A reconstruction of Iberia accounting for Western Tethys–North Atlantic kinematics since the late-Permian–Triassic

. CE3 The western European kinematic evolution re-sults from the opening of the western Neotethys and the Atlantic oceans since the late Paleozoic and the Mesozoic. Ge-ological evidence shows that the Iberian domain recorded the propagation of these two oceanic systems well and is there-5 fore a key to signiﬁcantly advancing our understanding of the regional plate reconstructions. The late-Permian–Triassic Iberian rift basins have accommodated extension, but this tectonic stage is often neglected in most plate kinematic models, leading to the overestimation of the movements be-10 tween Iberia and Europe during the subsequent Mesozoic (Early Cretaceous) rift phase. By compiling existing seismic proﬁles and geological constraints along the North Atlantic margins, including well data over Iberia, as well as recently published kinematic and paleogeographic reconstructions, 15 we propose a coherent kinematic model of Iberia that accounts for both the Neotethyan and Atlantic evolutions. Our model shows that the Europe–Iberia plate boundary was a domain of distributed and oblique extension made of two rift systems in the Pyrenees and in the Iberian intra-continental 20 basins. It differs from standard models that consider left-lateral strike-slip movement localized only in the northern Pyrenees in introducing a signiﬁcant strike-slip movement south of the Ebro block. At a larger scale it emphasizes the role played by the late-Permian–Triassic rift and magmatism, 25 as well as strike-slip faulting in the evolution of the western Neotethys Ocean and their control on the development of the Atlantic rift.


Introduction
Global plate tectonic reconstructions are mostly based on the knowledge and reliability of magnetic anomalies that record age, rate, and direction of sea-floor spreading (Stampfli and Borel, 2002;Müller et al., 2008;Seton et al., 2012).Where these constraints are lacking or their recognition is ambiguous, kinematic reconstructions rely on the description and interpretation of the structural, sedimentary, igneous and metamorphic rocks of rifted margins and orogens (e.g., Handy et al., 2010;McQuarrie and Van Hinsbergen, 2013).However, the required quantification and distribution of finite strain into deformed continents often remain uncertain due to the poor preservation of pre-kinematic markers.A well-known example of this problem is illustrated by the contrasting Mesozoic plate kinematic models proposed for the Iberian plate relative to Europe with significant implications for the reconstructions of the Alpine Tethys and Atlantic oceans (Olivet, 1996;Handy et al., 2010;Sibuet et al., 2004;Vissers and Meijer, 2012;Barnett-Moore et al., 2016;Nirrengarten et al., 2018).This movement is proposed to be imposed by the northward propagation of the North Atlantic rifting during the Triassic-Early Cretaceous period (Olivet, 1996;Stampfli et al., 2001;Handy et al., 2010;Barnett-Moore et al., 2016;Nirrengarten et al., 2018).Although all of the proposed reconstructions agree on the amplitude and kinematics of a 400-500 km left-lateral strike-slip motion between Europe and Iberia after the Variscan orogeny, its precise timing and spatial partitioning is debated and remains unresolved so far (e.g., Vissers and Meijer, 2012; Barnett-P.Angrand et al.: The Iberian kinematics since the late- Permian-Triassic Moore et al., 2016).Because of the lack of geological constraints, such a large strike-slip displacement has been supposedly exported along the North Pyrenean Fault (Fig. 1a) (e.g., Choukroune and Mattauer, 1978;Debroas, 1987Debroas, , 1990;;Lagabrielle et al., 2010;Jammes et al., 2009).Here again, previous studies were not conclusive so there are currently no firm geological constraints nor geophysical evidence across the Pyrenees to argue for significant transcurrent deformation during the Jurassic or the Cretaceous (Olivet, 1996;Masini et al., 2014;Canérot, 2016;Chevrot et al., 2018).Other studies have suggested that this conclusion might hold true back to the Permian, based on local geological evidence from the western Pyrenees (Saspiturry et al., 2019).
