The Nautanen deformation zone in the Gällivare area of northern Sweden is a highly Cu-mineralized, magnetite-rich, large-scale shear zone with a long-lived (~100 m.y.) deformation, hydrothermal alteration, and mineralization history. This composite structure hosts the Aitik porphyry Cu-Au-Ag ± Mo deposit and several Cu-Au ± Fe ± Ag ± Mo occurrences assigned to the iron oxide copper-gold (IOCG) deposit class. The Nautanen deformation zone was a locus for polyphase deformation and intermittent metasomatic-hydrothermal activity that overprinted middle Orosirian (ca. 1.90–1.88 Ga) continental arc-related volcanic-plutonic rocks. The deformation zone is characterized by intense shearing fabrics that form a series of subvertical to moderately W-dipping, NNW-SSE–trending, first-order shear zones with oblique reverse kinematics and related NNE-SSW–oriented second-order shear zones that control hydrothermal alteration patterns and Cu-Au mineralization.

Hydrothermal alteration in the study area formed during several phases. Volcanic-volcaniclastic rocks to the east and west of the Nautanen deformation zone display low to moderately intense, pervasive to selectively pervasive (i.e., patchy zones or bands, disseminations) sericite ± feldspar, amphibole + biotite + magnetite ± tourmaline, and K-feldspar + hematite alteration. Both the amphibole + biotite and K-feldspar + hematite associations occur adjacent to NNW- and NE-oriented deformation zones and are locally associated with minor sulfide. Within the deformation zone, a moderate to intense biotite + amphibole + garnet + magnetite + tourmaline + sericite alteration assemblage is typically associated with chalcopyrite + pyrrhotite + pyrite and forms linear and subparallel, mainly NNW-oriented seams, bands, and zones that locally appear to overprint possibly earlier scapolite + sericite ± feldspar alteration. Late-stage epidote ± quartz ± feldspar alteration (retrograde saussuritization) forms selectively pervasive zones and epidote veinlets across the area and is partly related to brittle faulting.

A magnetite-amphibole-biotite–rich, penetrative S1 foliation records shortening during early Svecokarelian-related deformation (D1) and can be related to ca. 1.88 to 1.87 Ga arc accretion processes and basin inversion that overlaps with regional peak metamorphism to near mid-amphibolite facies conditions and a potential initial Cu mineralization event. Folding and repeated shearing along the Nautanen deformation zone can be assigned to a second, late-Svecokarelian deformation event (D2 stage, ca. 1.82–1.79 Ga) taking place at a higher crustal level. This D2 deformation phase is related to late-stage accretionary processes active during a transition to a stage of postorogenic collapse, and it was accompanied by abundant, syntectonic intrusions. D2-related magmatism produced high-temperature and low-pressure conditions and represents a regional magmatic-hydrothermal event that controlled the recrystallization/remobilization of magnetite, biotite, and amphibole. Associated shear zone reactivation during D2 favors the utilization of the Nautanen deformation zone as a fluid conduit, which preferentially controlled the siting and formation of epigenetic Cu-Au mineralization with distinctive IOCG characteristics within second-order shear zones.

The northern Norrbotten ore province in northern Sweden is one of the most active mining areas in Europe and hosts several important ore deposits, including the type locality for iron oxide-apatite (IOA) deposits at Kiruna. Prominent examples of world-class deposits in the Gällivare area of north-central Norrbotten are the Malmberget IOA and Aitik Cu-Au-Ag-(Mo) deposits (Figs. 1, 2). Although traditionally referred to as “epigenetic” Cu-Au mineralization (e.g., Martinsson, 2004), many hydrothermal Cu-Au ± Fe ± Co ± Ag ± Mo occurrences in the Gällivare area and across Norrbotten have been assigned to the enigmatic “iron oxide copper-gold (IOCG)” class of deposits based on their geologic, structural, mineralogical, and hydrothermal alteration characteristics (e.g., Carlon, 2000; Billström et al., 2010; Williams, 2010; Martinsson et al., 2016). In the case of Aitik, it is thought to represent a porphyry Cu-Au deposit that was subsequently overprinted by an IOCG mineralization event (Wanhainen et al., 2012). Diagnostic features of Norrbotten IOCG deposits include their vein, fracture-infill, breccia, disseminated chalcopyrite ± chalcocite ± bornite ± pyrite ± native Au ± electrum mineralization style, an association with sodic-calcic, potassic-ferroan, and/or propylitic alteration, and their position within or adjacent to ductile-brittle deformation zones that transect older metasupracrustal or intrusive rocks (Edfelt et al., 2005; Smith et al., 2010; Martinsson et al., 2016). Additionally, some occurrences show a degree of stratigraphic or lithological control (Lindblom et al., 1996; Martinsson et al., 2016), while a relatively close (~1–8 km) proximity to known IOA mineralization is also common, although not always present. Fluid inclusion studies indicate Norrbotten IOCG mineralization formed from moderate- to high-salinity, mesothermal aqueous and aqueous-carbonic fluids derived from a mix of magmatic and basinal brine sources (e.g., Lindblom et al., 1996; Gleeson and Smith, 2010; Wanhainen et al., 2012; Smith et al., 2013; Martinsson et al., 2016).

At the regional scale, important tectonothermal features that represent favorable precursor and/or contemporaneous controls for Norrbotten IOA-IOCG mineralization include the province’s location on the landward side of a paleocontinental margin (e.g., Skirrow et al., 2018), subduction-related, bimodal, calc-alkaline to alkaline magmatism and coeval basin formation, the presence of long-lived (reactive) crustal-scale shear zones and related structures, and multiple overprinting metasomatic and hydrothermal alteration events driven by episodic Paleoproterozoic (ca. 1.90–1.78 Ga) magmatism and/or metamorphism (e.g., Bergman et al., 2001; Martinsson, 2004; Smith et al., 2013; Bergman and Weihed, 2020). Such features also highlight the broad similarity between the northern Norrbotten ore province and IOA- and IOCG-prospective terrains elsewhere, including the Carajás mineral province in Brazil (e.g., deMelo et al., 2017; Craveiro et al., 2019), the Cloncurry IOCG district and Olympic mineral province of Australia (e.g., Skirrow et al., 2019), the Great Bear magmatic zone, Northwest Territories, Canada (e.g., Corriveau and Mumin, 2010; Corriveau et al., 2016) the Akjoujt area of Mauritania (e.g., Kolb et al., 2008), and the Coastal belt in Chile (Sillitoe, 2003).

Genetic models for IOA and IOCG deposits (sensu stricto) generally differ and include the segregation and crystallization of Fe- and volatile-rich melts and associated fluids for IOA systems (Nystrom and Henriquez, 1994; Naslund et al., 2002; Knipping et al., 2015a; Martinsson et al., 2016; Troll et al., 2019) and magmatic-hydrothermal, skarn-metasomatic, and/or hydrothermal replacement processes for IOCG systems (e.g., Hitzman et al., 1992; Barton and Johnson, 1996; Sillitoe and Burrows, 2002; Pollard, 2006; Tornos et al., 2016; Martinsson et al., 2016). Additionally, studies have suggested that a genetic and zonal continuum exists between magnetiteapatite and copper sulfide-rich IOCG (sensu lato) systems that partly explains overlapping mineralogical and geochemical characteristics for some IOA and IOCG deposits as well as their spatial and/or temporal coincidence (Knipping et al., 2015b; Martinsson et al., 2016; Reich et al., 2016; Apukhtina et al., 2017; Barra et al., 2017). Furthermore, links between magmatic-hydrothermal processes that form porphyry Cu ± Au ± Mo deposits and IOCG-style mineralization have been proposed (e.g., Pollard, 2006; Mumin et al., 2010; Richards and Mumin 2013), while a possible connection with sediment-hosted stratabound/stratiform Cu ± Co ± Ag mineralization may also be relevant for systems purely associated with nonmagmatic basinal fluids (Hunt et al., 2007).

Despite specific geologic or geochemical variations among IOCG systems, a key unifying factor during their formation is the availability of proximal controlling structures that enhance permeability, hydraulic conductivity, and hydrothermal fluid flow, leading to increased mineralization potential in favorable depositional sites (e.g., Williams et al., 2005; Groves et al., 2010; Barton, 2014; Skirrow et al., 2019). For IOCG deposits, the most common structural pathways and traps include second- and third-order (or higher) fault systems associated with major shear zones, dilational jogs or bends developed along deformation zones, fault intersections, faults that exploit lithological contacts, synshearing en echelon fault arrays, and fault systems occurring close to or within fold hinge zones (e.g., Ford and Blenkinsop, 2008; Austin and Blenkinsop, 2009; Hayward and Skirrow, 2010; Lopez et al., 2014). The interaction between early-formed ductile structures in host rocks and brittle faulting during overprinting deformation events may represent a critical developmental step in deposit formation (e.g., Duncan et al., 2014). Likewise, spatially focused hydrofracturing producing “open space-infilled” breccias, veins, veinlets, and vugs is an important structural-mechanical component commonly observed at meso- to microscales (e.g., Hunt et al., 2007; Kreiner and Barton, 2011). Thus, from an ore-forming perspective, numerous IOCG districts and deposits contain structural features that are spatially interlinked at district to grain scales and which act as important pathways and controls for mineralizing hydrothermal fluids (e.g., Duncan et al., 2014; Lopez et al., 2014; Haddad-Martim et al., 2018).

The geology of the Gällivare area is dominated by a major, approximately NNW-trending, Paleoproterozoic shear zone named the Nautanen deformation zone (Witschard, 1996; Bergman et al., 2001). This composite structure hosts the Aitik Cu-Au-Ag deposit and several other Cu-Au ± Mo occurrences along its length, indicating that the Nautanen deformation zone played an important role in controlling and focusing Cu-Au ± Ag ± Mo mineralization in the area (Figs. 1, 2; Martinsson and Wanhainen, 2004; Smith et al., 2009; McGimpsey, 2010; Lynch et al., 2015). The Aitik deposit (southern Nautanen deformation zone; Figs. 2, 3) has proven reserves of 726 Mt @ 0.22 % Cu, 0.15 g/t Au, and 1.2 g/t Ag, and probable reserves of 461 Mt @ 0.23 % Cu, 0.14 g/t Au, and 1.3 g/t Ag (New Boliden, 2019). The smaller Nautanen Cu-Au deposit (northern Nautanen deformation zone; Figs. 2, 3) is composed of several orebodies that were mined from 1902 to 1907 with a total production of 71,835 t of ore (Årebäck and Dean 2018). New mineralization recently delineated by Boliden at the Nautanen North deposit (Fig. 2) has an indicated mineral resource of 12.7 Mt @ 1.54% Cu, 0.9 g/t Au, 6.0 g/t Ag, and 100 g/t Mo, and an inferred resource of 8.7 Mt @ 1.37% Cu, 0.6 g/t Au, 6.0 g/t Ag, and 98 g/t Mo (Årebäck et al., 2020).

