International Geology Review
For Peer Review Only
Tectonic evolution of Western Tethys from Jurassic to present day: coupling geological and geophysical data with seismic tomography models
Journal: International Geology Review
Manuscript ID Draft
Manuscript Type: Data Article
Date Submitted by the Author: n/a
Complete List of Authors: Hosseinpour, Maral; The University of Sydney, Geosciences Williams, Simon; The University of Sydney, Geosciences Seton, Maria; The University of Sydney, Geosciences Barnett-Moore, Nicholas; The University of Sydney, Geosciences Muller, Dietmar; The University of Sydney, Geosciences
Tectonic Reconstruction, Atlantic Kinematics, Western Tethys, Seismic Keywords: Tomography, Absolute Reference Frame
URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 1 of 73 International Geology Review
1 2 3 4 1 Tectonic evolution of Western Tethys from Jurassic to present 5 6 7 2 day: coupling geological and geophysical data with seismic 8 9 10 3 tomography models 11 12 13 14 4 For Peer Review Only 15 16 5 Maral Hosseinpour, Simon Williams, Maria Seton, Nicholas Barnett-Moore and R. 17 18 19 6 Dietmar Müller 20 21 7 22 23 24 8 EarthByte Group, School of Geosciences, University of Sydney NSW 2006, Australia 25 26 9 27 28 10 Abstract 29 30 31 32 11 The geodynamic evolution of the Western Tethys is characterised by multiple phases of 33 34 12 rifting, seafloor spreading, subduction and collisional events. Regional reconstructions are 35 36 37 13 highly dependent on the kinematic history of the major plates bounding the Atlantic and 38 39 14 Tethyan tectonic domains as well as small microplates resulted from the fragmentation of 40 41 15 northern Gondwanaland. The complexity of tectonic events in this area leads to major 42 43 44 16 discrepancies between competing models about the timing, location and polarity of 45 46 17 subduction zones, both for the Cenozoic evolution and for earlier phases. We focus on 47 48 49 18 unravelling the Mesozoic evolution of the western Tethys. We first reassessed kinematic 50 51 19 models for the Early Jurassic-Late Cretaceous opening of the Central, North Central and 52 53 20 North Atlantic and used these as boundary conditions on the kinematic reconstructions of 54 55 56 21 the Tethyan realm. We combine reconstructions of rifting and early seafloor spreading in 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 2 of 73
1 2 3 22 northern Pangea that incorporate quantitative estimates of continental extension, and 4 5 6 23 suggest a transtensional motion of Iberia relative to Europe in Early Cretaceous time to fit 7 8 24 in the refined plate configuration of Central-North Atlantic. We combined this regional 9 10 11 25 framework with a recently published model for the motion of smaller blocks within the 12 13 26 western Tethys; from this model, we created synthetic isochrons for extinct oceanic basins 14 For Peer Review Only 15 16 27 and built evolving topological plate boundaries based on the new rigid plate model to 17 18 28 derive a self- consistent and time-dependant model for the last 200 Myr. We then examined 19 20 29 the consistency of subduction history implied by the kinematic reconstructions, by 21 22 23 30 comparing reconstructed plate boundary configurations to mantle velocity structure 24 25 31 imaged by a range of seismic tomography models. Our results show that a satisfactory 26 27 28 32 match can be made between Cenozoic subduction events in western Tethys region and 29 30 33 observed shallow tomographic high velocity material. However, the match is less clear for 31 32 34 older subducted material. Correlations between surface reconstructions and deep Earth 33 34 35 35 structure suggest that mid-deep mantle seismic features under present day Northeast- 36 37 36 Central and Northwest Africa-Arabia may correspond with the Mesozoic subduction 38 39 40 37 systems in the Vardar Ocean, Alpine Tethys and Western Neotethys respectively. These 41 42 38 correlations support a model with intra-oceanic subduction of Vardar Ocean from Middle 43 44 39 Jurassic to Early Cretaceous, and mid-Early Cretaceous initiation of oceanic subduction in 45 46 47 40 the Ligurin-Piemont Ocean. The results from the plate-tomography comparison suggest the 48 49 41 existence of oceanic subduction in the Alpine Tethys Oceans in Late early Cretaceous time. 50 51 52 42 We investigated the uncertainties in the tectonic model in terms of absolute and relative 53 54 43 plate motion and surface velocities and showed that choice of absolute reference frame can 55 56 57 44 partially account for the lateral offset between the Vardar subduction zone and associated 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 3 of 73 International Geology Review