An alternative scenario has recently emerged (Tugend et al., 2015;Nirrengarten et al., 2018;Tavani et al., 2018), proposing a spatiotemporal partitioning of the deformation in a wider deformation corridor than the single Pyrenean belt.It suggests that the transcurrent deformation that results from the eastwards movement of Iberia occurred mainly during the Late Jurassic-Early Cretaceous in northern Iberia along a 100 km scale pull-part or en échelon rift basins formed by the NW-SE-trending Iberian massifs.Indeed, along these massifs, several extensional basins recorded major subsidence and strike-slip deformation during the late-Permian to middle-Cretaceous time interval (Alvaro et al., 1979;Salas et al., 2001;Aldega et al., 2019;Aurell et al., 2019;Soto et al., 2019).However, no geological evidence for lithosphere-scale strike-slip movements has yet been clearly defined in the intra-Iberian basins.
These basins separate the Ebro continental block from the greater Iberia to the south.The Ebro block is delimited to the north by the Pyrenean system.Extension then migrated and localized to the north (Rat et al., 2019) leading to oceanic spreading in the Bay of Biscay and hyper-extension in the Pyrenean rift basins (Jammes et al., 2009;Lagabrielle et al., 2010;Mouthereau et al., 2014;Tugend et al., 2014Tugend et al., , 2015)).
Moreover, the contribution of pre-Late Jurassic-Early Cretaceous extension phases might have been substantial to the overall crustal attenuation and movements of Iberia (Fig. 1a) (Fernández, 2019;Soto et al., 2019).Indeed, two major geodynamic events, the late-Permian-Early Triassic breakup of Pangea and the opening of the Neotethys and the Late Triassic-Early Jurassic central Atlantic magmatic event preceding the opening of the North Atlantic Ocean are recorded in Iberia and contributed to the finite crustal thinning.Therefore, all full-fit reconstructions considering that extension between Iberia and Newfoundland only initiated by Jurassic times in the North Atlantic realm invariably overestimate the amount of strike-slip motion required in the Pyrenees and northern Iberia from the Jurassic onward (Barnett-Moore et al., 2016;Nirrengarten et al., 2018).
Here, we examine the possible contribution of the late-Permian-Triassic extension to the plate reconstruction of Iberia between the Neotethys and the North-central Atlantic domains and its impact on the definition of the spatial and temporal distribution of strike-slip movement between Iberia and Europe.By integrating constraints from 270 to 100 Ma, our reconstructions bring to light the connection between the Tethyan and the Atlantic oceanic domains.
The Permian tectonic phase is contemporaneous with widespread magmatism related to the breakup of Pangea, and its transition toward diffuse extension.This is observed in present-day rifted margins of the North Atlantic such as the North Sea and Norwegian-Danish basins (Glennie et al., 2003), the Western Approaches (McKie, 2017), the Scottish Midland Valley (Upton et al., 2004), and the basement of Cenozoic collision belts around Iberia, for instance, in the Pyrenees (Lago et al., 2004a;Denèle et al., 2012;Vacherat et al., 2017;Saspiturry et al., 2019), the Iberian Range (Lago et al., 2004b), the Catalan Coastal Ranges (Solé et al., 2002), and the Betic Cordillera (Sánchez-Navas et al., 2017).
An expression of the continued lithospheric thinning and thermal instability associated with high heat flow during the Permian (McKenzie et al., 2015) and the Triassic (Peace et al., 2019a, and references therein) is recorded in the central and southern North Atlantic.Lithospheric extension prior to (or associated with the premises of the subsequent) Early Jurassic continental breakup in the central Atlantic then favored the drainage of mantle melt reservoir (Silver et al., 2006;Peace et al., 2019a), attested by the very rapid emergence of the widespread tholeiitic magmatic CAMP (Central Atlantic Magmatic Province) event at the Triassic-Jurassic boundary (200 Ma) in the central Atlantic (Olsen, 1997;Marzoli et al., 1999;McHone, 2000).The CAMP extends to Iberia as large-scale volcanic intrusions such as the Messejana-Plasencia dyke (Cerbiá et al., 2003) in Iberia and the Late Triassic-Early Jurassic ophitic magmatism in the Pyrenees (e.g., Azambre et al., 1987).Extension and salt movements in the North Sea basins during the Late Triassic further point to the propagation of the North Atlantic rift (Goldsmith et al., 2003).
The persistence of shallow-marine to non-marine deposition during this period contrasts with the large accommodation space that is required at a larger scale to sediment the giant evaporitic province in the late Permian (Jackson et al., 2019) and in the Late Triassic (Štolfová and Shannon, 2009;Leleu et al., 2016;Ortí et al., 2017).Therefore the subsidence appears much lower than that predicted by a simple isostatic model of crustal thinning (McKenzie, 1978).