Previous deposit-related studies in the Gällivare area have mainly focused on the mineralogical and geochemical characteristics of the Aitik, Malmberget, and Nautanen deposits (e.g., Zweifel, 1976; Monro, 1988; Wanhainen, 2005; Mc-Gimpsey, 2010; Lund, 2013; Waara, 2015; Edblom 2020) and other Cu-Au prospects, such as Liikavaara Östra and Salmijärvi, close to Aitik (Fig. 2; Sarlus, 2013; Warlo, 2016). Wanhainen et al. (2012) reported on the deformation history of the Aitik deposit and connected it to regional ore-forming events. Recently, Bauer et al. (2018) described the deformation history of the Malmberget deposit to the west of the Nautanen deformation zone (Fig. 2) and characterized the ductile-brittle modification of IOA mineralization there. Sarlus et al. (2017) presented a revised temporal framework for Paleoproterozoic tectonothermal processes in the Gällivare area based on the petrology, geochemistry, and geochronology of ca. 1.90 and 1.80 Ga syn- to late-orogenic intrusive suites. Likewise, petrogenetic constraints for ca. 1.90 to 1.88 Ga volcanosedimentary rocks hosting IOA and Cu-Au mineralization in the Gällivare-Malmberget area are reported by Lynch et al. (2018) and Sarlus et al. (2017, 2019), emphasizing coupled magmatic-sedimentary processes within a continental intra- to back-arc setting at that time.

Despite these efforts, a unifying tectonic-metallogenic model for IOA and/or Cu-Au ± Ag ± Mo mineral systems in the Gällivare area based on combined structural mapping, hydrothermal mineral characterization, and 3-D modeling has remained underdeveloped and/or unrevised. Specifically, work that constrains the structural character and kinematic evolution of the Nautanen deformation zone and identifies links between deformation and fluid flow events is justified for this structure, given its apparent role as a locus for Cu-Au ± Ag ± Mo mineralization and related hydrothermal alteration. Consequently, a better understanding of this regionally important metalliferous deformation zone may enhance the prospectivity of analogous Paleoproterozoic shear zones hosting IOCG mineralization in northern Fennoscandia and elsewhere (e.g., Niiranen et al., 2007; Corriveau and Mumin, 2010).

This study presents a structural investigation of the Gällivare area, with a specific focus on the Nautanen deformation zone and adjoining areas (Figs. 24). New structural mapping and microstructural analysis are integrated with airborne magnetic anomaly data (from the Geological Survey of Sweden) and 3-D geomodeling to constrain the structural and kinematic characteristics of the Nautanen deformation zone and to establish a revised deformational evolution for the area. Additionally, an assessment of mappable hydrothermal alteration occurring along the Nautanen deformation zone is presented to identify coupled deformation-hydrothermal processes and to highlight paragenetic relationships between specific structures and alteration mineral associations linked to Cu-Au mineralization. The results help constrain the deformation-alteration characteristics of one of the most Cu- and Au-endowed Paleoproterozoic shear zones in northern Fennoscandia and, therefore, establish a new tectonic-metallogenic framework for IOA- and IOCG-style mineralization in this part of the northern Fennoscandian Shield.

Regional geologic setting

The Mesoarchean to Paleoproterozoic geology of the northern Fennoscandian Shield reflects a complex geodynamic evolution that includes repeated extensional and compressional tectonic events and associated magmatism and metamorphism (e.g., Lahtinen et al., 2005, 2009; Bergman, 2018; Stephens and Bergman Weihed, 2020). The oldest rocks in the region belong to a ca. 3.5 to 2.7 Ga metamorphic complex that forms part of a composite Archean craton (Fig. 1; e.g., Lahtinen et al., 2005; 2009; Hölttä et al., 2008, 2019). Lithospheric-scale rifting of this craton is thought to have commenced by ca. 2.45 Ga along a major NW-directed line (present-day orientation) extending from Ladoga (Russia) to Lofoten (Norway; Martinsson, 1997). This extensional tectonism produced a large igneous province comprising ultramafic to mafic volcanic and plutonic rocks and associated sedimentary successions, including basal evaporitic deposits (e.g., Martinsson, 1997; Lahtinen et al., 2009; Melezehink and Hanski, 2012; Sharkov et al., 2017; Skyttä et al., 2019). Subsequent tectonic convergence coincided with the onset of the Svecokarelian (Svecofennian) orogeny, resulting in subduction-related magmatism at ca. 2.02 to 1.87 Ga, which formed several juvenile to mature continental arcs, intra- to back-arc extensional basins, voluminous intermediate to felsic volcanism-plutonism, and synvolcanic sedimentary deposits (Lahtinen, 1994, 2012; Nironen, 1997; Weihed et al., 2002; Lahtinen et al., 2005, 2009).

In northern Norrbotten, ca. 1.89 to 1.86 Ga continental arc-related magmatism is represented by calc-alkaline to alkalic volcanic successions of the Porphyrite and Kiirunavaara groups, respectively, and the comagmatic Haparanda and Perthite-monzonite intrusive suites (Bergman et al., 2001; Martinsson, 2004; Bergman, 2018). Throughout the region, synsubduction volcanic and sedimentary rocks are preserved within somewhat disconnected, linear to anastomosing supracrustal belts or domains. Based on space-time considerations, these supracrustal belts acted as a locus for large-scale, synorogenic composite shear zones and IOA-IOCG mineralization (Figs. 1, 2; e.g., Bergman et al., 2001). The partial overlap and geometric similarity (i.e., trend and form) between the synsubduction supracrustal belts and preorogenic (ca. 2.1 Ga), synrift greenstone successions also suggest that earlier-formed extensional basins may have provided a physiographic control for the location of Svecofennian supracrustal rocks and their contained structures (e.g., Skyttä et al., 2019).

The onset of accretion onto the Archean Norrbotten craton (e.g., Skellefte arc) occurred around 1.88 to 1.86 Ga, resulting in crustal shortening, basin inversion, deformation, and medium-grade regional metamorphism at upper greenschist-to middle amphibolite-facies conditions (Bergman et al., 2001; Weihed et al., 2002; Lahtinen et al., 2009; Bauer et al., 2011, 2013; Skyttä et al., 2012). Late Svecofennian granitoids in northern Norrbotten comprising ca. 1.81 to 1.78 Ga granites of the Lina suite and subordinate coeval mafic intrusive rocks (Bergman et al., 2001; Sarlus et al., 2017; Martinsson et al., 2018) represent a phase of intracrustal reworking and anatexis, and are locally associated with suggested low-pressure, high-temperature contact metamorphism (e.g., Sarlus, 2017; Sarlus et al., 2020). Local felsic magmatism in Norrbotten that is coeval with extensive I- to A-type felsic batholiths of the Transscandinavian igneous belt farther to the southwest is interpreted to have formed during eastward subduction along a N-S–oriented continental margin at around 1.81 to 1.65 Ga (Fig. 1; Andersson, 1991; Åhäll and Larson, 2000; Weihed at al., 2002).

Geology of the Gällivare-Malmberget area

The Gällivare-Malmberget area is composed of a deformed and metamorphosed package of late Orosirian (ca. 1.89–1.87 Ga) volcanic and sedimentary (epiclastic) rocks, comagmatic and variably deformed gabbroic, dioritic, and granitic intrusions, and a suite of late Orosirian to early Statherian (ca. 1.81–1.78 Ga) weakly deformed to massive granites, dolerites, and gabbros (Fig. 2; Sarlus et al., 2017; 2020). Witschard (1996) assigned the deformed supracrustal rocks to the informal “Porphyry group,” with meta-volcanosedimentary rocks in the east representing the lower parts of the succession, metavolcanic rocks at Malmberget forming the uppermost basal part, and felsic metavolcanic rocks at Malmberget and quartzites at Gällivare forming the middle to uppermost parts, respectively (Fig. 2). Subsequently, Martinsson and Wanhainen (2004) subdivided the sequence by assigning the meta-volcanosedimentary rocks east of Malmberget to the informal “Muorjevaara group” and the metavolcanic rocks at Malmberget to the “Kiirunavaara group,” with the latter inferred to be correlative with the volcanic sequence hosting the Kiirunanvaara IOA deposit at Kiruna based on lithological and geochemical constraints (Fig. 1; e.g., Martinsson et al., 2013). Definitive contact relationships between volcanic and epiclastic rocks are generally lacking, however, and recent geochronology results (e.g., Lynch et al., 2018; Sarlus et al., 2020) suggest that most of the supracrustal rocks in the Gällivare area were deposited contemporaneously from ca. 1885 to 1878 Ma (see below).

The Nautanen deformation zone and Cu-Au ± Ag ± Mo ± Fe mineralization

The Nautanen deformation zone, an approximately NNW-trending composite deformation zone, is a conspicuous feature of the Gällivare area (Figs. 14; Witschard, 1996; Martinsson and Wanhainen, 2004). About 10 km east of Gällivare, this zone forms a ~300-m- to 2-km-wide and at least 20-km-long, distinctive high-strain zone, characteristically delineated on aeromagnetic maps by planar, subparallel (striped), approximately NNW-aligned positive magnetic anomalies (e.g., Fig. 4). Internally, the Nautanen deformation zone is characterized by moderate to intense shearing, mylonitization, and structural transposition as well as pervasive metasomatic-hydrothermal alteration (Martinsson and Wanhainen, 2004; Lynch et al., 2015, 2018). Bergman et al. (2001) interpreted a mainly dextral-oblique shear sense for this composite structure based on the interpretation of regional structural and magnetic lineament geometries. The trend, scale, and internal structural character of the Nautanen deformation zone are partly replicated by other Paleoproterozoic shear zones in northern Norrbotten (e.g., Andersson et al., 2020).

Several Cu-Au ± Fe deposits and prospects occur within and adjacent to the Nautanen deformation zone (Martinsson and Wanhainen, 2004). Important examples include the Aitik Cu-Au-Ag-(Mo) deposit and the Nautanen, Nautanen North, Liikavaara Östra, and Ferrum deposits. Several styles of Cu-Au mineralization occur, including an older phase of disseminated and quartz vein-hosted related to porphyritic intrusions (e.g., Aitik; Wanhainen et al., 2006). Later phase mineralization is composed of shear zone-hosted, disseminated, semimassive, and pure sulfide veins/breccias (e.g., Aitik, Nautanen North; Martinsson and Wanhainen, 2004) and quartz ± tourmaline ± amphibole vein-hosted mineralization occurring mainly east of the Nautanen deformation zone (e.g., Ferrum prospect; Gustavsson and Johansson, 1984) or as a late-stage brittle overprint within the Nautanen deformation zone. Chalcopyrite with lesser bornite and chalcocite represent the main ore minerals and are typically associated with pyrrhotite, pyrite, and magnetite. Gold occurs as the native metal, amalgam (Au-Ag-Hg), and electrum (Au-Ag) as inclusions, grain boundary segregations, and fracture-infill associated with disseminated chalcopyrite, pyrite, and magnetite (Sammelin et al., 2011; Bark et al., 2013; Edblom, 2020).

Aitik represents Sweden’s largest Cu-sulfide deposit and is hosted within gneissic, variably altered metavolcaniclastic rocks (Wanhainen and Martinsson, 1999). Anomalous Cu and Au distributions are thought to be controlled by the structural remobilization of sulfides into second-order, NNE-SSW–trending, high-strain deformation zones (Wanhainen et al., 2003). In the structural footwall at Aitik, a quartz monzodiorite intrusion containing disseminated and localized quartz stockwork-style Cu-sulfide mineralization occurs (Wanhainen et al., 2006). Fluid inclusion data indicate that these veins formed at ~300°C and at approximately 3 kbar (lithostatic equivalent) during the early porphyry-style event that formed chalcopyrite and pyrite associated with potassic (biotite and microcline) and phyllic (sericite) alteration. Subsequent IOCG overprinting associated with amphibole-scapolite and K-feldspar alteration developed at 200° to 500°C and approximately 1 kbar (Wanhainen et al., 2012).