1 2 3 45 slab material in deep mantle; additional mismatches may be attributable to the limitations 4 5 6 46 of our methodology, such as the assumption that slabs sink vertically. 7 8 47 9 10 11 48 1. Introduction 12 13 49 14 For Peer Review Only 15 50 The Tethyan oceans evolved throughout the Mesozoic and into the Cenozoic within the area 16 17 18 51 bounded by Eurasia to the north and the continents comprising Gondwanaland to the west, south 19 20 52 and east (e.g. Africa, India, Antarctica and Australia). The western part of this oceanic domain, 21 22 53 the western Tethys, occupies the region between Eurasia, Africa and Arabia (Fig. 1) and is 23 24 25 54 arguably one of the most complex components of the Tethyan system, characterised by multiple 26 27 55 phases of rifting, seafloor spreading, subduction and collision. Only small fragments of oceanic 28 29 30 56 crust formed within the western Tethys are believed to be preserved in situ (Speranza et al., 31 32 57 2012), while remnants of its closure can be found in the Pyrenees, Alpine and Carpathians 33 34 58 orogenic belts in the north, the Anatolian plate in the east and the northwestern coast of Arabia 35 36 37 59 and Atlas Mountains in the south. These remnants of ancient ocean floor, arc complexes and 38 39 60 collisional systems provide the building blocks for piecing together the regional tectonic 40 41 61 evolution of the area. Constraining the broad scale tectonic evolution of the western Tethys 42 43 44 62 (including its size, shape and episodes of growth and contraction) requires an examination of the 45 46 63 large scale motions of the plates surrounding the western Tethys, such as Eurasia, Iberia and 47 48 64 Africa as well as the plates involved in the opening of the north and central Atlantic Ocean (Fig. 49 50 51 65 1). Reconstructions of the microplates that move independently from the major plates as well as 52 53 66 the basins that open and close between them must also be examined. In addition, an assessment 54 55 56 67 of how well any proposed subduction history reconciles the signatures of slab material imaged 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 4 of 73
1 2 3 68 by seismic tomography can also assist in reconciling the location of subduction zones through 4 5 6 69 time. 7 8 70 Recent tectonic models of the western Tethys have utilised advances in seismic tomography to 9 10 11 71 link the Cenozoic geological record with mantle structure (Bijwaard and Spakman, 2000; 12 13 72 Faccenna et al., 2003; Handy et al., 2010; Spakman et al., 1993; Spakman and Wortel, 2004), in 14 For Peer Review Only 15 73 particular examining the history of subduction along the Alpine and Hellenic arcs (Faccenna and 16 17 18 74 Becker, 2010; Faccenna et al., 2004; Kissling et al., 2006; van Hinsbergen et al., 2005). This 19 20 75 approach has rarely been used to study the less well constrained Mesozoic history of subduction 21 22 76 in the western Tethys, where seismic tomography can arguably play a more critical role 23 24 25 77 (Hafkenscheid et al., 2006; Van der Meer et al., 2010). This is due, in part, to the fragmented 26 27 78 geological record and uncertain subduction history at these times in the western Tethys. 