Two hypotheses may be invoked to explain the difference with the McKenzie model.(1) The first is a reduction in mantle density during lithospheric thinning, due to mantle phase transitions to lighter mineral phases because of crustal attenuation (Simon and Podladchikov, 2008) and/or due to the trapping of melt in the rising asthenosphere before breakup (Quirk and Rüpke, 2018) in addition to magmatic rethickening of attenuated crust by underplating.(2) Another possible hypothesis for the Permian-Triassic topographic evolution of the Iberian basins relies on the complex post-Variscan evolution of the Iberian lithosphere.Recent studies have shown that during the existence of the Pangea supercontinent (300 to 200 Ma), temperature in the asthenospheric mantle increased due to the thermal insulation by the continental lid (Coltice et al., 2009;Ganne et al., 2016).This thermal insulation would be responsible for the accumulation of magmatic material of the CAMP (see Peace et al., 2019a, and references therein).Such a mantle thermal anomaly could have further inhibited lithospheric mantle re-equilibration after late-Variscan mantle delamination over a long time span.This model requires a strong impermeability of the overlying lithosphere (Silver et al., 2006).As a consequence of the Pangea breakup and magmatic emission at the Triassic-Jurassic boundary, the lithospheric mantle started to cool and thicken, causing isostatic subsidence of the thinned Iberian crust and resulting in topographic drop.
This argues for a protracted period of ∼ 100 Myr (late Carboniferous to Late Triassic) of continental lithosphere thinning and magmatism prior to Early Cretaceous breakup of the North Atlantic but contemporaneous with the Tethyan evolution.One main consequence is that the late-Permian-Triassic extension has been so far underestimated in plate reconstructions, despite evidence for continuous extension.

From late-Permian-Early Triassic rifting to Late Jurassic-Early Cretaceous rifting in Iberia
The Permian-Triassic basins of Iberia are exposed in the inverted Mesozoic rift basins of the Basque-Cantabrian and Pyrenean belts, the Iberian Ranges, the Catalan Range and the Betic Cordillera (Alvaro et al., 1979;Lagabrielle et al., 2020) (Figs. 1b and 3a).The coincidence between the orientations of the Alpine orogenic segments and the spatial distribution of Permian-Triassic depocenters (Figs.1b and 3a) suggests that the Cenozoic orogenic cycle largely inherits the earliest stages of the Tethyan rift evolution.In addition, these Permian-Triassic depocenters are superposed over Variscan structures (Fig. 1b), suggesting antecedent tectonic control of the Tethyan continental rift segment by the late-Variscan evolution.
We analyze subsidence reconstructed based on a compilation of well data and a synthetic stratigraphic section in the Aquitaine Basin (Brunet, 1984), the Cameros and Iberian basins (Salas and Casas, 1993;Salas et al., 2001;Omodeo-Salé et al., 2017), western Iberia (Spooner et al., 2019), and the Betics (Hanne et al., 2003) to estimate 1D mean tectonic subsidence evolution in these areas (Fig. 3b; see Supplement for individual tectonic subsidence curves in each region).For each region, we calculated the mean tectonic subsidence, following the approach of Spooner et al. (2019) for which wells that do not sample the entire stratigraphy are corrected based on the oldest well of the region.We then calculated the mean  (Brunet, 1984), Betics (Hanne et al., 2003), Cameros Basin (Salas and Casas, 1993;Salas et al., 2001;Omodeo-Sale et al., 2017), Maestrat Basin (Salas and Casas, 1993;Salas et al., 2001), and West Iberia (Spooner et al., 2018).See Supplement for individual tectonic subsidence curves in each region.crustal stretching (β factor, Fig. 3c) for each tectonic subsidence curve based on isostatic calculation (Watts, 2001).
During the late-Permian-Early Triassic, a first phase of significant tectonic subsidence, up to 500 m, is recorded in the Maestrat Basin and on the Iberia paleomargin of the Betic basins (Salas and Casas, 1993;Van Wees et al., 1998;Salas et al., 2001;Hanne et al., 2003;Soto et al., 2019) (Fig. 3bc).This phase is contemporaneous with the westward migration of marine deposition in the Iberian basins during the Middle Triassic (Anisian-Carnian, 240-230 Ma).Sopeña et al. (1988) argue that Tethyan rifting propagated westward into the Iberian continental interior.The same evolution is suggested by the stratigraphy and the depositional evolution constraints from the Catalan and Basque-Cantabrian basins (Sopeña et al., 1988), and in the Aquitaine domain (Fig. 3b), although it is ill-defined for Permian times.