The historical Nautanen mining area (Fig. 2) is composed of several orebodies that are hosted by strongly altered and deformed rocks within the Nautanen deformation zone (Ros, 1980; Martinsson and Wanhainen, 2004). Mineralization typically consists of disseminated chalcopyrite with magnetite + pyrrhotite ± pyrite, or occurs as vein-style mineralization. Typical hydrothermal alteration minerals are scapolite, amphibole, microcline, epidote, biotite, sericite, and garnet, with varying occurrences of barite, carbonates, quartz, and tourmaline (Martinsson and Wanhainen, 2004). Garnet is often more abundant near the mineralization and forms large, preto syntectonic porphyroblasts (Ros, 1980; Danielsson, 1985; Tollefsen, 2014; Waara, 2015). The Nautanen North deposit (Fig. 2) is hosted by volcanosedimentary rocks and is strongly controlled by a NNW-striking, subvertical to steeply E-dipping, high-strain zone (Årebäck and Dean, 2018). Nautanen North is composed of two high-grade zones with mainly disseminated and fracture-controlled chalcopyrite ± magnetite ± pyrrhotite ± pyrite mineralization. In general, highly mineralized zones are enveloped by a lower-grade zone with disseminated chalcopyrite, pyrite, magnetite, and molybdenite (Årebäck and Dean, 2018).

Timing of tectonothermal and mineralization events

Radiometric dating of volcanic and plutonic rocks in the Gällivare-Malmberget area provides constraints on the timing of Paleoproterozoic magmatism and regional lithostratigraphic correlations (Fig. 4). Sarlus et al. (2020) reported zircon U-Pb secondary ion mass spectrometry (SIMS) ages of 1885 ± 6 and 1881 ± 6 Ma for Kiirunavaara group metavolcanic rocks hosting the Malmberget IOA deposit. These ages are identical to ca. 1.89 to 1.88 Ga zircon U-Pb dates obtained for correlative metavolcanic rocks hosting the Kiruna IOA deposit (Westhues et al., 2016). To the east of Malmberget, Lynch et al. (2018) obtained a zircon U-Pb SIMS age of 1878 ± 7 Ma for an andesitic metavolcanic rock within the Nautanen deformation zone. The dated horizon is intercalated with volcaniclastic-epiclastic rocks similar to those to the east and west of the shear zone and constrains a depositional age of ca. 1.87 Ga for the clastic units. Crystallization ages (zircon U-Pb SIMS) of 1882 ± 6 Ma for an andesitic metavolcanic rock about 40 km east-southeast of Gällivare and 1868 ± 6 Ma for a rhyolite porphyry interbedded and overlain by quartzite about 20 km south-southeast of Gällivare (Claeson et al., 2018a, b) further constrain the timing of synorogenic volcanism and sedimentation.

Igneous ages for plutonic rocks at Gällivare highlight two major phases of bimodal intrusive magmatism at ca. 1.89 to 1.87 and ca. 1.81 to 1.78 Ga. In the Aitik area (Figs. 2, 5), crystallization ages for plutonic rocks include a zircon U-Pb thermal ionization mass spectrometry (TIMS) age of 1887 ± 8 Ma for a foliated quartz monzodiorite in the footwall of the deposit (Wanhainen et al., 2006) and a zircon U-Pb SIMS age of 1883 ± 5 Ma for a foliated granite ~3 km to the southwest (Sarlus et al., 2017). To the east of Aitik, at the Liikavaara Östra Cu-(W-Au) deposit, Warlo et al. (2020) reported a crystallization age (zircon U-Pb SIMS) of 1874 ± 5 Ma for a foliated granodiorite in the deposit’s structural footwall. This age is identical to a zircon U-Pb SIMS age of 1874 ± 4 Ma obtained for a quartz monzonite (Perthite-monzonite suite) located ~50 km west of Gällivare (Kathol and Hellström, 2018). Both ages also overlap with a reported 1870 ± 12 Ma date (zircon U-Pb SIMS) for a weakly foliated diorite in the Nautanen area (Sarlus et al., 2017). To the west of Aitik, Sarlus et al. (2017) also obtained an age of 1883 ± 5 Ma (zircon U-Pb SIMS) for a gabbroic unit forming part of a larger ultramafic-mafic intrusive complex. Crystallization ages (zircon U-Pb TIMS, SIMS, and laser ablation-inductively coupled plasma-mass spectrometry [LA-ICP-MS] dating) ranging from ca. 1.81 to 1.78 Ga have been reported for voluminous granite and subordinate gabbro-dolerite and monzonite-syenite intrusions in the general Gällivare-Malmberget area (Bergman et al., 2002; Sarlus et al., 2017; Claeson et al., 2018c).

Metamorphic overprinting affected the Gällivare area during two separate events at ca. 1.88 to 1.87 and ca. 1.80 to 1.78 Ga (Fig. 5). The first (M1) is recognized as a regional prograde metamorphic event that reached greenschist- to upper amphibolite-facies conditions (Bergman et al., 2001; Bauer et al., 2018). Based on structural and mineralogical relations at Malmberget, Bauer et al. (2018) suggested that peak M1 regional metamorphism coincided with the first major com pressional deformation phase (D1) in the area. Subsequently, a second compressional phase (D2) defined by approximately E-W–directed deformation developed during mainly high-temperature, low-pressure metasomatism (i.e., M2 event). Rim domains on magmatic zircons from Malmberget volcanic rocks yielded ages of 1797 ± 7 and 1775 ± 6 Ma and were interpreted by Sarlus et al. (2020) to represent a contact metamorphic event at ca. 1.80 to 1.78 Ga. In the Nautanen area, zircon crystals from a ca. 1.87 Ga diorite also record a younger, ca. 1.79 to 1.78 Ga tectonothermal event (Sarlus et al., 2017).

Some radiometric ages have been determined for depositproximal minerals in the Gällivare area. At the Aitik deposit, 16 titanite U-Pb and molybdenite Re-Os ages record episodic magmatic-hydrothermal activity at 1876 ± 10, 1848 ± 8, 1805 ± 6 to 1773 ± 2, and 1728 ± 7 Ma (Wanhainen el al., 2005). The younger age marks emplacement of an undeformed pegmatite with minor molybdenite and chalcopyrite, suggesting a relatively late phase of magmatic-hydrothermal activity that postdates compressive deformation and metamorphism (Fig. 4). Fracture-hosted apatite-stilbite-calcite ± monazite mineralization at the Malmberget deposit, dated to 1730 ± 6.4 Ma (Romer, 1996), further highlights a ca. 1.74 to 1.73 Ga stage of postmetamorphic brittle deformation and fluid flow. Smith et al. (2009) reported U-Pb LA-ICP-MS ages ranging from 1785 ± 21 to 1777 ± 20 Ma for hydrothermal titanite and allanite in mylonitic and altered schist hosting Cu-Au mineralization. A comparable (although less precise) age range of ca. 1.83 to 1.78 Ga is also reported for hydrothermal titanite at the Malmberget IOA deposit (Storey et al., 2007).

Structural mapping was conducted along 10 approximately E-W–oriented transects across the Nautanen deformation zone and adjoining areas to the east of Malmberget. A total of 2,500 structural measurements and 70 oriented bedrock samples were taken from 950 outcrop observations. Regional geology and geophysical maps (1:50,000-scale) from the Geological Survey of Sweden (e.g., Witschard, 1996) were used as a mapping base and for locating outcrops. Field mapping was performed digitally using FieldMove (Petroleum Experts Ltd.) on portable iPad Mini devices (Apple). Structural measurements were taken with Brunton Geo compasses using the dip-direction/dip convention. As magnetic disturbance is prevalent along the Nautanen deformation zone, magnetic declination was frequently checked by taking bearings on topographical and geologic maps using known reference lines. We estimate the error on structural readings in highly magnetic outcrops falls in the range of 5° to 10°. Follow-up characterization of microstructures and alteration types was performed by transmitted and reflected light optical microscopy of 65 thin sections.

Alteration mineral associations were recorded and mapped at outcrops in order to identify spatial and/or paragenetic links between mapped structures and alteration types, and to assess coupling between deformation and hydrothermal activity. Additionally, unoriented drill core from the Nautanen North, Nautanen, and Aitik deposits that intersects chalcopyrite mineralization was inspected to further assess alterationdeformation links at three key deposits from north to south along the Nautanen deformation zone (e.g., Fig. 2). In this paper, “alteration mineral association” refers to petrographically characteristic groupings of two or more hydrothermal minerals formed under either equilibrium or disequilibrium conditions, with key indicator minerals listed in decreasing relative abundance (e.g., Gifkins et al., 2005). Corresponding generic and/or compositional alteration terms (e.g., sericitic, potassic) have been used sparingly and are based on those listed in Thompson and Thompson (1996). For the listed alteration associations, the “+” symbol adjoins commonly associated minerals, while the “±” symbol precedes minerals only occasionally or more locally developed. By way of contrast, vein-type constituent minerals are quoted and adjoined with the “-” symbol.

Structural analysis and processing of field data were performed using Move 2017 (Petroleum Experts Ltd.) and Arc-GIS (ESRI). For geologic 3-D modeling, structural measurements were imported into Move 2017 and the locations and configurations of regional-scale ductile shear zones and brittle deformation zones were subsequently digitized onscreen using the field observations and magnetic anomaly maps. Due to the sparse nature and uneven distribution of the structural data, modeling of large shear systems was performed in GoCAD (Paradigm), utilizing an implicit approach with the Sparse plugin (Mira Geoscience). This software generates construction surfaces that honor structural measurements from field mapping and geophysical map trace interpretations. Applying the same method when modeling the folded Nautanen deformation zone footwall and hanging-wall complexes from structural measurements resulted in unrealistic fold geometries that did not reflect those observed in the field. Laurent et al. (2016) showed that most implicit modeling algorithms at the time were not able to reproduce realistic polydeformed fold structures such as refolded folds. This modeling limitation may also be relevant for strongly noncylindrical folds such as those east and west of the Nautanen deformation zone. Hence, an explicit modeling approach was chosen to model these structures where, prior to modeling, the fold systems were analyzed for fold axis plunges for all parts of the fold and fold axis curvatures were determined. Consequently, relevant fold structures were modeled in Move 2017 based on their map traces derived from magnetic anomaly maps and the obtained fold axis geometries. Fold modeling utilized the extrusion method in Move 2017, allowing fold axes plunges and trends to be determined and allowing the manual altering of the shape of fold axes based on surface map traces.

Results from structural geology mapping show a variety of superimposed ductile and brittle structures that record a protracted, multiphase deformation history. Based on different deformation intensities and styles, the study area is subdivided into three structural domains (Fig. 4): the eastern volcanosedimentary domain, the Nautanen deformation zone domain, and the western volcanosedimentary domain. The overall structural architecture of the study area is visualized in the 3-D structural framework model (Fig. 6; App. 2).

