarios: (a) MD, petrographically coherent crystals with high-Tc values; (b) MD, petrographically coherent crystals with low-Tc values; (c) MD, petrographically incoherent crystals low-Tc values; (d) SD, petrographically coherent crystals with high-Tc values; (e) SD, petrographically coherent crystals with low-Tc values; (f) SD, petrographically incoherent crystals low-Tc values. Figure 13. Open in new tabDownload slide A summary of the AMS possibilities resulting from the combination of various scenarios: (a) MD, petrographically coherent crystals with high-Tc values; (b) MD, petrographically coherent crystals with low-Tc values; (c) MD, petrographically incoherent crystals low-Tc values; (d) SD, petrographically coherent crystals with high-Tc values; (e) SD, petrographically coherent crystals with low-Tc values; (f) SD, petrographically incoherent crystals low-Tc values. On the other hand, the magnetic results for samples with xenomorphic oxide crystals systematically show thermomagnetic curves with low Curie temperature (∼ 200 °C), suggesting high-concentrations of Ti in the titanomagnetites (e.g. Dunlop & Özdemir 1997). Mineral texture in petrographic and SEM observations suggest that these titanomagnetites formed late in the history of magma crystallization, occupying the interstices left by the early crystallized minerals (Figs 13b and c). Under these conditions, the magnetic fabric varies from well to poorly defined, depending on whether the late Ti-mag crystallization occurred in concordance with the pre-existing silicate fabric or not (Hargraves et al.1991). In several sites, we observe a concordance between late, high-Ti titanomagnetite crystals and the silicate fabrics (e.g. site COP5), where these crystals form along the main axis of elongated silicate crystals (e.g. plagioclase and pyroxene) resulting in a well-defined AMS tensor (Fig. 13b). In contrast, there is a second group of sites with late Ti-mag crystals displaying incoherent orientation with the silicate fabrics. In this case, late crystals form randomly, not following the main silicate fabrics and yielding poorly defined AMS results (Fig. 13). Another factor that may control the AMS is the domain structure of titanomagnetites. Rocks with MD Ti-mag may display a normal AMS fabric since its magnetization is dominantly governed by self-demagnetization and shape anisotropy. In contrast, elongated small SD Ti-mag grains will show an inverse AMS fabric, because the magnetization in these grains is saturated along the easy axis. This will result in an exchange between the long axis of the crystal and the maximum AMS axis (Fig. 13d; Rochette 1987; Rochette et al.1992). But intermediate AMS fabrics may arise when mixing small SD and large MD grains in the same rock. Ferré et al. (2002) and Ferré (2002) demonstrate that different mixtures of SD-MD Ti-mag grains result in intermediate fabrics. Although this kind of fabric can be observed by a contribution of different minerals, when Ti-mag dominates the fabric the magnetic lineation does not reflect the flow direction, and there is no straightforward relationship between the shape or degree of the anisotropy and rock texture. Sites of lava flow with a typical MD Ti-mag shows a ‘normal’ magnetic fabric and parallel AMS and AARM main axes. On the other hand, the site with a dominant SD behaviour (COP09) shows an ‘inverse’ magnetic fabric, where K1 is parallel to the M3 axis of the AARM tensor (Fig. 9). This is commonly associated with the ‘SD effect’ that has been long discussed in studies of magnetic anisotropy (Potter & Stephenson 1988; Rochette et al.1999), but its experimental observation is rare due to other factors that can also generate this type of fabric (Tarling & Hrouda 1993; Chadima et al.2009; Cañón-Tapia & Mendoza-Borunda 2014). Here, the combined study of AMS and AARM allows us to eliminate this ambiguity, once SD Ti-mag (the main magnetic carrier observed in our sample collection) present magnetic fabric whose axis of maximum remanence is parallel to the minimum susceptibility. The petrographic analysis of site COP09 exhibits disseminated Fe-Ti oxides less than 2 |$\mu$|m diameter, reinforcing the presence of SD Ti-mag (see Supporting Information Table ST1). In the ignimbrite samples, the magnetic domain structure factor may also have affected the magnetic fabric. FORC diagrams with MD behaviour are exclusively related to ‘parallel’ fabric, whereas FORC diagrams with mixtures of MD–SD grains are linked to ‘transverse’ fabric. This fabric has been associated with very intense crystal-bearing pyroclastic fluxes by other authors (Giordano et al.2008; Ort et al.2013; Agrò et al.2015; Ort et al.2015). However, the ‘SD effect’ also occurs in the ignimbrites. It may produce a transverse fabric that is not related to the flow regime, but instead to the composite fabrics of SD and MD magnetite grains present in the same site. Thus, any geological interpretation from AMS data should carefully consider the carriers of magnetic anisotropy and not only macroscopic factors. 6 CONCLUSIONS The comparison between petrographic properties, rock magnetism, and magnetic anisotropies revealed that the CVCC extrusive and pyroclastic rocks present varied magnetic mineralogy, formed in different stages of the magmatic evolution. Some sites are characterized by idiomorphic magnetite crystals of pure magnetite or low-Ti titanomagnetites, with elevated Curie temperature. Other sites present high-Ti titanomagnetites with irregular forms, typical of late crystallization. Also, there is a great granulometric variation in the titanomagnetite crystals, with a predominance of large MD crystals in some sites and SD crystals in others. In all cases, titanomagnetite dominates the magnetic fabric of the rocks. The sites with late-crystallized high-Ti titanomagnetite have an undefined fabric (E1 − 2, E2 − 3 > 26°), while the sites with early-crystallized magnetite and low-Ti titanomagnetite have well-defined magnetic fabric. This probably occurs because the earliest minerals register the magma flow (or the pyroclastic flux in the case of ignimbrites). Among the sites with pure magnetite or low-Ti titanomagnetite, some have a significant amount of crystals with SD domain structure. In one of these sites, the fabric has an inverse behaviour (e.g. COP09), showing perpendicular AMS and AARM fabrics. Further AARM measurements at sites where SD grains were identified will be required to determine the influence of these grains on the final AMS fabric, eventually producing intermediate fabric orientations. This work demonstrates the complexity in the interpretation of magnetic anisotropy data in volcanic rocks, suggesting that such studies need to be accompanied by a detailed investigation of the magnetic mineralogy, and the minerals carrying the magnetic fabrics. ACKNOWLEDGEMENTS This work was supported by CNPq grant 144580/2016-0 to TRM, CAPES AUXPE 2043/2014 and CNPq grant 454609/2014-0 to GAH, CNPq/CAPES grant 400724/2014-6 and CNPq grants 441766/2014-5 and 303015/2015-2 to CAS. We thank D. Brandt, G. Moreira e P. Jaqueto by the support during sample preparation and data acquisition. We thank two anonymous reviewers for their helpful and constructive comments. REFERENCES Agrò A. , Zanella E. , Le Pennec J.