During the Late Triassic (220-200 Ma), the regional tectonic subsidence in all regions is found to be associated with the deposition of evaporites that spread all over Iberia, in the Betics, western Iberia, and in the Aquitaine Basin (Fig. 3a).The distribution of salt terranes in Iberia and its surroundings (Fig. 3a) highlights a very large subsiding domain for this period.A maximum mean subsidence of 700 m is inferred in the Maestrat Basin for Triassic times.The relatively rapid subsidence in the Triassic contrasts with the slower subsidence observed during the Early-Middle Jurassic.A notable exception is depicted by the slight increase in subsidence between 200 and 150 Ma in the Betics (Fig. 3b-c), consistent with rifting across the Iberia-Africa boundary (Ramos et al., 2016;Fernández, 2019).
A third Late Jurassic-Early Cretaceous phase (150-110 Ma) is marked by the increase in tectonic subsidence in the Iberian basins, coeval with the expected timing of strikeslip deformation and rifting in the Cameros (e.g., Rat et al., 2019;Aurell et al., 2019) and Columbrets (Etheve et al., 2018) basins as well as the initiation of mantle exhumation in the Atlantic domain (Fig. 1a) (Murillas et al., 1990;Mohn et al., 2015).The most recent extension is recorded in the Aquitaine Basin at 120-100 Ma, which reflects the onset of oceanic spreading in the Bay of Biscay (Fig. 3b-c).
Subsidence analyses show thinning events in Iberia that reveal control by Tethys and Atlantic rifting (late Permian-Late Triassic) and later by the intra-Iberian-Pyrenean rift events (Late Jurassic-Early Cretaceous).In the Iberian Basin, this latter event is characterized by a relatively large and short-lived subsidence (1.5 km in 30 Myr) localized in narrow basins that suggests the strike-slip nature of the boundary between Ebro and Iberia in the Late Jurassic.The long-lasting rift evolution, however, shows an average low stretching factor of about 1.2.

Previous kinematic models
We compile and implement previous kinematic models involving Iberia (Fig. 4).The objective is to establish a coherent kinematic model of Iberia that considers both the evolution of the Neotethyan and Atlantic regions.Global kinematic models (e.g., Müller et al., 2019) are based on the assimilation of geological and geophysical information at a large scale to allow a dynamic understanding of Earth's plate tectonics but do not aim to solve regional tectonic issues such as strain partitioning between Iberia and Europe.In contrast, regional models are focused on the reconstruction of the North Atlantic (e.g., Barnett-Moore et al., 2016;Nirrengarten et al., 2018;Peace et al., 2019b) or are interested in the reconstruction of the Alpine orogen with inferences on the kinematics of the Tethys and Adria (Schmid et al., 2008;Handy et al., 2010;Van Hinsbergen et al., 2020).
North Atlantic reconstructions use offshore geophysical constraints from the northwest Iberian margins and pay relatively little attention to the geological evolution of the Pyrenees and other orogenic domains in Iberia (e.g., Sibuet et al., 2004;Barnett-Moore et al., 2016;Nirrengarten et al., 2018).However, these models give fundamental insights into the geometry of the North Atlantic full-fit reconstruction and the timing of oceanic spreading.The nature of some magnetic anomalies in the southern North Atlantic has been the matter of considerable debate (Olivet, 1996;Sibuet et al., 2004;Vissers and Meijer, 2012;Barnett-Moore et al., 2016).Here, we adopt the reconstruction of Nirrengarten et al. (2018), who propose a model based on the re-evaluation of magnetic anomalies that are considered not oceanic before C34 (83 Ma) and therefore not suitable for kinematic studies (Nirrengarten et al., 2017).
Reconstructions of the Alpine domain (Schmid et al., 2008;Handy et al., 2010) and at a larger scale of the Tethys domain (Van Hinsbergen et al., 2020) are keys to understanding the evolution of past oceanic domains now inverted in Alpine orogenic systems (e.g., Paleotethys, Neotethys, Meliata, Pindos, and Vardar oceans).These models however do not account for the recent reconstruction of the southern North Atlantic presented above that have an impact on the movement of Iberia of interest for our study.