Structures in the eastern volcanosedimentary domain

Volcanic and epiclastic rocks within the eastern volcanosedimentary domain are characterized by tight to isoclinal folding of primary depositional structures (S0), compositional banding, alteration banding (from selective replacement of primary bedding), and tectonic foliations (Fig. 7). The most common structural fabric in the eastern volcanosedimentary domain is a variably intense, planar penetrative foliation, here designated S1. It is generally approximately NW- to NNW-aligned and moderate to steeply SW to WSW-dipping, and it tends to parallel inferred primary bedding, laminae, and compositional banding in meta-volcanosedimentary rocks (Fig. 7A) that were transposed into the plane of deformation. Additionally, S1 has a similar orientation to planar foliations in eastern volcanosedimentary domain-hosted dioritic intrusions (see “Structures associated with intrusive rocks”, below). The intensity of S1 varies between outcrops and is controlled by host rock lithologies. In areas with relatively high strain, the foliation represents a schistose to gneissic fabric defined by an alignment of amphibole, biotite, feldspar, and quartz (Fig. 7B). In lower-strain areas, S1 planar structures are developed as weak foliations cutting inferred bedding contacts (S0) at a low angle (Fig. 7C) or they are parallel to S0 structures, forming a composite S0/1 fabric. Rarely in the eastern volcanosedimentary domain, tight to isoclinal and asymmetric intrafolial folds (F1) occur within S1 composite fabrics and have dominantly northwest to north-northwest orientations (Fig. 7A).

To the immediate east of the Nautanen deformation zone domain, the dominant structure is a distinctive, asymmetrical, and overturned NNW-aligned synform with an approximate wavelength of 5 km (Figs. 24, 6; App. 2; e.g., Zweifel, 1976). The western limb appears to be truncated by Nautanen deformation zone-related, approximately NNW-trending shear zones and faults. Fold vergences are typically eastward, with axial surfaces approximately N- to NNW-aligned (subparallel to the Nautanen deformation zone trend) and generally steeply dipping toward the west (Figs. 3, 4, 6; App. 2). The fold shapes are noncylindrical and fold axes are locally curvilinear and doubly plunging toward the north and south (Figs. 2, 4, 6; App. 2). In general, first-order fold structures are accompanied by parasitic, asymmetric folding (Fig. 7E). In the southern part of the eastern volcanosedimentary domain, parasitic fold axes commonly plunge with gentle to moderate angles (5°–35°) toward the south-southeast, whereas in the northern parts, parasitic fold axes have doubly plunging geometries (typically approx. northwest and southeast). Mineral lineations developed as dynamic recrystallization and stretching of minerals on foliation planes and are gently to moderately plunging (10°–40°; Fig. 4). Overall, the structural relationships between the first- and second-order folding, S1 foliations with rare F1 folds, and the folding of S1 foliation suggest a later timing, and these later structural units are designated here as F2 folds.

A discordant, bedding plane-orthogonal crenulation cleavage, here designated S2, is defined by realignment of micas and occurs at a few localities in the eastern volcanosedimentary domain (Fig. 7D). It is generally approximately N-aligned, subvertical, and axial planar to upright and moderately approximately S- to SSE-plunging F2 folds. The S2 cleavage is associated with L2 bedding-cleavage intersection lineations that typically have a moderate plunge to the south and south-southeast, similar to the F2 fold axes (Fig. 4). Locally, near the hinge zones of first-order F2 folds, relatively intense S2/L2 deformation has developed elongated and stretched L>S-to L-tectonites that form mullion-like features along bedding surfaces (Fig. 7F). Additionally, in eastern F2 limb areas, local bedding plane surfaces containing subordinate L2 intersection lineations show slickensides that indicate top-block reverse movement toward the north and north-northwest (Fig. 7G).

Brittle structures (S3) in the eastern volcanosedimentary domain are only weakly developed and consist of the following: (1) locally developed fracture sets that tend to follow earlier planar structures, (2) numerous approximately NNW- to E-aligned, generally subvertical amphibole and quartz vein sets, of which the latter are locally chalcopyrite-pyrrhotite-pyrite–bearing (e.g., Ferrum Cu-Au prospect), and (3) joint sets developed in intrusive and supracrustal rocks. Additionally, discordant, approximately N- to NNE-aligned deformation zones form distinct negative anomalies in aeromagnetic data (Fig. 4) that cut all previously described folds and cleavages. These crosscutting, locally Nautanen deformation zone-related, high-strain zones segment the eastern volcanosedimentary domain into several localized structural blocks.

Structures in the Nautanen deformation zone domain

Structurally, the Nautanen deformation zone is characterized by a conspicuous, approximately NNW- to NW-aligned, steep to locally moderate approximately W-dipping, penetrative foliation with varying but generally strong intensity (Fig. 8A). This locally mylonitic foliation is defined by the dynamic recrystallization of quartz, an orientation of amphibole lenses, preferentially aligned magnetite grains and biotite aggregates, and the stretching of lithic clasts parallel to the strain fabric in volcaniclastic rocks. Locally observed primary bedding, compositional banding, and selective alteration of primary bedding are typically transposed into steep orientations parallel to the dominant Nautanen deformation zone NNW-aligned foliation and locally produce a composite S0/1/2 fabric. In low-strain blocks within the Nautanen deformation zone, the foliation is only weakly developed and can overprint S0/1 at an oblique angle, consistent with a later S2 designation (Fig. 8B).

The dominant shearing direction within the Nautanen deformation zone strikes approximately north-northwest to south-southeast and shows mainly west-block-up reverse kinematics, common oblique dextral components, and less common sinistral components (Fig. 8F, G), evident as asymmetric foliation deflections around garnet porphyroblasts (Fig. 8C), local asymmetric kink bands (Fig. 8D), asymmetric folding, S-C fabrics, and offset of layers (e.g., App. 1). Several garnet porphyroblasts show syntectonic microstructures and mineral inclusions with traces of rotation, following the definition of Passchier and Trouw (2005). The NNW-SSE–trending reverse shear zones are interconnected by N-S–trending, subvertical, second-order shear zones with mainly dextral kinematics evident from asymmetric intrafolial folding, asymmetric foliation deflections around porphyroblasts, and kink bands (Figs. 2, 6; Supplementary online material 1). The shear fabrics and internal character of these second-order zones are similar to the primary (first-order) reverse structures but can be distinguished in the field based on a clear deviation in orientation. Analogous structures have been suggested to control quartz-tourmaline veins in the Nautanen north deposit (Årebäck and Dean, 2018) and the distribution of gold and copper grades in the Aitik deposit (Wanhainen et al., 2003). These second-order, syn-S2 structures are partly intruded and crosscut by pegmatite dikes (Fig. 9A, B). Additionally, local tensional features such as en echelon quartz veins with an approximate north-northeast to south-southwest orientation were observed along the western margin of the Nautanen deformation zone (Fig. 8E; see below).

Mineral lineations in the Nautanen deformation zone are defined by stretching of silicate minerals, and their orientations tend to change along strike. In the northern parts of the Nautanen deformation zone (vicinity of the Nautanen deposits), lineations plunge 50° to 70° toward the south and southwest. Local deviations with gentle to horizontal plunges were also observed (Figs. 2, 8F). In the southern parts, around the Aitik deposit, mineral lineations plunge downdip on the foliation planes (30°–60°) toward the west to locally southwest (Figs. 2, 8G). Minor asymmetric, shear-related F2 folds are mainly evident from asymmetrically folded hornblende, magnetite- and epidote-filled veinlets. The axial planes of these minor F2 folds are typically parallel to the main S2 shear fabrics developed in the volcanic-volcaniclastic host rocks.

From north to south along the Nautanen deformation zone, dip angles of the reverse, first-order S2 shear zones decrease (from 85° in the north to 40° in the south) and a subtle change in orientation can be observed. At the Aitik deposit (e.g., Fig. 2), first-order shear zones are more N-S–oriented and are moderately dipping (40°–55°) toward the west with distinct, approximately W-plunging mineral lineations (Figs. 2, 3, 9A). The first-order shear zones also show reverse, west-block-up kinematics, evident from asymmetric folds in the hanging wall (Fig. 9B; App. 1; e.g., Wanhainen et al., 2012) and asymmetric foliation deflections around andalusite porphyroblasts (Fig. 10). The andalusite porphyroblasts show internal, intertectonic microstructures according to the definition of Passchier and Trouw (2005; Fig. 10). The main shear zones (first order) form distinct, approximately 2- to 20-m-wide deformation zones with relatively intense foliations defined by the alignment of mainly muscovite, which forms a continuous, partly anastomosing cleavage. The mylonitic shear zones exhibit sharp contacts to the hanging-wall and footwall rocks and can contain angular to subrounded clasts of wall rock, hence forming local cataclastic breccias with fine-grained, mica-rich matrix. Brittle slip surfaces within the muscovite fabric are common and show a certain amount of late, brittle reactivation.

Brittle deformation in the Nautanen deformation zone is composed of localized, subvertical fault zones and zones with increased fracture frequency. Both host rocks and previous structures have been affected by brittle fracturing (S3) and jointing. These fractures typically form sharp, planar features that are generally continuous on an outcrop scale and that exhibit highly varying fracture frequencies ranging from 10-cm to 15-m spacing. Slickensides are rare and cm-scale displacements show that the overall amount of movement was low. The brittle fractures typically form a conjugate pattern with three dominating orientations: a subvertical NW-SE–striking set, a subvertical NNE-SSW–striking set, and a subhorizontal set (Fig. 4). Locally, a brittle reactivation of shear zones is manifested as narrow, foliation-parallel fault zones.

Structures in the western volcanosedimentary domain

The western volcanosedimentary domain is characterized by alternating layers of sedimentary, volcanosedimentary, and volcanic rocks that contain locally distinct planar crossbedding (S0) and tectonic foliations (S1), forming a composite S0/1 planar fabric. Like the eastern volcanosedimentary domain, bedding and S1 foliations in the western domain have been folded into inclined to overturned and asymmetric, approximately N- to NNE-oriented folds (F2), with open to closed interlimb angles (Figs. 24). The dominant structures in the western volcanosedimentary domain are a repetition of antiforms and synforms with noncylindrical, curvilinear fold axes showing approximately north-south to north-northeast–south-southwest orientations and gentle plunges toward north and south (5°–20°). A major F2 fold axis in the northwest part of the western volcanosedimentary domain is doubly plunging, with gentle and moderate plunges toward north and south, respectively (Fig. 2). Local crenulation lineations, small-scale fold axes, and mineral stretching lineations are developed on phyllitic foliation planes and are oriented subparallel to the larger-scale F2 fold axis (Fig. 11). Planar S2 fabrics that are parallel to F2 fold axial surfaces have not been observed in the western volcanosedimentary domain.

Brittle deformation in the western volcanosedimentary domain is dominated by a large, NW-SE–striking, subvertical fault zone that divides the foliated and folded volcanic and sedimentary rocks to the north from mainly undeformed granites to the south (Figs. 2, 6). This fault zone corresponds to a low magnetic anomaly zone (Fig. 4), is internally characterized by intense fracturing and fracture-parallel barren quartz veining of the volcanosedimentary rocks, and forms a distinctive topographical depression in the area. The subparallel fractures are narrow, sharp, planar features with high fracture frequencies. No clear displacement or kinematics could be determined in the field, but displacement patterns on the magnetic anomaly map (Fig. 4) tentatively suggest reverse, south-side-up movement.

Structures associated with intrusive rocks

Intrusive rocks in the study area have varying strain intensities reflecting their relative ages and emplacement histories. For Haparanda-type dioritic to granodioritic intrusions, a commonly developed penetrative foliation is observed as a preferential alignment of amphibole, biotite, and/or feldspar crystals (Fig. 12A). In the eastern volcanosedimentary domain, this planar fabric is predominantly NNW- to NW-aligned, is steeply SW-dipping, and tends to follow the orientations of dominant S1 foliations occurring in adjacent country rocks. Locally, dioritic dikes intruding metavolcaniclastic rocks display evidence of apparent relative dextral-oblique emplacement associated with drag folding (e.g., Fig. 12B), whereas high-strain zones crosscutting dioritic intrusions produce a strong transposition of the penetrative foliation, suggesting foliation development preceded shear formation. Along the western margin of the Nautanen deformation zone, dioritic intrusions also occur as narrow, elongate, NNW-aligned bodies that are oriented subparallel to, and deflected by, folding and Nautanen deformation zone-related shearing.