-L. , Temel A. , 2015 . Magnetic fabric of ignimbrites: a case study from the Central Anatolian Volcanic Province , Geol. Soc. London , 396 ( 1 ), 159 – 175 . Google Scholar Crossref Search ADS WorldCat Archanjo C.J. , Launeau P. , 2004 . Magma flow inferred from preferred orientations of plagioclase of the Rio Ceará-Mirim dyke swarm (NE Brazil) and its AMS significance , Geol. Soc. London , 238 ( 1 ), 285 – 298 . Google Scholar Crossref Search ADS WorldCat Bascou J. , Camps P. , Dautria J.M. , 2005 . Magnetic versus crystallographic fabrics in a basaltic lava flow , J. Volc. Geotherm. Res. , 145 ( 1-2 ), 119 – 135 . Google Scholar Crossref Search ADS WorldCat Biedermann A.R. , Lowrie W. , Hirt A.M. , 2013 . A method for improving the measurement of low-field magnetic susceptibility anisotropy in weak samples , J. appl. Geophys. , 88 , 122 – 130 . Google Scholar Crossref Search ADS WorldCat Borradaile G. , 1987 . Anisotropy of magnetic susceptibility: rock composition versus strain , Tectonophysics , 138 ( 2-4 ), 327 – 329 . Google Scholar Crossref Search ADS WorldCat Borradaile G. , Henry B. , 1997 . Tectonic applications of magnetic susceptibility and its anisotropy , Earth-. Sci. Rev. , 42 ( 1-2 ), 49 – 93 . Google Scholar Crossref Search ADS WorldCat Borradaile G.J. , Jackson M. , 2010 . Structural geology, petrofabrics and magnetic fabrics (AMS, AARM, AIRM) , J. Struct. Geol. , 32 ( 10 ), 1519 – 1551 . Google Scholar Crossref Search ADS WorldCat Cañón-Tapia E. , 2004 . Flow direction and magnetic mineralogy of lava flows from the central parts of the peninsula of Baja California, Mexico , Bull. Volcanol. , 66 ( 5 ), 431 – 442 . Google Scholar Crossref Search ADS WorldCat Cañón-Tapia E. , Mendoza-Borunda R. , 2014 . Magnetic petrofabric of igneous rocks: Lessons from pyroclastic density current deposits and obsidians , J. Volc. Geotherm. Res. , 289 , 151 – 169 . Google Scholar Crossref Search ADS WorldCat Cañón-Tapia E. , Walker G.P. , Herrero-Bervera E. , 1995 . Magnetic fabric and flow direction in basaltic pahoehoe lava of Xitle volcano, Mexico , J. Volc. Geotherm. Res. , 65 ( 3-4 ), 249 – 263 . Google Scholar Crossref Search ADS WorldCat Cañón-Tapia E. , Walker G.P. , Herrero-Bervera E. , 1996 . The internal structure of lava flows—insights from AMS measurements i: near-vent a’a , J. Volc. Geotherm. Res. , 70 ( 1-2 ), 21 – 36 . Google Scholar Crossref Search ADS WorldCat Chadima M. , Cajz V. , Tỳcová P. , 2009 . On the interpretation of normal and inverse magnetic fabric in dikes: examples from the Eger Graben, NW Bohemian Massif , Tectonophysics , 466 ( 1-2 ), 47 – 63 . Google Scholar Crossref Search ADS WorldCat Cole J. , Milner D. , Spinks K. , 2005 . Calderas and caldera structures: a review , Earth-Sci. Rev. , 69 ( 1-2 ), 1 – 26 . Google Scholar Crossref Search ADS WorldCat Day R. , Fuller M. , Schmidt V. , 1977 . Hysteresis properties of titanomagnetites: grain-size and compositional dependence , Phys. Earth planet. Inter. , 13 ( 4 ), 260 – 267 . Google Scholar Crossref Search ADS WorldCat Dunlop D. , Özdemir , 1997 . Rock magnetism: Fundamentals and Frontiers , Cambridge University Press , 573 p. Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Dunlop D.J. , 2002 . Theory and application of the day plot (Mrs/Ms versus Hcr/Hc) 1. theoretical curves and tests using titanomagnetite data , J. geophys. Res. , 107 ( B3 ), EPM – 4 . OpenURL Placeholder Text WorldCat Elston W.E. , Smith E.I. , 1970 . Determination of flow direction of rhyolitic ash-flow tuffs from fluidal textures , Bull. geol. Soc. Am. , 81 ( 11 ), 3393 – 3406 . Google Scholar Crossref Search ADS WorldCat Ferré E.C. , 2002 . Theoretical models of intermediate and inverse AMS fabrics , Geophys. Res. Lett. , 29 ( 7 ), 31 – 1 . Google Scholar Crossref Search ADS WorldCat Ferré E.C. , Bordarier C. , Marsh J.S. , 2002 . Magma flow inferred from AMS fabrics in a layered mafic sill, Insizwa, South Africa , Tectonophysics , 354 ( 1-2 ), 1 – 23 . Google Scholar Crossref Search ADS WorldCat Folguera A. , Ramos V. , 2000 . Control estructural