Reconstruction workflow
A plate reconstruction from the late Permian to middle Cretaceous is presented in Figs. 5 and 6, based on kinematic modeling using GPlates version 2.1 (Müller et al., 2018).The rotation poles of the main plates are summarized in Table 1 (see also the Supplement for GPlates files).
Our compilation is as follows: (1) the reconstruction of the western Tethys prior to the Late Jurassic is constrained by the kinematic evolution of the Mediterranean region since   Because there is no motion during the 270-250 Ma interval for the North America and Africa plates relative to Europe (Domeier and Torsvik, 2014), we extended the full-fit of these models to 270 Ma.
These input models were then updated according to the following constraints (Table 2): (1) age of rifting, mantle exhumation, onset of oceanic spreading in the Atlantic; (2) the present-day position of ophiolite bodies and the timing of rifting, oceanic spreading and subduction for the Tethyanrelated oceanic domains (Paleotethys, Neotethys, Pindos, Meliata, Vardar); (3) at 100 Ma, Iberia is close to the presentday position relative to Europe, so that the late-Mesozoic-Cenozoic Pyrenean shortening is essentially orthogonal.

Implementations of the pre-existing models
A critical step in determining the pre-rifting configuration is the restoration of rifted margins.Here, we adopted the reconstructed continental crust geometry of Nirrengarten et al. (2018).Polygons from the model of Nirrengarten et al. (2018) that are based on Seton et al. (2012) were re-defined such that they include new smaller polygons (continental micro-blocks) separated by deformed areas in Iberia and Adria to account for internal deformation (Fig. 1b).
The kinematics of these continental blocks (e.g., the Ebro block) has been reconstructed using geological constraints inferred from the tectono-sedimentary evolution of intra-and peri-Iberian basins (see Sect. 3 and Fig. 3), which allows defining periods of deformation and subsidence related to extension or transcurrent deformation.
Because a full-fit reconstruction in southwest Iberia leads to significant overlapping between the Flemish Cap and Galicia, we use the Nazaré Fault (Pereira et al., 2017) to segment western Iberia.This allows us to minimize the overlap of northwest Iberia (Galicia) over the Flemish Cap or to have a gap between southwest Iberia and Newfoundland.The Neotethys Ocean opening initiated in the early Permian in the northern Gondwana margin, resulting in the northward drift of the Cimmerian terrane and the subduction of the Paleozoic Paleotethys Ocean (Stampfli et al., 2001;Stampfli and Borel, 2002).This occurred contemporaneously with the establishment of the Carboniferous-Permian magmatic activity in the North Sea rift and Midland Valley rift areas (Evans et al., 2003;Heeremans et al., 2004;Upton et al., 2004).
As the Neotethys rift propagated westwards, diffuse continental rifting took place in the whole of western Europe defined by the position of the Paleozoic Variscan and Caledonian orogenic belts in the west, the Tornquist suture in the east and a diffuse transtensional transfer zone along the Africa-Iberia-Adria boundary (Fig. 5a).This is recorded by several late-Permian rift domains located in the southern North Atlantic (Rasmussen et al., 1998;Leleu et al., 2016), in the Adriatic (Scisciani and Esestime, 2017) in the North Sea (Hassaan et al., 2020), in the Germanic rift basins, including the Zechstein Basin (Evans, 1990;Van Wees et al., 2000;Jackson et al., 2019), and in Iberia (Figs. 2, 3, and 5a).A recent study (Sandoval et al., 2019) also showed a high pre-Early Jurassic thinning in North Atlantic basins (e.g., Galicia Interior, Porcupine, and Orphan basins).
Back-arc extension associated with the subduction of the Paleotethys (Van Hinsbergen et al., 2020) (Fig. 5b) triggered the extension and formation of oceanic basins in the Pindos and Meliata domains during the Early (250 Ma) and Late Triassic (Carnian, 220 Ma), respectively (Channell and Kozur, 1997; Stampfli et al., 2001).As proposed by Schmid et al. (2008), the Pindos Ocean was probably a western branch of the Neotethys rather than a unique ocean.The strike-slip reactivation of the Tornquist Zone could also be a far-field effect of Paleotethys closure, as suggested by Phillips et al. (2018).