Relatively small dolerite intrusions occur locally in the study area and show varying degrees of strain. The dolerites are generally equigranular, although examples with porphyritic textures with larger plagioclase laths occur locally. Plagioclase laths show either random orientations or are oriented into a preferred orientation which may change over centimeter to meter scales. At the Aitik deposit, dolerites are emplaced as sills parallel to S2 foliations developed within the major N-S–aligned reverse shear zone that separates the hanging wall and the mineralized zone (Fig. 12C). Contacts with wall rocks are sharp and the dolerites locally show pinch-and-swell structures (Fig. 12C). From the character of their foliation intensities and emplacement geometries, the dolerites are interpreted as syntectonically emplaced during a D2 deformation event.

Granites of the Lina suite and subordinate aplite-pegmatite bodies typically occur as sheetlike intrusions, sills, and dikes that intrude parallel to earlier-formed S0-1 planar structures (Fig. 12D) and the contacts of larger dioritic and granitic intrusive bodies. Grain size varies from fine to very coarse (5 mm to >3 cm) over a distance of a few 10s of centimeters, with an accompanying transitional gradation from granite to pegmatite parts. A varying degree of strain is observable in these granitic rocks, ranging from a weak to moderate penetrative foliation (S2) to nonfoliated (massive). Foliation intensity varies sometimes from centimeters to meters. Typically, the granitic bodies crosscut earlier-formed (D1) planar foliations and shear fabrics in the country rocks. Along intrusive contacts, recrystallization of amphibole, magnetite, and biotite are common features (see below, “Vein types and vein-related alteration”). Pinch-and-swell and boudinaged shapes are common (Fig. 9), while drag folds are also observable. At Aitik, pegmatitic dike swarms mimic the orientations of subvertical, NE-SW–striking, second-order structures developed between the first-order shear zones and have orientations that are axial planar, parallel to asymmetric F2 folds (Fig. 9). Overall, the variable degree and character of the foliations preserved in these granitic bodies, along with their local boudinaged shapes and orientations that are parallel with F2 fold axes trends, suggest a syn-D2 emplacement.

Metavolcanosedimentary rocks within and adjacent to the Nautanen deformation zone have been affected by varying degrees of metasomatic-hydrothermal alteration (e.g., Martinsson and Wanhainen, 2004; McGimpsey, 2010; Tollefsen, 2014). Based on magnetic anomaly signatures (Fig. 4), and assuming that magnetic mineral distributions in IOCG-prospective terranes reflect episodes of hydrothermal fluid flow (e.g., Corriveau et al., 2010), a preliminary assessment of the area indicates that relatively intensely developed alteration is spatially confined to the Nautanen deformation zone domain and overlaps with high-strain deformation (Figs. 3, 13). Outside the Nautanen deformation zone domain, alteration intensity is generally lower and is confined to localized areas with known Cu ± Au mineralization (e.g., Liikavaara Östra, Ferrum) or controlling/favorable structural-lithological features. In general, two alteration styles occur across the study area (Table 1): irregular to banded replacement zones and patches occurring over 10s of centimeters to 100s of meters (i.e., pervasive to selectively pervasive style), and narrow alteration zones or halos associated with hydrothermal veins and breccias (i.e., selectively developed).

Hydrothermal alteration within the Nautanen deformation zone

The main hydrothermal alteration developed within the Nautanen deformation zone domain is composed of a moderate to intense amphibole + biotite + magnetite ± tourmaline association forming banded and linear replacement zones within the deformed metavolcanic-volcaniclastic rocks (Figs. 13, 14A; Table 1). Due to the texturally destructive nature of the alteration, it is difficult to determine if one or more specific protoliths were preferentially affected by this replacement. Locally, this association appears to overprint a more pervasive and possibly earlier-formed scapolite ± sericite association (Fig. 14A). Scapolite, however, is also associated with synmineralization alteration (see below, “Structural alteration links and Cu ± Au mineralization”), while sericite locally occurs as a retrograde hydrothermal phase (replacing feldspar), suggesting at least two generations of these minerals. Amphibole + biotite + magnetite banding consists of individual, millimeter to centimeter-scale, laterally continuous bands that merge to form broader (10s of meters) linear alteration zones. Biotite, amphibole, and magnetite aggregates in this association are typically stretched and dynamically recrystallized and help define the main N- to NNW-oriented composite tectonic S1/2 foliation in the Nautanen deformation zone. This suggests that the amphibole + biotite + magnetite association formed prior to or during either D1 or D2 shearing. Magnetite abundance typically ranges from ~3 to 10 vol % within these alteration bands or zones.

A conspicuous alteration feature within the Nautanen deformation zone domain is the local presence of disseminated porphyroblasts and irregular, aggregated (granoblastic) zones of reddish-pink garnet (almandine-spessartine varieties; Monro, 1988; Tollefsen, 2014; Waara, 2015). Garnet porphyroblasts generally range from 0.2 to 5 cm in diameter (rarely, up to 30 cm), are typically subhedral, and often appear flattened and stretched parallel to the dominant NNW-oriented mylonitic foliation (Figs. 8C; 14A). Locally, garnet porphyroblasts deflect the main NNW-aligned S1/2 foliation and show curved internal trails of foliation traces, indicating synkinematic, metasomatic growth (Figs. 8C, 14A). Rare curving and partly folded garnet tails, combined with asymmetric S1/2 foliation deflections, suggest a degree of oblique grain rotation, mostly with a dextral shear sense in the XZ plane of the strain ellipsoid (see Fig. 8C). Overall, the garnet porphyroblasts show a spatial association with biotite, magnetite, tourmaline, and pyrite and are typically intergrown with these minerals. The garnets are thus considered to be paragenetically related to the more dominant amphibole + biotite + magnetite ± tourmaline ± pyrite ± chalcopyrite association. Waara (2015) reported that quartz, biotite, pyrite, and magnetite make up the main inclusions in the Nautanen deformation zone garnet porphyroblasts.

K-feldspar ± biotite ± magnetite ± garnet ± hematite alteration also occurs throughout the Nautanen deformation zone domain and the eastern volcanosedimentary domain (Figs. 13, 14B). In general, it forms reddish-pink, 0.01- to 1-m-wide irregular patches and bands and is locally associated with a paragenetically later epidote ± quartz ± amphibole ± feldspar (albite) association (Fig. 14B; see below). K-feldspar ± biotite ± magnetite ± garnet alteration is locally associated with disseminations and thin (<5-mm-thick), stringer-like seams of fine- to medium-grained (<0.1–3 mm) biotite, pyrite, and chalcopyrite. Sulfide abundance is typically 1–2 vol %, but locally increases to ~5 to 8 vol %. Both pyrite and chalcopyrite replace or overprint magnetite and occur on the margins of garnet porphyroblasts and in pressure shadows, where present. Locally, dark green amphibole forms 1- to 10-mm-thick veins associated with biotite, magnetite, and scapolite.

Black tourmaline (probably schorl to dravite varieties; e.g., Frietsch et al., 1997) is a relatively common alteration mineral across the Nautanen deformation zone domain, where banded amphibole + biotite + magnetite ± garnet alteration locally grades into tourmaline-rich zones (Fig. 14D, E). These zones are typically 0.3 to 2 m wide, have irregular to banded forms, and are associated with amphibole, magnetite, quartz, and sulfide. Locally, prismatic tourmalines are oriented subparallel to the dominant NNW-oriented, steeply dipping S1/2 foliations and locally overprint sericite-muscovite grains (Fig. 14E). In both the Nautanen deformation zone and eastern volcanosedimentary domains, tourmaline also forms planar to lenticular aggregated alteration bands parallel to variably oriented S0-S1 structures in volcanic-volcaniclastic rocks, occurs within or adjacent to quartz ± amphibole ± sulfide veins (see below), and is an accessory mineral in weakly deformed to massive granitic pegmatite.

Throughout the Nautanen deformation zone (and adjacent areas), a paragenetically late epidote ± quartz alteration association overprints the previously described alteration styles and occurs as irregular patches, nodules, and veinlets (Fig. 14B, C), with epidote generally forming fine-grained (<1 mm) anhedral and granular disseminations and aggregates. Epidote-rich zones may also contain rare (~1–2 vol %) subhedral to euhedral allanite. Locally, epidote forms distinct aggregates with sharp contacts, suggesting replacement of quartzofeld-spathic wall-rock clasts (Fig. 14C). In the Nautanen deformation zone, linear epidote-quartz zones/bands and veins tend to mimic the orientation of steep, N- to NNW-aligned S1/2 foliations (Fig. 14C, F), but more irregular, patchy epidote alteration and veining that lacks a preferred orientation also occurs locally.

Metavolcanosedimentary rocks in the eastern volcanosedimentary domain and western volcanosedimentary domain are less pervasively altered compared to analogous units in the Nautanen deformation zone domain. In the western volcanosedimentary domain, K-feldspar ± magnetite ± sericite alteration of amphibole- and feldspar-bearing, intermediate volcanic-volcaniclastic rocks predominates and tends to follow earlier-formed NNE- to NE-oriented S0-S1 fabrics or deformation zones (Fig. 13). A spatial association with overprinting epidote ± quartz alteration is also evident. Amphibole + magnetite + tourmaline ± biotite alteration is less prominent in the western volcanosedimentary domain and is typically confined to more local irregular zones and veins.

Vein types and vein-related alteration

Several hydrothermal vein types occur throughout the study area, although their distributions are highly variable (Fig. 15; Table 1). Generally, quartz veins are more abundant within the eastern volcanosedimentary domain and western volcanosedimentary domain, while magnetite ± amphibole ± pyrite veinlets and breccia zones are more prevalent within the Nautanen deformation zone domain. Weakly to moderately developed alteration haloes are generally confined to within 1 to 2 cm of vein margins, although the degree of alteration halo intensity also varies. Overall, vein sets are typically aligned subparallel to earlier-formed planar structures (i.e., mainly NNE- to NNW-directed) or are locally discordant to the shear fabrics with an approximate east to west orientation. Most veins are steeply dipping (>80°) and thus mimic the generally steeply dipping character of planar structures across the study area.

In the Nautanen deformation zone domain, magnetite-amphibole veinlets form sheeted to stockwork and local breccia-like vein networks. They are spatially associated with approximately NNW-oriented biotite + amphibole + magnetite banding and, locally, they are moderately to tightly folded and sheared (Fig. 15A, B). At the contact and within S2 shear zones, amphibole grains and veins are transposed into the shear zone producing strained amphibole aggregates that define the shear zone fabric (Fig. 15A, B). At the contacts to granitic dikes, amphibole ± magnetite can occur as irregular patches containing large crystals, suggesting remobilization due to granite emplacement (Fig. 15A). Amphibole ± magnetite ± garnet veinlets and irregular to linear aggregated bands and lenses also occur along the Nautanen deformation zone and thus a continuum of amphibole- and magnetite-rich veinlets associated with pervasive to selectively pervasive iron oxide-alkali alteration banding is locally intensely developed.

