Timing of formation of WNW–ESE-striking thrust systems
The inversion of WNW–ESE-striking thrust systems and related structures (Figure 2a and e and Figure 3d and f) and their controlling relationship over Caledonian folds and thrusts and post-Caledonian normal faults (Figure 2i and Figure 4a–b) imply a pre-Caledonian origin for WNW–ESE-structures. Their strike possibly indicates that they formed during an episode of NE–SW-oriented contraction. Here, we review possible origins for the WNW–ESE-striking thrust systems off the Faroe Islands.
Late Paleoproterozoic origin. The 240 km wide, WNW–ESE-striking Ammassalik Belt in southeastern Greenland is the southeastern continuation of the late Paleoproterozoic Nagssugtoqidian Orogen in western Greenland (Bridgewater & Myers, 1979; Chadwick & Vasudev, 1989; Chadwick et al., 1989; van Gool et al., 2002; van Gool et al., 2005). The belt consists dominantly of steeply-dipping, strike-slip (dominantly sinistral), E–W-striking shear zones, which are crosscut by gently dipping, top-SSW shear zones and nappe stacks, which formed in the Paleoproterozoic under greenschist facies conditions (Bridgewater & Myers, 1979; Chadwick & Vasudev, 1989; Chadwick et al., 1989; Kalsbeek et al., 1993). The regional strike and kinematics of structures in the Ammassalik Belt are comparable to those of the WNW–ESE-striking thrust systems off the Faroe Islands (Figure 2a–i, Figure 4a, and Figure 5a).
Smaller scale similarities include mylonitic fabrics (with sharp boundaries between intensely deformed and little deformed material) and imbricate structures, which are common both within the gently NNE-dipping shear zones of the Ammassalik Belt (Bridgewater & Myers, 1979) and within the major thrust systems west and southwest of the Faroe Islands (e.g., asymmetric folds and duplexes separated by mylonitic thrust surfaces; Figure 2a–i). In addition, fold structures within shear zones in southeastern Greenland are of comparable size (hundred to a few hundreds of meters width and height) and geometry (south- to SSW-verging asymmetric, up to isoclinal; Chadwick & Vasudev, 1989, e.g., their figures 9 and 13) to those observed within the thrust systems in the study area (Figure 2a–i). Moreover, the broad occurrences of sheared basic dykes within the shear zones at the Ammassalik Belt (Bridgewater & Myers, 1979) may very well be present in the study area too and help enhance the acoustic impedance contrast allowing the imaging of the intra-thrust shear fabrics on the interpreted seismic reflection data (e.g., amplitude contrast between the mylonitic shear zones and asymmetric folds).
In northern Scotland, Paleoproterozoic structures in the Lewisian Complex include WNW–ESE-striking, amphibolite-facies, mylonitic shear zones formed during the Laxfordian Orogeny, such as the several kilometer-wide, > tens of kilometers long Laxford (or Tarbet), Canisp (or Stoer), Gairloch, Diabaig, Ness, and Langavat shear zones (Attfield, 1987; Beach, 1974; Coward et al., 1980; Coward & Park, 1987; Evans, 1965). These are also tightly folded into WNW–ESE-elongated, dome-shaped anticlines in the same way as the WNW–ESE-striking thrust systems on the Faroe Island continental shelf, which were reworked by NE–SW-striking Caledonian folds (Figure 2a–i and Figure 4a). In addition, the double vergence of contractional structures off the Faroe Islands, i.e., (dominantly) top-SSW and (subsidiarily) top-NNE transport direction (Figure 1b–c and Figure 2a–i), is consistent with the alternating top-SSW and top-NNE vergence of folds and shear zones onshore northern Scotland (e.g., top-NNE Laxford Shear Zone and top-SSW Diabaig Shear Zone; Figure 1c; Beach, 1974; Coward et al., 1980; Coward & Park, 1987). Furthermore, some Laxfordian shear zones in northern Scotland also show top-SSW tectonic transport, e.g., Diabaig Shear Zone (Beach, 1974; Coward et al., 1980). Although most Laxfordian shear zones in northern Scotland display steep, subvertical geometries and evidence of (dominantly dextral) strike-slip movements (Coward & Park, 1987), these may reflect Caledonian reactivation/overprinting due to NW-SE-oriented Caledonian contraction and/or portions of Paleoproterozoic shear zones that were folded during the Caledonian Orogeny similarly to the thrust systems off the Faroe Islands rather than exotic terranes or inliers (e.g., Friend et al., 2008; Storey et al., 2010; Figure 5b).