During the Late Triassic-Early Jurassic (Fig. 5c-d) the opening of the Ionian Basin (Tugend et al., 2019) triggers northward displacement of Adria relative to Iberia and Africa and induces transtension between Adria and Iberia.This is consistent with Triassic basins of eastern Betics and Catalonia that developed at the emplacement of the future Alpine Tethys, the rifting of which started from the Late Triassic (220 Ma) (Stampfli and Borel, 2002;Schmid et al., 2008).The large rift-related subsidence in the Iberian basins (Fig. 3b) is kinematically consistent with the stretching lin-eations documented from Triassic strata (Soto et al., 2019).Ebro is already individualized from Iberia and moved eastwards relative to Iberia and Europe through right-lateral and left-lateral strike-slip movements, respectively.

Early Jurassic (200-160 Ma)
This period marks a gradual change from Tethyan-dominated to Atlantic-dominated tectonism in Iberia.As the Neotethys propagated in the Vardar Ocean, the Pindos and Meliata oceans started to close (Fig. 5c) (Channell and Kozur, 1997).Major dynamic changes occurred with the CAMP event (Olsen, 1997;Marzoli et al., 1999;McHone, 2000;Leleu et al., 2016;Peace et al., 2019b) that led to breakup in the central Atlantic Ocean during the 190-175 Ma interval (Pliensbachian-Toarcian) (Fig. 5c-d) according to Labails et al. (2010) and Olyphant et al. (2017).The propagation of the central Atlantic rift northwards caused extension to propagate in the southern North Atlantic (Murillas et al., 1990;Leleu et al., 2016) and laterally, eastward in the Alpine Tethys (Schmid et al., 2008;Marroni et al., 2017) by some reactivation of Triassic Neotethyan rift structures.Evidence for nearly synchronous intrusions of MORB-type gabbro, in a western branch of the Alpine Tethys, is described at 180 Ma in the internal zones of eastern Betics (Puga et al., 2011), associated with the rapid subsidence in the Betics (Fig. 3b).However, whether this is related to incipient oceanic spreading or magmatism in a hyper-extended margin is controversial.By contrast, both the thermal and stratigraphic evolutions (also Fig. 2) suggest that central Iberia remained little affected by the propagation of the Early Jurassic Atlantic rift Iberian basins (Aurell et al., 2019;Rat et al., 2019).A kinematic change from oblique to orthogonal E-W extension in the Alpine Tethys is marked by the onset of oceanic spreading between the Bajocian-Bathonian (170-166 Ma) and the Oxfordian (161 Ma) as suggested by the ages of MORB magmatism in the Alps (Schaltegger et al., 2002) and first postrift sediments (Bill et al., 2001).As such the Jurassic Alpine Tethys has temporal and genetic affinities with the Atlantic Ocean evolution rather than the Neotethys.The required differential movement between the opening of the Alpine oceanic domains, and the central Atlantic and the closure of the Neotethys and short-lived Vardar Oceans from 160 Ma onward induced the reactivation of the former diffuse transfer zone between Iberia and Africa into a localized transform plate boundary (Fig. 6a).

Late Jurassic-Early Cretaceous (160-100 Ma)
A major tectonic change occurred in the Late Jurassic-Early Cretaceous when the southernmost North Atlantic successfully rifted the continental domain located offshore of southwest Iberia in present-day coordinates (between 160 and 100 Ma; Fig. 6), as recorded by mantle exhumation and subsequent oceanic spreading at 147 and 135-133 Ma, respectively, in Gorringe Bank (Sallarès et al., 2013).Oceanic opening then migrated northward, as attested by mantle exhumation and oceanic offshore of southwest Galicia between 139.8 and 129.4 Ma (Mohn et al., 2015) and 121-112 Ma (Bronner et al., 2011;Vissers and Meijer, 2012) (Fig. 6b).At that time, the east-directed movement of Iberia relative to Ebro induced left-lateral transtensional faulting in a corridor shaped by the Iberian basins (Tugend et al., 2015;Aurell et al., 2019;Rat et al., 2019).We further infer a residual strike-slip movement between Ebro and Europe in the Pyrenean basins until the mid-Cretaceous (118 Ma) when the Bay of Biscay opened and the rotation of Iberia occurred (Sibuet et al., 2004;Barnett-Moore et al., 2016).The eastwards motion of Iberia relative to Adria resulted in the closure of the southern Alpine Tethys (Fig. 6c).Eastward rotation of Africa induces subduction along the northern Neotethyan margin (Schmid et al., 2008) (Fig. 6b-d).