Irregular to linear aggregated bands and lenses of generally fine- to coarse-grained (<1 cm) garnet associated with amphibole, biotite, magnetite, and rare chalcopyrite and pyrrhotite also occur in the Nautanen deformation zone (Fig. 15C, D). These zones have a gradational to locally sharp (veinlike) appearance and are locally discordant to the main NNW-oriented alteration banding and mylonitic foliation. In these veins, garnet is more typically subhedral to euhedral and appears less deformed, suggesting two broadly contemporaneous generations of garnet may be present within the Nautanen deformation zone domain, although this effect may represent local strain partitioning during garnet growth and amphibole + magnetite alteration. Biotite and magnetite in these zones are also typically coarse grained and subhedral and locally crosscut earlier biotite-magnetite fabrics, suggesting two generations of biotite, magnetite, and possibly amphibole. Scapolite ± amphibole ± magnetite veins occur locally in the Nautanen deformation zone, show folded to linear forms, and typically appear to predate amphibole-garnet-biotite-magnetite veins and breccias (Fig. 15E). In the eastern volcanosedimentary domain, distinctive amphibole ± quartz ± tourmaline veins with albitic haloes occur. Quartz is generally confined to the center of these veins and may have been precipitated during a later phase of vein opening.

Quartz veins in both the eastern volcanosedimentary domain and Nautanen deformation zone domain include the following: (1) planar to often wavy (folded?) quartz ± magnetite veins that locally crosscut S1/2 foliations and alteration banding, (2) generally planar quartz-tourmaline ± amphibole veins (and patchy segregations) that are locally chalcopyrite-pyrrhotite-pyrite–bearing and associated with weakly developed amphibole halos, and (3) barren, milky-white, planar quartz veins that crosscut other structures. The latter veins are about 0.02 to 1 m thick, are typically undeformed, and contain medium to coarse (~2–15 mm) anhedral and blocky quartz. Throughout the study area, epidote ± quartz ± K-feldspar veinlets associated with weak to moderate epidote ± albite halos occur (Fig. 15F). In general, they crosscut other veins and foliations, and are locally discordant to earlier-formed primary and tectonic structures.

Structural-alteration links and Cu ± Au mineralization

Drill core from the Nautanen North, Nautanen, and Aitik deposits (Fig. 2) further highlights spatial and paragenetic links between ductile-brittle deformation, hydrothermal alteration, and Cu-Au ± Ag mineralization within the Nautanen deformation zone domain (Fig. 16). Overall, the main alteration association proximal to chalcopyrite-bearing core intervals is a biotite + amphibole + magnetite + sericite ± garnet ± tourmaline ± scapolite ± quartz ± chlorite association with rare, dark green pyroxene (Table 1). Although the relative abundance and intensity of these minerals varies within and between the deposits, the paragenetic characteristics and textural style of this ore-stage iron oxide-alkali association is relatively consistent between these deposits, suggesting a degree of commonality in terms of source components, fluid composition, and/or host-rock response during fluid-rock interaction. From a structural perspective, the spatial juxtaposition, superimposition, elongated (stretched) character, and textural association of main-stage alteration and ore minerals with planar ductile-brittle structures similar to those striking north to northwest (based on surface mapping) supports a space-time genetic link between Nautanen deformation zone-related high-strain deformation, hydrothermal fluid flow, and Cu ± Au mineralization and/or a link to remobilization of preexisting 1.89 Ga Cu-Au mineralization in the case of the Aitik deposit.

Weakly to moderately developed amphibole + scapolite ± albite ± sericite veining or banding is generally evident in tectonized (schistose to gneissic) amphibole- and feldspar-bearing wall rocks where potassic-ferroan and/or sulfide mineralization is lacking (Fig. 16A, B). The former association partly defines a planar schistosity or banding (composite S1/2 fabric) and is transected by similarly oriented veins, seams, and undulose shear bands containing disseminated and stringer-type chalcopyrite, pyrite, and magnetite. Locally, the schistose to gneissic (banded) wall rocks are overprinted by biotite ± sericite ± magnetite alteration zones with disseminated and replacement stringer chalcopyrite ± pyrrhotite that, combined, define a planar tectonic fabric (S1/2). In these biotite-rich zones, chalcopyrite stringers locally form moderate to tight Z folds (F2) with axial planes parallel to the composite S1/2 fabric (Fig. 16C). At the Nautanen North deposit, disseminated pyrite, chalcopyrite, and pyrrhotite associated with biotite and magnetite alteration become more abundant in thin (~1–3 mm) veinlets and biotite ± quartz ± tourmaline seams (Fig. 16D). The latter locally mantle garnet porphyroblasts that appear somewhat stretched or elongated parallel to the main S1/2 foliation (Fig. 16D). Kinematic indicators evident on surfaces orthogonal to the main foliation (e.g., low-amplitude asymmetric bands, shear-related dilation features, assymetrically rotated grains, intrafolial shear folds, kink-bands) suggest an oblique dextral sense of shear (Fig. 16D). Elsewhere (Aitik), biotite ± tourmaline alteration is associated with irregular and partly diffuse quartz–K-feldspar–chalcopyrite veins or bands oriented parallel to S1/2 foliations in sericite + biotite-altered wall rocks (Fig. 16E).

Moderate to intense biotite + sericite alteration of feld-spathic wall rocks locally manifests as a wavy to asymmetrically banded or folded texture in the northern Nautanen deformation zone, with related synkinematic fabrics also apparent in adjacent quartz-chalcopyrite-tourmaline ± pyrrhotite veins (Fig. 16F, G). For the latter vein type, prismatic tourmaline and intergrown aggregates of subhedral chalcopyrite and pyrrhotite form subparallel inclusion trails and bands within asymmetrically deformed quartz vein interiors (Fig. 16F). Similar asymmetric shear fabrics with shear-induced folds that verge in a direction parallel to the main sense of shear are defined and accentuated by alternating biotite + sericite + garnet alteration bands. Larger garnet crystals appear to have behaved in a more rigid manner during shearing, while the preferential deformation and alignment of micaceous minerals and chalcopyrite developed (Fig. 16G). Individual wavy quartz-chalcopyrite-pyrite-magnetite-tourmaline ± K-feld-spar veins associated with biotite + tourmaline ± K-feldspar ± magnetite halos also locally display internal fabrics (e.g., mineral aggregate elongation, asymmetric seams) that suggest synkinematic emplacement and strain partitioning during hydrofracturing (Fig. 16H). Other wavy and more diffuse quartz-chalcopyrite-pyrite-tourmaline ± magnetite veins display a degree of boudinage and bifurcation indicative of synkinematic emplacement and contain lenticular and stretched S2 foliation-parallel chalcopyrite aggregates that locally extend beyond vein margins (Fig. 16I). For alteration zones where scapolite is more dominant (replacing plagioclase feldspar), synshearing fabrics are less evident, although associated magnetite, chalcopyrite, sericite, and rare pyroxene display a preferred alignment consistent with ubiquitous S2 foliations in the Nautanen deformation zone domain (Fig. 16J). Overall, however, the elongation, stretching, and/or “channeling” of anhedral chalcopyrite, pyrrhotite, and pyrite grains and aggregates into synkinematic planar structures that partly control the distribution of associated biotite + garnet + sericite + magnetite + tourmaline + quartz alteration indicate that Cu-Au mineralization along the Nautanen deformation zone is structurally controlled and developed or was remobilized during a major phase of deformation (Fig. 16K–M).

Linear bands, patches, and zones of moderate to intense, brick-red K-feldspar + hematite (± biotite ± tourmaline ± sericite) alteration also occurs at proximal settings to deposits along the Nautanen deformation zone, where it is locally associated with disseminated and fracture-filling chalcopyrite (Fig. 16N, O). Intermittent bands of this “red rock”-type alteration (Williams, 1994), representing the replacement of wall rock albite, are generally aligned parallel to dominant S2 foliations, while the intensity of internal shear fabrics is generally less well preserved compared to biotite ± sericite ± amphibole-rich alteration zones. Locally, however, segregated feldspar crystals or clasts show evidence of synkinematic rotation (as σ-clasts), while stepped or dilatant aplite vein dikes emplaced parallel to the main shear fabric also occur (Fig. 16N). Overall, the variable distribution of red rock potassic alteration within the Nautanen deformation zone and adjacent domains may reflect primary compositional variation in meta-supracrustal wall rocks (i.e., more albite-rich felsic horizons) and/or physicochemical changes during hydrothermal fluid evolution as alteration and mineralization progressed (e.g., Corriveau et al., 2016).

Like other areas along and adjacent to the Nautanen deformation zone, a paragenetically late epidote + quartz alteration association preferentially overprints or crosscuts a more pervasively developed K-feldspar + hematite alteration (Fig. 16O). In general, these epidote ± quartz veinlets and patches are randomly oriented, although veinlets may partly follow the trend of the main tectonic fabric in high-strain zones (Fig. 16N).

Geometries and kinematics within the Nautanen deformation zone, and structural interpretation

The structural character, geometry, and kinematics of the Nautanen deformation zone suggest it is composed of a set of parallel, variably spaced, NNW-SSE–oriented, moderately to steeply W-dipping, first-order reverse shear zones and associated N-S–trending, subvertical, second-order shear zones. Based on the southward plunge of L2 mineral lineations in the northern and central parts of the Nautanen deformation zone and the overall dextral kinematics preserved in the first-and second-order structures, an overall oblique, reverse sense of slip with a dextral component can be interpreted for this composite structure, suggesting local NNE-SSW–shortening directions (Fig. 17). The approximately N-S–trending, second-order shear zones are interpreted as Riedel shears that formed during the oblique, dextral-reverse, west-block-up shearing (Fig. 2). From a vertical (cross-section) perspective, the geometric and kinematic properties suggest that the Nautanen deformation zone represents a reverse duplex structure with an oblique horizontal component (Fig. 3). This deformational-kinematic assessment for the Nautanen deformation zone is consistent with southwest-side-up movements interpreted by Wanhainen et al. (2012) in Aitik and Bergman et al. (2001) from S-C fabrics in correlative volcanic rocks located 25 km south of Gällivare.

In contrast to the dextral oblique reverse shearing recorded in the northern and central parts of the Nautanen deformation zone, the southern sector around Aitik preserves L2 mineral lineations on NNE-oriented S2 shear planes that mainly plunge moderately downdip toward the west, suggesting a more E-W–directed shortening (Fig. 17). In addition, the overall N-S- to NNW-SSE–trending orientation of F2 fold axes in the study area suggests a local E-W- to ENE-WSW–directed shortening during a regional D2 deformation event. This is also in agreement with observations by Wanhainen et al. (2012) on the Aitik deposit. An approximately E-W–crustal shortening direction was suggested by Andersson et al. (2020) to have produced the latest regional ductile deformation event affecting Svecofennian supracrustal rocks to the west of the Nautanen deformation zone. This idea is also supported by the general north to north-northeast orientation of interpreted F2 folds preserved at the Malmberget IOA deposit (Bauer et al., 2018).