The strong similarities (e.g., strike, transport direction, fold and shear zone geometries, mylonitic fabrics) of structures in the Paleoproterozoic Ammassalik Belt in southeastern Greenland and coeval Laxfordian shear zones in northern Scotland with the mapped thrust systems off the Faroe Islands suggest that they are all part of the same (Laxfordian–Nagssugtoqidian) orogen. We therefore propose that the WNW–ESE-striking thrust systems west and southwest of the Faroe Islands formed during the late Paleoproterozoic (Figure 5a).
Considering the occurrence of WNW–ESE-striking Inverian (ca. 2.49–2.48 Ga) shear zones in northern Scotland, which are crosscut by parallel Laxfordian shear zones (e.g., Attfield, 1987; Coward & Park, 1987), it is possible that the observed offshore thrust systems initiated during the Inverian Orogeny. However, more data and further work are needed to test this hypothesis. A late Paleoproterozoic age (ca. 1.8 Ga) for the presented thrust systems is therefore considered as a minimum age.
We note that the proposed relationship with late Paleoproterozoic Laxfordian–Nagssugtoqidian (and/or Inverian?) belts do not preclude a link of the mapped thrust systems with the Tornquist Zone and related fault segment, the Teisseyre–Tornquist Zone (Krzywiec et al. 2022; Mazur et al., 2015) and Sorgenfrei–Teisseyre Zone (Phillips et al., 2018; Ponikowska et al., 2024). This would either mean that the Tornquist Zone formed along inherited late Paleoproterozoic thrusts, or that it had already been formed in the late Paleoproterozoic. The latter is possible considering the WNW–ESE strike of major late Paleoproterozoic Svecokarelian–Svecofennian structures in southern Baltica (Bergh et al., 2012; Nordgulen & Saintot, 2008; Saintot et al., 2011), i.e., parallel to the Tornquist Zone (Abramovitz et al., 1999; Cotte et al., 2002; Coward, 1990; Graversen, 2009; Hossein Shomali et al., 2006; Narkiewicz et al., 2015; Pegrum, 1984; Phillips et al., 2018), and the proximity (< 100 km) of the southernmost Svecokarelian–Svecofennian structures in surface outcrops to the Sorgenfrei–Tornquist Zone in southeastern Sweden (Saintot et al., 2011). In addition, southern Baltica was also the locus of the late Paleoproterozoic Transscandinavian Igneous Belt, which also consists of WNW–ESE-striking structures in southernmost Sweden and Denmark, e.g., possibly late Paleoproterozoic WNW–ESE-striking gneissic fabrics on the island of Bornholm (Johansson et al., 2015), i.e., in the vicinity of the Sorgenfrei–Tornquist Zone.
Late Neoproterozoic origin. A possible event is the late Neoproterozoic episode of deformation recorded by amphibolite-facies, top-west movements along the east-dipping Barnhill Shear Zone in northwestern Scotland (Storey et al., 2004) and in the Walls Metamorphic Series in the Shetland Islands (Walker et al., 2020). Although the N–S strike and top-west kinematics of the Barnhill Shear Zone differs from that of the major WNW–ESE-striking thrust systems west and southwest of the Faroe Islands, they may reflect folding of the shear zone during the Caledonian Orogeny as observed for the thrust systems in the study area (Figure 2e, Figure 4a, and Figure 5b).
In eastern Europe, late Neoproterozoic deformation was recorded southwest of the WNW–ESE-striking Teisseyre–Tornquist Zone (Belka et al., 2003; Narkiewicz & Petecki, 2017; Zelazniewicz et al., 2009). In addition, the imaginary prolongation of the Sorgenfrei–Tornquist Zone to the west-northwest lines up with the Wyville–Thomson fault zone (Figure 1a–b). However, late Neoproterozoic deformation along the Teisseyre–Tornquist Zone is only preserved in exotic Gondwanan terranes in the southwest, and, thus far, the Tornquist Zone was only traced as far as the southern North Sea, where it controlled the formation and geometry of a series of WNW–ESE-striking, post-Caledonian rift basins (e.g., E–W-striking Farsund Basin; Figure 1b) and associated normal faults when it was reactivated as a major strike-slip fault (Pegrum, 1984; Phillips et al., 2018). More work is therefore needed to further examine this potential relationship.
Another fold-and-thrust belt that displays similar characteristics as those in the study area is the Timanian Orogen in northern Baltica. The Timanides are characterized by tens of km wide, tens of km thick, thousands of km long, dominantly top-SSW thrust systems, which extend from northwestern Russia to northern Norway, Svalbard and the western Barents Sea (Koehl et al., 2022; Koehl et al., 2023a; Koehl et al., 2023b; Koehl, 2024a; Koehl & Stokmo, 2024; Olovyanishnikov et al., 2000; Siedlecka & Siedlecki, 1967). However, the Timanian Orogen is located some distance (c. one thousand km) from the study area.