del volcán Copahue (38°S-71°O): implicancias tectónicas para el arco volcánico cuaternario (36-39°S) , Rev. Asoc. Geol. Argentina , 53 ( 3 ), 229 – 244 . OpenURL Placeholder Text WorldCat Folguera A. , Ramos V.A. , Melnick D. , 2003 . Recurrencia en el desarrollo de cuencas de intraarco.: Cordillera Neuquina (37° 30'-38°S) , Rev. Asoc. Geol. Argentina , 58 ( 1 ), 3 – 19 . OpenURL Placeholder Text WorldCat Folguera A. , Vera E.R. , Vélez L. , Tobal J. , Orts D. , Agusto M. , Caselli A. , Ramos V. , 2016 . A review of the geology, structural controls, and tectonic setting of Copahue volcano, Southern Volcanic Zone, Andes, Argentina , in Copahue Volcano , pp. 3 – 22 ., eds Tassi F. , Vaselli O. , Caselli A. , Springer . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Geshi N. , Shimano T. , Chiba T. , Nakada S. , 2002 . Caldera collapse during the 2000 eruption of Miyakejima Volcano, Japan , Bull. Volcanol. , 64 ( 1 ), 55 – 68 . Google Scholar Crossref Search ADS WorldCat Giordano G. , Porreca M. , Musacchio P. , Mattei M. , 2008 . The Holocene Secche di lazzaro phreatomagmatic succession (Stromboli, Italy): evidence of pyroclastic density current origin deduced by facies analysis and AMS flow directions , Bull. Volcanol. , 70 ( 10 ), 1221 – 1236 . Google Scholar Crossref Search ADS WorldCat Graham J. , 1954 . Magnetic anisotropy, an unexploited petrofabric element , Geol. soc. Am. Bull. , 65 , 1257 – 1258 . OpenURL Placeholder Text WorldCat Groeber P. , 1925 . La región de copahue y su glaciación diluvial , Soc. Argentina Estud. Geogr. , 1 ( 3 ), 92 – 110 . OpenURL Placeholder Text WorldCat Gudmundsson M.T. et al. . , 2016 . Gradual caldera collapse at Bárdarbunga volcano, Iceland, regulated by lateral magma outflow , Science , 353 ( 6296 ), aaf8988 , doi: 10.1126/science.aaf8988 . Google Scholar Crossref Search ADS PubMed WorldCat Hargraves R. , Johnson D. , Chan C. , 1991 . Distribution anisotropy: the cause of ams in igneous rocks? . Geophys. Res. Lett. , 18 ( 12 ), 2193 – 2196 . Google Scholar Crossref Search ADS WorldCat Harrison R.J. , Feinberg J.M. , 2008 . FORCinel: An improved algorithm for calculating first-order reversal curve distributions using locally weighted regression smoothing , Geochem.Geophys. Geosyst. , 9 ( 5 ), doi:10.1029/2008GC001987 . OpenURL Placeholder Text WorldCat Hirt A. , Evans K. , Engelder T. , 1995 . Correlation between magnetic anisotropy and fabric for devonian shales on the Appalachian Plateau , Tectonophysics , 247 ( 1-4 ), 121 – 132 . Google Scholar Crossref Search ADS WorldCat Hounslow M. , 1985 . Magnetic fabric arising from paramagnetic phyllosilicate minerals in mudrocks , J. Geol. Soc. 142 ( 6 ), 995 – 1006 . Google Scholar Crossref Search ADS WorldCat Hrouda F. , 1982 . Magnetic anisotropy of rocks and its application in geology and geophysics , Geophys. Surv. , 5 ( 1 ), 37 – 82 . Google Scholar Crossref Search ADS WorldCat Hrouda F. , 1994 . A technique for the measurement of thermal changes of magnetic susceptibility of weakly magnetic rocks by the CS-2 apparatus and KLY-2 kappabridge , Geophys. J. Int. , 118 ( 3 ), 604 – 612 . Google Scholar Crossref Search ADS WorldCat Hrouda F. , 2003 . Indices for numerical characterization of the alteration processes of magnetic minerals taking place during investigation of temperature variation of magnetic susceptibility , Stud. Geophys. Geod. , 47 ( 4 ), 847 – 861 . Google Scholar Crossref Search ADS WorldCat Jackson M. , 1991 . Anisotropy of magnetic remanence: a brief review of mineralogical sources, physical origins, and geological applications, and comparison with susceptibility anisotropy , Pure appl. Geophys. , 136 ( 1 ), 1 – 28 . Google Scholar Crossref Search ADS WorldCat Jelínek V. , 1977 . The Statistical Theory of Measuring Anisotropy of Magnetic Susceptibility of Rocks and its Application , Geofyzika Brno . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jelinek V. , 1981 . Characterization of the magnetic fabric of rocks , Tectonophysics , 79 ( 3-4 ), T63 – T67 . Google Scholar Crossref Search ADS WorldCat Jelínek V. , Kropáček V. , 1978 . Statistical processing of anisotropy of magnetic susceptibility measured on groups of specimens , Stud. Geophys. Geod. , 22 ( 1 ), 50 – 62 . Google Scholar Crossref Search ADS WorldCat Khan M.A. , 1962 . The anisotropy of magnetic susceptibility of some igneous and metamorphic rocks , J. geophys. Res. , 67 ( 7 ), 2873 – 2885 . Google Scholar Crossref Search ADS WorldCat Kodama K.P. , 1995 . Magnetic fabrics , Rev. Geophys. , 33 ( S1 ), 129 – 135 . Google Scholar Crossref Search ADS WorldCat LaBerge R.D. , Porreca M. , Mattei M. , Giordano G. , Cas R.A. , 2009 . Meandering flow of a pyroclastic density current documented by the anisotropy of magnetic susceptibility (AMS) in the quartz latite ignimbrite of the Pleistocene Monte Cimino volcanic centre (central Italy) , Tectonophysics , 466 ( 1-2 ), 64 – 78 . Google Scholar Crossref Search ADS WorldCat Linares E. , Ostera H.A. , Mas L. , 1999 . Cronologia potasio-argon del complejo efusivo Copahue-Caviahue, Provincia del Neuquen , Rev. Asoc. Geol. Argentina , 54 ( 3 ), 240 – 247 . OpenURL Placeholder Text WorldCat MacDonald W.D. , Ellwood B.B. , 1987 . Anisotropy of magnetic susceptibility: Sedimentological, igneous, and structural-tectonic applications , Rev. Geophys. , 25 ( 5 ), 905 – 909 . Google Scholar Crossref Search ADS WorldCat Magee C. , O’Driscoll B. , Petronis M. , Stevenson C. , 2016 . Three-dimensional magma flow dynamics within subvolcanic sheet intrusions , Geosphere , 12 ( 3 ), 842 – 866 . Google Scholar Crossref Search ADS WorldCat Martín-Hernández F. , Lüneburg C. , Aubourg C. , Jackson M. , 2004 . Magnetic fabric: methods and applications—an introduction , Geol. Soc. London , 238 ( 1 ), 1 – 7 . Google Scholar Crossref Search ADS WorldCat Mazzoni L. , Licitra D.T. , 2000 . Significado estratigráfico y volcanológico de depósitos de flujos piroclásticos neógenos con composición intermedia en la zona del lago Caviahue, Provincia del Neuquén , Rev. Asoc. Geol. Argentina , 55 ( 3 ), 247 – 249 . OpenURL Placeholder Text WorldCat Melnick D. , Folguera A. , Ramos V.A. , 2006 . Structural control on arc volcanism: the Caviahue–Copahue complex, Central to Patagonian andes transition (38°S) , J. South Am. Earth Sci. , 22 ( 1-2 ), 66 – 88 . Google Scholar Crossref Search ADS WorldCat Muñoz J. , Stern C. , 1988 . The Quaternary volcanic belt of the southern continental margin of South America: transverse structural and petrochemical variations across the segment between 38°S and 39°S , J. South Am. Earth Sci. , 1 ( 2 ), 147 – 161 . Google Scholar Crossref Search ADS WorldCat Ort M.H. , De Silva S.L. , Jiménez C N. , Jicha B.R. , Singer B.S. , 2013 . Correlation of ignimbrites using characteristic remnant magnetization and anisotropy of magnetic susceptibility, Central Andes, Bolivia , Geochem. Geophys. Geosyst. , 14 ( 1 ), 141 – 157 . Google Scholar Crossref Search ADS WorldCat Ort M.H. , Newkirk T. , Vilas J. , Vazquez J. , 2015 . Towards the definition of AMS facies in the deposits of pyroclastic density currents , Geol. Soc. London , 396 ( 1 ), 205 – 226 . Google Scholar Crossref Search ADS WorldCat Pesce A. , 1989 . Evolución volcano-tectónica del complejo efusivo Copahue-Caviahue y su modelo geotérmico preliminar , Rev. Asoc. Geol. Argentina , 44 ( 1-4 ), 307 – 327 . OpenURL Placeholder Text WorldCat Potter D.K. , Stephenson A. , 1988 . Single-domain particles in rocks and magnetic fabric analysis , Geophys. Res. Lett. , 15 ( 10 ), 1097 – 1100 . Google Scholar