The local deviation of σ1-orientations along the Nautanen deformation zone is suggested to be a result of strain partitioning during regional D2 compressive deformation. To the west of the Nautanen deformation zone, two relatively large, ca. 1.88 Ga intrusive bodies occur, namely the ultramafic-mafic Dundret intrusive complex and the Aitik granite (Fig. 17; e.g., Sarlus et al., 2017, 2019). Sarlus et al. (2017, 2019) reported a penetrative, tectonic, NW-striking and moderately SW-dipping foliation developed in the Aitik granite, whereas the Dundret complex shows tectonic foliations around its margin with the center apparently unaffected by deformation. Based on this structural evidence and U-Pb geochronology results, Sarlus et al. (2019) interpreted that both intrusions were likely affected by D1 deformation and were emplaced at ca. 1.88 Ga, prior to Nautanen deformation zone-related shearing. Their emplacement ages at ca. 1.88 Ga and their foliated character suggests that the Dundret and Aitik intrusions may have acted as rigid bodies during a subsequent phase of compressive D2 deformation (e.g., Wanhainen et al., 2012). An E-W–directed shortening (D2) may have partitioned strain around their margins (Fig. 17). This mechanism might also explain the curvilinear appearance of the Nautanen deformation zone at the current exposure level, progressing from a north-northwest direction in the north to a north-south direction around Aitik and into a southwest direction as it splays in the south (Fig. 4, 17). Comparable rigid body deformation and strain partitioning have been reported for intrusion and supracrustal rocks farther south in the Skellefte district (Skyttä et al., 2012).

Relative timing and style of deformation

Crosscutting field relationships, the presence of strained and unstrained minerals of several generations, and previously published absolute age dating provide a framework for the timing of major deformation and fluid flow events in the Nautanen-Gällivare area (e.g., Fig. 5). Overall, structural and field relationships suggest that two major, independent, compressive, ductile to ductile-brittle deformation events occurred (D1 and D2), coincident with protracted and episodic late Orosirian (ca. 1.90–1.80 Ga) orogenesis. This idea is consistent with previous structural interpretations for the Malmberget-Gällivare area by Bergman et al. (2001), Wanhainen et al. (2012), and Bauer et al. (2017, 2018), and agrees with observations by Andersson et al. (2020) for the western supracrustal belt located about 20 to 50 km to the west and northwest of Gällivare.

An early penetrative tectonic foliation (S1), preserved in intermediate to felsic volcanic-volcaniclastic rocks, is defined by an alignment of strained amphibole-biotite-magnetite ± sericite, suggesting these minerals formed during D1-related deformation and M1 regional metamorphism. Depositional ages of ca. 1885 to 1878 Ma for the metasupracrustal rocks (Lynch et al., 2018; Sarlus et al., 2020) provide a lower (oldest) age limit of ca. 1.88 Ga for the commencement of the S1-forming M1-D1 event. Sarlus et al. (2017) dated the Aitik granite (1883 ± 5 Ma) and Nautanen diorite (1870 ± 12 Ma) located at the southern and northern ends of the Nautanen deformation zone, respectively (Fig. 17). Both intrusions preserve penetrative tectonic foliations, suggesting these intrusions were emplaced pre- or syn-D1 compressive deformation at ca. 1.88 to 1.87 Ga (e.g., Witschard, 1996; Wanhainen et al., 2006). This idea is further supported by igneous ages of ca. 1.89 to 1.88 Ga for deformed dioritic to granitic intrusions in the Gällivare area (Sarlus et al., 2017). An upper (youngest) age limit for the cessation of M1-D1–related deformation in the Gällivare area is provided by a ca. 1868 Ma porphyritic rhyolite located about 20 km to the south-southeast (Claeson et al., 2018b). This well-preserved unit containing euhedral feldspar phenocrysts and depositional (eutaxitic) layering (Claeson et al., 2018b) and overlain by quartzite-arkosic rocks similar to inferred stratigraphically upper-level quartzite at Gällivare (e.g., Witchard, 1996) forms part of a narrow N- to NW-trending volcanic tract extending over 25 km. This lithostratigraphic relationship, combined with other igneous age constraints (e.g., Sarlus et al., 2020), suggests M1-D1 events at Gällivare developed in the time span from ca. 1.88 to 1.87 Ga and thus represent a relatively short-lived (~10 Ma) tectonothermal episode. Strained ca. 1853 to 1848 Ma pegmatite and quartz-molybdenite-chalcopyrite veins at the Aitik deposit (Wanhainen et al., 2005; Stein, 2014) were deformed by a poorly resolved post- D1 deformation event at ca. 1.85 Ga or later (e.g., Fig. 5), but the exact timing and nature of this minor deformation event remain unclear. Although deformation of these minor units in the southern Nautanen deformation zone domain can be accounted for by D2 events at ca. 1.80 Ga (see below), shear zone-related deformation, metamorphism, and magmatism at ca. 1.86 to 1.85 Ga is recorded from the Pajala deformation belt about 100 km east of Gällivare (Fig 1; Bergman et al., 2006; Hellström and Bergman, 2016). This tentatively suggests an alternative scenario in which some degree of localized deformation at ca. 1.85 Ga may also have occurred in the Gällivare area between the main D1 and D2 deformation events (Fig. 5).

S1-designated tectonic foliations in metasupracrustal rocks to the east and west of the Nautanen deformation zone are folded (e.g., Figs. 2, 7E, 8D). This structural relationship implies that a separate, later deformation event (D2) affected D1-related structures, or that a substantial change in the direction and character of D1 deformation progressively evolved. Within the Nautanen deformation zone domain, two deformation phases are not readily distinguishable, as tectonic foliations are transposed into mainly NNW-oriented shear zones forming a composite S1/2 fabric. The 1813 ± 9 Ma dolerite dike that intrudes into the main shear zone in Aitik (Sarlus et al., 2017) contains syntectonic intrusive features (e.g., pinch-and-swell structures) that indirectly constrain the timing of D2-related deformation. Furthermore, Sarlus et al. (2017) dated syntectonic Lina-type granites at 1795 ± 4 and 1801 ± 3 Ma in the Malmberget area immediately to the west of the Nautanen deformation zone, while ca. 1.79 to 1.78 Ga metasomatic titanite ages from the Nautanen area have also been reported (Smith et al., 2009). An undeformed ca. 1728 Ma pegmatite dike at the Aitik deposit further constrains a prolonged history of dike emplacement and magmatic-hydrothermal activity along the Nautanen deformation zone and further constrains D2 deformation (Wanhainen et al., 2005). Given the reported dates for post-1.85 Ga syn- to posttectonic intrusive rocks in the Nautanen-Gällivare area, we propose that D2-related deformation and hydrothermal fluid flow developed within the time bracket 1822 to 1780 Ma. (e.g., Fig. 5).

Intratectonic andalusite porphyroblasts at Aitik (e.g., Fig. 10) indicate that peak metamorphism in the area postdates the first deformation phase (D1) but predates the second deformation phase (D2). Combined with the penetrative S1 fabric defined by amphibole-biotite-magnetite-sericite banding, the D1 event is suggested to coincide with a regional Barrovian/Buchan-type metamorphic event (M1). The character of D2-related structures, including a locally weakly developed, spaced S2 foliation and the relatively common association with brittle to ductile-brittle structures, suggests D2 deformation happened under relatively shallow crustal, brittle-plastic conditions. The abrupt change of metamorphic grade across the Nautanen deformation zone from amphibolite facies in the west to greenschist facies to the east (Bergman et al., 2001) indicates that the shear zone was reactivated after peak metamorphism and a substantial amount of displacement likely occurred after peak M1 metamorphism. Comparable late-orogenic brittle reactivation and juxtaposition of different earlier metamorphic facies has also been suggested for the Kiruna region (Andersson, 2019). Tollefsen (2014) suggested early regional metamorphism and Mn- and Ba- rich metasomatism in the Nautanen-Gällivare area was followed by a retrograde metamorphic or metasomatic event and contact metamorphism. To support this idea, Tollefsen (2014) reported pressure-temperature (PT) conditions in metamorphic mineral assemblages (based on petrographic studies, X-ray fluorescence [XRF] analysis, and THERMOCALC) ranging from 589° ± 41° to 620° ± 39°C and 4 ± 2.6 to 4.3 ± 1.5 kbar for the regional metamorphic event (M1 in this study). Contact metamorphism near granite intrusions was estimated at 668° ± 40°C and 3.3 ± 1.3 kbar in the Nautanen deformation zone and 681° ± 14°C and 2.5 ± 0.2 kbar in the eastern volcanosedimentary domain. Tollefsen (2014) interpreted retrograde metamorphism at temperatures ranging from 474° ± 43° and 510° ± 61°C for a pressure of 3 and 3.5 kbar, as an effect of hydrothermal activity along the Nautanen deformation zone. These estimates are consistent with the interpretation by Bauer et al. (2018) that D1 deformation coincided with regional metamorphism in the area, with D2 representing a more localized intrusion-related metasomatic event. Wanhainen et al. (2012) reported peak PT conditions of 500° to 600°C and 4 to 5 kbar for a metamorphic event that overprinted the Aitik deposit, and hydrothermal conditions of 200° to 500°C and ~1 kbar for the IOCG event, and relates this to a late phase of hydrothermal activity. Abundant, ca. 1.8 Ga felsic and mafic intrusions in the area (Sarlus et al., 2017) may have provided an important heat source for high-temperature, low-pressure, D2-related deformation enabling the formation of circulating hydrothermal systems within suitable structural pathways. Storey and Smith (2017) relate an IOCG mineralizing event in the Kiruna area to a phase of 1.8 to 1.7 Ga postorogenic collapse based on an Nd isotope study in titanites. In summary, structural and alteration mapping along the Nautanen deformation zone is consistent with previous studies and suggests moderate-grade regional metamorphism at 1.89 to 1.86 Ga (D1) followed by a high-temperature, magmatic-hydrothermal event that coincided with E-W–directed crustal shortening at 1.82 to 1.79 Ga (D2) at the transition to postorogenic collapse.

Structural controls on hydrothermal alteration and Cu-Au mineralization

Field crosscutting relationships, mesoscopic rock and mineral textures, structural associations, and variable strain intensity indicate that hydrothermal alteration developed episodically in the study area and that several generations of the same minerals occur in different settings. For example, amphibole-biotite-magnetite are typically highly strained and define the main S1 foliation in the area, suggesting formation during D1 deformation. In contrast, moderate K-feldspar ± scapolite ± amphibole ± biotite ± garnet ± magnetite alteration tends to exploit Nautanen deformation zone mylonitic rocks and forms replacement patches that partly crosscut the tectonic fabric. Locally, this association occurs together with largely unstrained, but locally folded, epidote-amphibole ± quartz ± feldspar veinlets. These paragenetic relationships and the strained and non-strained character of this potassic-calcic association, together with folding and syntectonic trails in garnets, suggest that this hydrothermal stage was syn- to late-tectonic during D2, consistent with the regional IOCG overprint proposed by Wanhainen et al. (2012). Waara (2015) has suggested a link between garnet growth and early-stage IOCG mineralization and shows that certain zonations within Nautanen deformation zone-hosted garnets are enriched in manganese as a result of multiple phases of hydrothermal fluids. However, the relevance of these garnets as part of the IOCG system remains unclear. Waara (2015) also infers from magnetite inclusions in garnets that magnetite might have formed early in the IOCG system. These observations are consistent with the occurrence of strained magnetite defining the S1/2 foliation and coarse-grained magnetite (± hematite) in veins and patches, suggesting a remobilization of magnetite during the D2 hydrothermal overprint. At the contacts to granitic dikes, amphibole ± magnetite can occur as irregular patches containing large crystals, suggesting remobilization due to granite emplacement (Fig. 15A). These remobilization patterns surrounding granitic sheets and dikes may be related to local contact metamorphism, as suggested by Tollefsen (2014), and may mimic the remobilization of amphibole, biotite, and magnetite during a ca. 1.80 Ga tectonothermal event (D2) at the Malmberget deposit as suggested by Bauer et al. (2018). This would further suggest that ca. 1.80 Ga intrusive rocks played an important role in element leaching and/or remobilization. Nevertheless, we cannot exclude that magnetite was also introduced by D2-related magmatic-hydrothermal activity. Hence, it remains equivocal whether this phase of magmatism provided only heat or also contributed fluids, metals and/or ligands to the developing mineral system.