Thus, although a late Neoproterozoic origin is possible for the interpreted thrust systems, a formation during the late Paleoproterozoic is more probable. It is therefore probable that the Barnhill Shear Zone in northern Scotland represents (or formed along) a folded portion of an inherited late Paleoproterozoic thrust system. This suggests that late Neoproterozoic deformation reactivated and/or overprinted preexisting late Paleoproterozoic orogens and related structures.
Mid-Cenozoic origin. The positive relief at the location of major WNW–ESE- and NE–SW-striking macrofolds and thrust systems suggests recent activity along both structural trends. This is also suggested by the occurrence of multiple, high-angle, shallow (1.5–4.0 seconds TWT) reverse faults at the Wyville–Thomson and Ymir ridges (Boldreel & Andersen, 1993 their figures 4 and 5, Boldreel & Andersen, 1998 their figures 4 and 5; Johnson et al., 2005 their figure 6; Kimbell et al., 2016 their figure 2c; Stoker et al., 2015; Jolley et al., 2021; see Figure 1b–c for location). However, we did not find any convincing evidence of Cenozoic brittle reverse faulting within the Wyville–Thomson Ridge as suggested in Boldreel and Andersen (1993; 1998) and Kimbell et al. (2005; ramp-anticline complex).
The onlap of Cenozoic (Paleogene) volcanic lava flows and sedimentary rocks (e.g., Boldreel & Andersen, 1993 their figures 4 and 5, Boldreel & Andersen, 1998 their figures 4 and 5; Johnson et al., 2005 their figure 6; Kimbell et al., 2016 their figure 2c; Jolley et al., 2021) onto metamorphosed and intensely folded basement rocks at depth (present study; Figure 3a–f) may suggest some contraction-related uplift during Cenozoic times. However, it must be noted that this may simply suggest the existence of paleotopography rather than contraction and uplift, and that lava flows do not behave like sediments and thickness variations and onlap features of lava flow successions therefore do not necessarily have the same geological implications as for sedimentary successions. Notably, the impact of existing topography on the thickness of lava flows significantly differs from that on sediments (e.g., Ganci et al., 2018; Rizo, 2018; Richardson & Karlstrom, 2019). It is therefore possible that Paleogene lava flows in the study area mimic existing paleotopography onto which they were emplaced (e.g., lava flow sequence thickness variations in Jolley et al., 2021).
In addition, the existence of topographic relief at present seafloor along both WNW–ESE- and NE–SW-striking macrofolds and thrust systems and the lack of ongoing tectonic contraction suggests that most (if not all) of present-day topography was partly inherited, possibly from the mid-Paleozoic Caledonian Orogeny (present study) and/or partly related to isostatic adjustments (Smallwood, 2008). Magmatic-underplating-related uplift in the Faroe–UK region during the Cenozoic was ruled out (Smallwood, 2008). Together with the absence of pervasive contractional deformation structures within Cenozoic volcanosedimentary successions (i.e., not as pervasive as suggested by Boldreel & Andersen, 1993; Boldreel & Andersen, 1998; Johnson et al., 2005; Kimbell et al., 2016; and Stoker et al., 2015; Figure 2a–i), this indicates that Cenozoic contraction/transpression was, at most, mild if any. Thus, a Cenozoic origin for the observed WNW–ESE-striking thrust systems and related structures in metamorphosed basement rocks at depth (e.g., asymmetric folds, duplexes, mylonitic shear zones) can be ruled out (Figure 2a–i and Figure 3a–f).
Nevertheless, it is possible that some Cenozoic contractional structures are present in the study area (e.g., Ymir Ridge; Boldreel & Andersen, 1993; Boldreel & Andersen, 1998; Johnson et al., 2005; Kimbell et al., 2016), the strike of which matches that of the interpreted WNW–ESE-striking thrust systems (Figure 1b–c, Figure 2a–i, and Figure 4a), and that some of the topography at the structural highs in the study area is Cenozoic (e.g., Stoker et al., 2005; Stoker et al., 2013). Thus, the proposed late Paleoproterozoic thrust systems and their Caledonian overprints (Figure 5a–b) may have been mildly reactivated/overprinted during Cenozoic contraction/transpression.
By contrast, there is no evidence for extensional reactivation at Wyville–Thomson Ridge (Figure 2a–i). Hence, we may safely dismiss the presumed influence of an early Paleocene rifting event (Ziska & Varming, 2008) in shaping the ridge and related structures, and an origin of the ridge along an inverted normal fault (Boldreel & Andersen, 1993).