Crossref Search ADS WorldCat Riel B. , Milillo P. , Simons M. , Lundgren P. , Kanamori H. , Samsonov S. , 2015 . The collapse of Bárharbunga caldera, Iceland , Geophys. J. Int. , 202 ( 1 ), 446 – 453 . Google Scholar Crossref Search ADS WorldCat Roberts A.P. , Pike C.R. , Verosub K.L. , 2000 . First-order reversal curve diagrams: A new tool for characterizing the magnetic properties of natural samples , J. geophys. Res. , 105 ( B12 ), 28 461 – 28 475 . Google Scholar Crossref Search ADS WorldCat Roberts A.P. , Heslop D. , Zhao X. , Pike C.R. , 2014 . Understanding fine magnetic particle systems through use of first-order reversal curve diagrams , Rev. Geophys. , 52 ( 4 ), 557 – 602 . Google Scholar Crossref Search ADS WorldCat Rochette P. , 1987 . Magnetic susceptibility of the rock matrix related to magnetic fabric studies , J. Struct. Geol. , 9 ( 8 ), 1015 – 1020 . Google Scholar Crossref Search ADS WorldCat Rochette P. , Vialon P. , 1984 . Development of planar and linear fabrics in Dauphinois shales and slates (French Alps) studied by magnetic anisotropy and its mineralogical control , J. Struct. Geol. , 6 ( 1-2 ), 33 – 38 . Google Scholar Crossref Search ADS WorldCat Rochette P. , Jackson M. , Aubourg C. , 1992 . Rock magnetism and the interpretation of anisotropy of magnetic susceptibility , Rev. Geophys. , 30 ( 3 ), 209 – 226 . Google Scholar Crossref Search ADS WorldCat Rochette P. , Aubourg C. , Perrin M. , 1999 . Is this magnetic fabric normal? A review and case studies in volcanic formations , Tectonophysics , 307 ( 1-2 ), 219 – 234 . Google Scholar Crossref Search ADS WorldCat Schindelin J. et al. . , 2012 . Fiji: an open-source platform for biological-image analysis , Nature Methods , 9 ( 7 ), 676 – 682 ., Google Scholar Crossref Search ADS PubMed WorldCat Sruoga P ., Consoli, C.V. , 2011 . El Volcán Copahue , in Congreso Geológico Argentino, Neuquén, Argentina , pp. 613 – 620 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Stephenson A. , Sadikun S.t. , Potter D. , 1986 . A theoretical and experimental comparison of the anisotropies of magnetic susceptibility and remanence in rocks and minerals , Geophys. J. Int. , 84 ( 1 ), 185 – 200 . Google Scholar Crossref Search ADS WorldCat Tarling D. , Hrouda F. , 1993 . The Magnetic Anisotropy of Rocks , Chapman and Hall , London, UK , 217 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tauxe L. , 1998 . Paleomagnetic principles and practice. kluwer academic publishers, dordrecht (299pp.). rise and growth of the tibet plateau , Science , 294 , 1671 – 1677 . OpenURL Placeholder Text WorldCat Varekamp J.C. , deMoor J.M. , Merrill M.D. , Colvin A.S. , Goss A.R. , Vroon P.Z. , Hilton D.R. , 2006 . Geochemistry and isotopic characteristics of the Caviahue-Copahue volcanic complex, Province of Neuquén, Argentina , Geol. Soc. Am. , 407 , 317 . OpenURL Placeholder Text WorldCat Wiegand M. , Trumbull R.B. , Kontny A. , Greiling R.O. , 2017 . An AMS study of magma transport and emplacement mechanisms in mafic dykes from the Etendeka Province, Namibia , Tectonophysics , 716 , 149 – 167 . Google Scholar Crossref Search ADS WorldCat Wing-Fatt L. , Stacey F.D. , 1966 . Magnetic anisotropy of laboratory materials in which magma flow is simulated , Pure appl. Geophys. , 64 ( 1 ), 78 – 80 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of The Royal Astronomical Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Mineralogical control on the magnetic anisotropy of lavas and ignimbrites: a case study in the Caviahue-Copahue field (Argentina)
JF - Geophysical Journal International
DO - 10.1093/gji/ggz483
DA - 2020-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/mineralogical-control-on-the-magnetic-anisotropy-of-lavas-and-DbAT8XYjDN
SP - 821
VL - 220
IS - 2
DP - DeepDyve
ER -