Until 120 Ma (Early Cretaceous) eastward accommodation space is constantly created by the formation of rift segments in the southwest Alpine domain (Valaisan domain and southeast basins of France) and then in the Provence domains (Tavani et al., 2018).In the southern part of the Western Alps, reactivation of Tethyan normal faults are shown to be Late Jurassic-Early Cretaceous in age (Tavani et al., 2018).At 110 Ma, deformation migrates in the South Provence Basin making a straighter continuity of the Pyrenean system toward the east (Tavani et al., 2018).
6 Implications for strike-slip movements and the Europe-Iberia plate boundary 6.1 Amount of strike-slip displacement  , 86 km was accommodated.We quantify 99 and 19 km for the 140-120 and 120-100 Ma time intervals, respectively, leading to a total of 128 km of strike-slip movement during the Lower Cretaceous, in the range of amounts deduced from offshore and onshore geological observations (Olivet, 1996;Canérot, 2016).By 118 Ma, most of the strike-slip faulting is terminated as extension became orthogonal and Ebro is close to its present-day position (Jammes et al., 2009;Mouthereau et al., 2014).The maximum strain rate of 5 km Myr −1 is obtained for the 140-120 Ma time interval, revealing progressive strain localization in the Pyrenean basins before mantle exhumation (Jammes et al., 2009;Lagabrielle et al., 2010;Mouthereau et al., 2014;Tugend et al., 2014).The Iberia-Ebro boundary has played as right-lateral and left-lateral kinematics.The rapid eastward displacement of Ebro during the late-Permian to Late Jurassic period (Figs. 5  and 6) induces a total of 67 km (12, 33, 17, and 5 km during the 270-250, 250-200, 200-180, and 180-160 Ma time interval, respectively) right-lateral strike-slip between Ebro and Iberia (i.e., Galicia).This displacement has been partitioned with extension within the Iberian basins along a NWdirected intra-continental deformation corridor.This is consistent with stretching markers in Triassic rocks in this area (Soto et al., 2019).From 160 to 100 Ma, the northward propagation of the central Atlantic spreading ridge into the southern North Atlantic resulted in a net left-lateral slip of 245 km and increasing strain rates of up to 9 km Myr −1 , indicating that the southern Ebro boundary became the main tectonic Solid Earth, 11, 1313-1332, 2020 https://doi.org/10.5194/se-11-1313-2020The cumulated left-lateral displacement from both rift systems, corrected for the right-lateral displacement in the Iberian basins, is 456 km, consistent with the absolute 400-500 km required from the closure of the Atlantic between Iberia and Newfoundland.

Strike-slip structures in the intra-Iberian basins
Despite the requirement of 245 km left-lateral strike-slip displacement along the Iberia-Ebro boundary from 160 to 100 Ma, there is no simple geological evidence in support of a unique major crustal-scale fault in the Iberian Range-Basque-Cantabrian Basin system.
Several studies have suggested that a left-lateral shear zone can be recognized along the Iberian Range and the Basque-Cantabrian rifts system.Geological evidence includes the High Tagus Fault in the Iberian Range (Aldega et al., 2019;Aurell et al., 2019) and the Ventaniella Fault in the Basque-Cantabrian region (e.g., Tavani et al., 2011).The latter fault is often considered in recent reconstructions to accommodate the Iberia-Ebro movement alone (Tugend et al., 2015;Nirrengarten et al., 2018).However, the estimated left-lateral displacement along the Ventaniella Fault is only in the order of magnitude of a few kilometers (Tavani et al., 2011) and therefore cannot be used as a North Pyrenean Fault equivalent.
In the Basque-Cantabrian Basin, the Ventaniella Fault is part of a NW-SE fault system that acted as left-lateral shear zone during the Late Jurassic-Early Cretaceous and has been subsequently inverted with a right-lateral kinematic during the Cenozoic (De Vicente et al., 2011;Tavani et al., 2011;Cámara and Flinch, 2017).These faults have a Triassic origin (Tavani and Granado, 2015).Tectonic activity along these faults gets younger NE-ward (Ubierna fault: Late Jurassic-Early Cretaceous; Zamanza-Oña fault: Early-Middle Cretaceous; salt tectonics in the center of the Basin, Cámara and Flinch, 2017).We suggest the Iberia-Ebro displacement to have possibly been distributed along these structures.