Influence on NW–SE-striking post-Caledonian transfer zones and tectonosedimentary evolution
The interpreted WNW–ESE-striking thrust systems align with post-Caledonian transfer zones on the continental shelf, e.g., the Wyville–Thomson fault zone aligns with the Sula Transfer Zone in the southeast (Bird et al., 2016) and the southeastward prolongation of the Munkagrunnur fault zone coincides with the Judd Fault (also Judd Transfer Zone; Lamers & Carmichael, 1999; Stoker et al., 2018; Sørensen, 2003; Figure 1c). The NW–SE-striking Sula transfer zone supposedly accommodates a switch of polarity of the main post-Caledonian, Devonian–Triassic normal faults, e.g., between the southeast-dipping Sula–Sgier Fault in the southwest and the northwest-dipping Shetland Spine Fault in the northeast (Bird et al., 2016; Figure 1c). Basement-seated thrust systems represent outstanding, (at least) hundreds of kilometers long, tens of kilometers wide, possibly tens of kilometers thick zones of weakness in the crust (Figure 1a–c, Figure 2a–i, and Figure 4a) and it is therefore probable that they have had a considerable impact on the formation and evolution of subsequent structures. In the present case that is on accommodating switches of polarity of post-Caledonian faults as transfer zones (Figure 5c). It is therefore proposed that the interpreted, WNW–ESE-striking, late Paleoproterozoic thrust systems continue southeast of the Faroe Islands, where they may be located at shallower crustal level and directly controlled the formation and evolution of the post-Caledonian transfer zones, as previously speculated by Kimbell et al. (2005). This is further supported by the occurrence of multiple NW–SE-striking transfer zones in northern Scotland (e.g., North Coast Transfer Zone), the Faroe Islands, and between the Faroe Islands and the Shetland Islands (Bird et al., 2016; Kimbell et al., 2005; Lamers & Carmichael, 1999; Moy & Imber, 2009; Rumph et al., 1993; Stoker et al., 2018; Sørensen, 2003; Wilson et al., 2010; Figure 1c), thus suggesting the existence of widespread WNW–ESE-striking fabrics in basement rocks in the region (Figure 5c).
An interesting feature is the rectangular basin shape created by the NNE-dipping Judd Fault and Rona Fault at the southern edge of the Faroe–Shetland Basin. The former parallels the studied potential late Paleoproterozoic thrust systems, whereas the latter parallels Caledonian structures such as the Moine Thrust (Figure 1b–c). It is therefore possible that the rectangular basin geometry is controlled by the two preexisting structural trends at depth.
Previous works have suggested various episodes of post-Caledonian uplift and erosion and rifting in the northeastern Atlantic region, including coeval contraction and extension at times (e.g., Dean et al., 1999; Lamers & Carmichael, 1999; Stoker, 2016; Stoker et al., 2017; Stoker et al., 2018). It is of course possible that the late Paleoproterozoic thrust systems described herein were mildly reactivated and/or overprinted during these events. However, this is not clear from the data analyzed. As suggested by Dean et al. (1999) and Stoker (2016), a possible explanation is that the studied WNW–ESE-striking thrust systems were reactivated as strike-slip faults, which may explain the possibly occasionally coeval timing of contraction and extension along the margin. This would however be challenging to identify if the structures were only mildly reactivated/overprinted. Nevertheless, there should probably be clues such as high-angle to sub-vertical disruption surfaces reflecting truncation of the structures by recent strike-slip faults. This is not obvious from the data analyzed. More detailed work should therefore be performed in the Faroe–Shetland Basin area, where the relative abundance of exploration wells and seismic reflection data (including 3D data) would provide a more reliable tie between tectonic movements and sedimentary deposition along specific structures. A serious limitation might be the lack of deep seismic reflection data and poor seismic imaging below the Cenozoic volcanic successions in this area, thus impeding the correlation of shallow post-Caledonian brittle faults and sedimentary strata with deep, basement-seated structures.
The lack or mild character of post-Caledonian reactivation/overprinting of the studied WNW–ESE-striking late Paleoproterozoic thrust systems is in line with observations along the potentially related Tornquist Zone. For example, Phanerozoic sedimentary strata above the Teisseyre–Tornquist Zone are generally not disturbed by post-Caledonian tectonic events, dipping gently to the southwest (Mazur et al., 2015). Nonetheless, the Tornquist Zone and shows evidence of mild Late Cretaceous–early Cenozoic contraction (e.g., Ponikowska et al., 2024), i.e., possibly similar to the study structures. Other potential episodes of mild reactivation along the Teisseyre–Tornquist Zone include Devonian–Mississippian extension, Variscan contraction in the Pennsylvanian, and Permian–Triassic extension (Krzywiec et al., 2022), the former and latter of which coincide with well-established extensional events in the study area off the Faroe Islands and northern Scotland.