While most mineralization and alteration zones in the study area display north-northwest–south-southeast orientations similar to that of the Nautanen deformation zone, the lateral distribution and geometries of hydrothermal alteration patterns and Cu-Au orebodies throughout the Nautanen deformation zone implies second-order, N- to NNE-oriented structures also played an important role during mineralization. Wanhainen et al. (2003) and Sammelin (2011) reported that both elevated Cu and Au grades at the Aitik deposit are controlled by NNE-SSW–trending, high-strain deformation zones, here interpreted as second-order structures to the Nautanen deformation zone. Similarly, Cu-Au mineralization at Nautanen North is hosted by an approximately NNW-SSE–striking shear zone (Årebäck and Dean, 2018), here interpreted as a second-order structure.

Fluid flow associated with the D2 event might have caused the (partial?) remobilization of earlier D1-related scapolite, magnetite, biotite, and amphibole (or their associated components). Nevertheless, it cannot be ruled out if all D2-related magnetite, biotite, and amphibole is related to remobilization or if new minerals were also added during syn-D2 hydrothermal activity. Critically, in the Gällivare area, D2-related fluids were preferentially channeled within the evolving Nautanen deformation zone composite shear zone, which acted as a hydraulic and permeability conduit, leading to laterally confined, deformation zone-hosted Cu-Au mineralization and related iron oxide-alkali metasomatism. Early, ca. 1.88 to 1.87 Ga hydrothermal alteration and ore formation is recorded in the Nautanen deformation zone, especially in the Aitik deposit (e.g., Wanhainen et al., 2006; 2012). This raises questions on the role of fluid-induced element leaching and remobilization. It also raises the question of how much of the copper budget of the Nautanen deformation zone formed during earlier, pre- to syn-D1 magmatic-hydrothermal events (e.g., Aitik, Liikavaara Östra) and how much was introduced (via magmatism) or redistributed during the subsequent fluid-rich, high-temperature, D2 deformation event (e.g., Nautanen). Thus, it remains unclear if the majority of metals were deposited early in the orogenic cycle (pre- to syn-D1) with a certain proportion redistributed during later D2 events, or if additional new metals were added by syn-D2 magmatic-hydrothermal processes.

Based on coupled structural and alteration features, the majority of Cu-Au ± Ag deposits in the Nautanen deformation zone domain display characteristics consistent with IOCG deposits or systems (sensu stricto) as described by Groves et al. (2010). Key diagnostic features include the following: (1) Cu + Au as economic metals; (2) hydrothermal characteristics and structural controls, commonly with breccias; (3) abundant low-Ti Fe oxide (magnetite, hematite) and/or Fe silicate gangue minerals (grunerite, Fe actinolite, fayalite); (4) light rare earth element (LREE) enrichment and low-S sulfides (chalcopyrite, bornite, chalcocite); (5) a lack of abundant quartz veins and silica-leaching of wall rocks; and (6) temporal relationship with magmatism, yet no close spatial association with causative intrusions.

Further features associated with IOCG (sensu stricto) mineralization (Groves et al., 2010) include the presence of previously deformed and hydrothermally altered country rocks that host Cu-Au and IOA mineralization, where the early alteration association is composed of scapolite, albite, amphibole, and disseminated or veinlet-hosted magnetite, and a paragenetic association between ore-stage chalcopyrite, pyrite, and pyrrhotite mineralization and variably intense potassicferroan to potassic-calcic hydrothermal alteration association in high-level systems (i.e., amphibole + biotite + sericite + garnet + K-feldspar + tourmaline association; e.g., Table 1). Notwithstanding the mineralogical and structural features of the Nautanen deformation zone and its contained mineralization that generally conform to IOCG (sensu stricto) system characteristics, differences between this system and the IOCG definition by Groves et al. (2010) include abundant, late- but syntectonic quartz-sulfide-tourmaline veining.

According to Groves et al. (2010), giant IOCG (sensu stricto) systems form in intracontinental settings at about 100 km from active craton or continental margins (e.g., Grainger et al., 2008). The relative location of the continental margin at the time of IOCG-style Cu-Au mineralization in the Nautanen deformation zone (~1.80 Ga) is unknown, but a subduction-related arc system is thought to have been active farther to the west at that time (i.e., west of the present-day Norwegian coast; e.g., Bergman et al., 2001; Angvik, 2014; Bergman and Weihed, 2020). This idea would also coincide with far-field stresses related to a distant orogeny that Groves et al. (2010) proposed is typical for IOCG systems. Sarlus et al. (2019) and Andersson (2019) interpret the Northern Norrbotten area as a continental back-arc area that underwent subsequent basin inversion during Svecokarelian-cycle orogenesis. This is comparable with the interpretation of Haywood and Skirrow (2010), suggesting that IOCG deposits can form in reactivated back-arc settings. A comparable tectonic setting might be the Mt. Isa area in Australia, where IOCG deposits are assigned to back-arc extension and subsequent basin inversion with related high-temperature, low-pressure metamorphism (Giles et al., 2006; Tiddy and Giles, 2020).

The Malmberget deposit occurs only 6 km west of the Nautanen deformation zone and represents a Kiruna-type IOA deposit (e.g., Martinsson et al., 2016). The deposit formed prior to D1 deformation, within a setting that is tentatively suggested to have undergone back-arc extension at ca. 1.90 to 1.89 Ga (Bauer et al., 2018; Sarlus et al., 2020). Subsequently, the Malmberget orebodies were metamorphosed and deformed during D1-related compressive deformation at ca. 1.87 Ga and subsequently refolded during the D2 deformation phase at ca. 1.80 Ga (Bauer et al., 2018). This suggests that IOA-type magnetite mineralization formed about 80 to 100 m.y. earlier than most IOCG deposits in the Nautanen deformation zone (excluding Aitik porphyry Cu mineralization at ca. 1.88 Ga), and that a clear tectonic-metallogenic hiatus exists between these major mineralization episodes (e.g., Fig. 5). A similar time gap between massive magnetite and Cu-Au formation was reported for the Gruvberget deposit, 40 km north of the study area (Martinsson et al., 2016). Such a temporal and genetic separation for these closely juxtaposed mineral systems is broadly consistent with IOA-IOCG metal-logenic relationships as discussed by Groves et al. (2010). The later D2-related magmatic-hydrothermal activity, however, may have facilitated remobilization of earlier-formed ore-forming components (e.g., Fe, Ca, P, K), including porphyry Cu-Au mineralization at Aitik (e.g., Wanhainen et al., 2012), thus providing a potential indirect genetic relationship between the IOA and hydrothermal Cu-Au deposits in the Gällivare area. The envisaged time gap between these mineral systems (~80–100 Ma), their formation at different stages of the Svecokarelian orogenic cycle (early for the IOA deposits, late for some of the Cu-Au deposits), and textural/structural relationships tentatively indicating some remobilization-reprecipitation of earlier-formed IOA- and/or metamorphic-related minerals within the Nautanen deformation zone (e.g., magnetite, amphibole, scapolite), suggests ca. 1.90 to 1.89 Ga synorogenic magmatism and IOA mineralization, followed by M1-D1–related structural modification and basin inversion, provided favorable precursor conditions for the subsequent IOCG-style mineralization event.

The Nautanen deformation zone in the Gällivare-Malmberget area of northern Norrbotten represents a significant Cu-Au–mineralized, crustal-scale shear zone that displays a long-lived (~80–100 m.y.) deformational history comprising several major, overprinting, reactivation episodes. The Nautanen deformation zone and its surroundings record structural evidence for two distinct deformation events that occurred during the Svecokarelian orogeny: an early, ca. 1.88 to 1.87 Ga deformation event (D1) accompanied regional peak metamorphism (M1) and the formation of an amphibole-biotite-magnetite-rich penetrative tectonic fabric (S1), and a subsequent late-Svecokarelian, ca. 1.82 to 1.79 Ga deformation event (D2) resulted in the folding and transposition of S1 foliations under relatively high-temperature, low-pressure conditions into F2 folds and composite N- to NNW-oriented S2 shear zones. The overall tectonic setting is interpreted as an inverted intra- to back-arc basin where two-stage basin inversion was caused by multiple arc accretion, resulting in early regional metamorphism and related deformation (D1) and a later high-temperature, low-pressure compressional magmatic-metasomatic event. D2 was accompanied by extensive magmatic-hydrothermal activity, generating heat and fluid that controlled the recrystallization/remobilization of amphibole, magnetite, biotite, and scapolite. The Nautanen deformation zone was reactivated during D2 deformation, which favored the development and utilization of the Nautanen deformation zone as a major fluid conduit that strongly controlled the distribution of main ore-stage iron oxide-alkali (biotite + amphibole + K-feldspar ± scapolite ± garnet) alteration and the formation of structurally focused Cu-Au ± Ag mineralization, preferentially within second-order, semibrittle faults. Since earlier-formed (pre-D2) intrusion-related Cu-Au mineralization occurs within the Nautanen deformation zone (e.g., Aitik, Liikavaara Östra), the potential that some or all of the metal budget of the Nautanen deformation zone initially formed during earlier magmatic-hydrothermal events (i.e., pre- to syn-D1 deformation) cannot be discounted. Overall, however, the mineralogical and structural characteristics of numerous Cu-Au deposits (e.g., Nautanen) located within a major hydrothermally altered shear zone (Nautanen deformation zone) is broadly consistent with IOCG-type mineralization as defined by Groves et al. (2010) and suggests that a major IOCG metallogenic epoch was active in the area at ca. 1.80 Ga. Within this context, a direct genetic link between Nautanen deformation zone-hosted IOCG mineralization and the nearby ca. 1.89 Ga Malmberget IOA deposit is not likely, although an indirect link via the remobilization of IOA-related ore-forming components (e.g., Fe, Ca, K, P) by IOCG-mineralizing systems at ca. 1.80 Ga may be established.

Dave Coller (Earth Tectonics Limited, Ireland) is acknowledged for contributions, discussions, and assistance in the field. We would like to thank Boliden Mineral AB and LKAB for the support during this project and acknowledge Peter Karlsson, Roger Nordin, Sofia Höglund, and Greg Joslin for helpful assistance, contributions, and discussions. Johan Jönberger (SGU) is thanked for reprocessing the magnetic anomaly data shown in Figure 4. Constructive comments from journal reviewer Chris R. Siron and editor Larry Meinert are most appreciated. Petroleum Experts Ltd. is thanked for donating the Move 2017 software. This work was carried out within the project, “Multi-scale 4-dimensional geological modeling of the Gällivare area,” financed by VINNOVA, Boliden, and LKAB. The project is part of the SIO-program, “Swedish Mining and Metal Producing Industry Research and Innovation Programme 2013-2016.”

Tobias E. Bauer received an M.Sc. degree in geology from Ludwig-Maximilian University of Munich, Germany, in 2008 and a Ph.D. degree in ore geology from Luleå University of Technology, Sweden, in 2013. Since that time, he has taught and conducted research at Luleå University of Technology, where he is currently Associate Professor in Ore Geology. His research has included structural controls on ore deposits and the regional structural and tectonic evolution of mineralized belts. In addition, he is active in the field of 3-D and 4-D modeling applied to mineral belts, mineral deposits, and related structures.

Gold Open Access: This paper is published under the terms of the CC-BY 3.0 license.

Supplementary data