The role of the weak Triassic evaporites in efficiently decoupling deformation in the pre-salt basement from the thinskinned extension in sedimentary cover has been emphasized largely in the Pyrenees (e.g., Grool et al., 2019;Duretz et al., 2019;Jourdon et al., 2020;Lagabrielle et al., 2020).Salt tectonics has also been suggested to have been particularly significant from the Jurassic through the Early Cretaceous in Mesozoic basins that shaped the NW-directed boundary between Ebro and Iberia, including the Basque-Cantabrian Basin (Cámara and Flinch, 2017), Parentis Basin (Ferrer et al., 2012), Cameros Basin (Rat et al., 2019), and Maestrat Basin (Vergés et al., 2020).The surface expression of the crustal strike-slip movements is inferred to have been limited in supra-salt layers.
The mechanism responsible for the independent movement of Ebro relative to Europe and Iberia prior to the opening of the southern North Atlantic remains unclear.The most likely hypothesis is that before the opening of the Alpine Tethys, the Ebro continental block was related to the regional eastward rotation of the Africa-Adria system.This rotation caused Ebro to move eastward, accommodating left-lateral and right-lateral strike-slip kinematics in the Pyrenean and Iberian basins, respectively.

Conclusions
To resolve several long-lasting problems of the Mesozoic kinematics of Iberia, we propose to better account for deformation associated with late-Permian-Triassic rifting and the role of the Ebro continental block in accommodating complex strain partitioning along the Iberia-Europe plate boundary and replace Iberia in a refined plate reconstruction between the Atlantic and Tethys domains.We show that (1) left-lateral strike-slip movement did occur in the Pyrenees from the late Permian to the Early Cretaceous but ended as the Bay of Biscay opened; (2) late-Permian-Triassic exhttps://doi.org/10.5194/se-11-1313-2020Solid Earth, 11, 1313-1332, 2020 tension in the Atlantic and Iberia (including Ebro) is key to quantifying the strike-slip movement in Iberia that is otherwise not well resolved from the geological constraints in Iberian basins and from full-fit reconstructions in the Jurassic.Salt tectonics that decouples syn-rift Iberian basins' evolution from their basement likely explains the lack of geological constraints.
The diffuse deformation across the Iberia-Europe plate boundary prior to oceanic spreading in the Bay of Biscay and hyper-extension in the Pyrenees appears to result mainly from transcurrent deformation partitioned with subordinate fault-perpendicular extension.The major intra-Iberia NWtrending strike-slip fault system outlined by spatially disconnected rift basins (Basque-Cantabrian, Cameros, Maestrat, and Columbrets basins) played a significant role in the Late Jurassic-Early Cretaceous, in addition to the North Pyrenean rift system.
By integrating the position of Iberia in the Tethyan and Atlantic evolution and propagating the effect of the eastward movement of Iberia into the Alpine Tethys, our reconstructions further imply that (1) Ebro was part of Adria before the onset of the Alpine Tethys opening, (2) the southern Alpine Tethys started to close in the Early Cretaceous (145 to 100 Ma), and (3) the boundary between Iberia and Africa localized as a transform plate boundary at 160 Ma, connecting the Alpine oceanic domains with the central Atlantic.
Data availability.This study is based on data compilation.Data used in this study can be found in the appropriate references.Kinematic model files are available online in the Supplement.
Author contributions.This article was mostly written by PA and FM.PA carried out the compilation of data, interpretation, and kinematic model and figures, in tight collaboration with FM.The text benefits from the expertise and contribution of EM and RA.
Competing interests.The authors declare that they have no conflict of interest.

Table 1 .
Rotation poles of the main plates or continental blocks used in the study.Poles with no references are from this study.See the complete list in the GPlates rotation file in Supplement.NW AFR: Northwest Africa; NAM: North America; EUR: Europe; WIB: West Iberia; APU: Apulia; NE AFR: Northeast Africa.

Table 2 .
Geodynamic and timing constraints used in the kinematic reconstruction model.

Table 3 .
Quantification of strike-slip displacement between the European and Ebro and between the Iberia (Galicia) and Ebro.