28 29 79 Here, we test quantitative reconstructions of the plate boundary development within the Western 30 31 32 80 Tethys from 200 Ma to the present, with a particular focus on the pre Cenozoic phases. We 33 34 81 combine the global reconstruction of Seton et al. (2012), with minor modifications in this area by 35 36 37 82 Shephard et al. (2013), with a slightly modified recent kinematic model for the Western Tethys 38 39 83 that extends to the Jurassic (Schettino and Turco, 2006; Schettino and Turco, 2011). As part of 40 41 84 the model integration, we revised our global model for early plate motion of North America 42 43 44 85 relative to Africa (Central Atlantic), Iberia (North Central Atlantic) and Eurasia (North Atlantic) 45 46 86 to place the Western Tethys reconstructions within a regional and global framework. We create 47 48 87 alternative sets of continuously closed plate polygons, with plate boundaries imbedded, in one 49 50 51 88 million year intervals. To examine the robustness of Western Tethys reconstruction scenarios, we 52 53 89 compare reconstructed plate boundaries to several S and P wave deep mantle tomography 54 55 90 models, and using two alternative absolute reference frames, one based on a combination of 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 5 of 73 International Geology Review
1 2 3 91 hotspot trail fitting and paleomagnetic data (e.g. Torsvik et al. (2008b)) and another based on 4 5 6 92 linking surface plate motions to subducted slab remnants mapped from seismic tomography (van 7 8 93 der Meer et al, 2010). We examine how well alternative scenarios reconcile present day mantle 9 10 11 94 velocity structure, particularly for the controversial subduction histories of the Vardar and Alpine 12 13 95 Oceans (Schettino and Turco, 2011; Seton et al., 2012), and how these correlations are dependent 14 For Peer Review Only 15 96 on the tomography model and the absolute reference frame used. 16 17 18 97 (Figure 1 near here) 19 20 98 21 22 99 2. Regional tectonics 23 24 25 100 26 27 101 Reconstructions of the Western Tethys are best expressed by separating the area into two major 28 29 102 boundary domains (Atlantic and Tethyan domains) and a series of small basins, which opened 30 31 32 103 and subducted in between these two main tectonic realms (Fig. 2). The Atlantic domain includes 33 34 104 the Central, North Central, North and Northeast Atlantic, and the Tethyan domain includes the 35 36 37 105 Vardar Ocean and western Neotethys. The smaller basins or sub domains are the Pyrenean Rift 38 39 106 and Bay of Biscay, the Ligurian Ocean, Piemont Ocean, the Valais Ocean, the Ionian Sea and 40 41 107 inner Tauride Ocean (Fig. 2). 42 43 44 108 In the following section we briefly discuss the most important tectonic events in each domain 45 46 109 and its relevance and importance to our tectonic model. We highlight the main controversies in 47 48 110 each domain and identify alternative ‘end member’ scenarios, where possible. For those areas 49 50 51 111 with alternative scenarios, we compare the reconstructed subducted history with seismic 52 53 112 tomography images of the lower mantle. 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 6 of 73
1 2 3 113 Various lines of evidence available to constrain the rifting, seafloor spreading and subduction 4 5 6 114 history of the Atlantic domains and the Western Tethys basins are compiled in Table 1 and 2. 7 8 115 The following text briefly details and discusses the key points for each domain, and outline how 9 10 11 116 we chose the optimum parameters for the spreading history in the major ocean basins involved in 12 13 117 our reconstruction. 14 For Peer Review Only 15 118 (Figure 2 and Table 1 and 2 near here) 16 17 18 119 19 20 120 2.1 Atlantic Domain 21 22 121 23 24 25 122 In a wider plate kinematic context we must consider the implications of the Late Triassic to 26 27 123 Jurassic opening of the Atlantic Ocean, as the major plate motions between North America and 28 29 124 Africa/Iberia/Eurasia directly controlled the episodes of growth or contraction of the Western 30 31 32 125 Tethys domain This, in turn, dictates how we can implement the timing and direction of seafloor 33 34 126 spreading in the Ligurian and western Piemont oceans between Eurasia and Africa. For this 35 36 37 127 reason, it is necessary to consider models not only for the Western Tethys itself, but also for the 38 39 128 early Atlantic opening. In this section, we review and test different models for the early opening 40 41 129 of Central, North Central and Northeast Atlantic to converge on a scenario compatible with the 42 43 44 130 most recent observations for all the plates involved in this domain, as well as the most recent 45 46 131 reconstruction of the blocks and basins in the Western Tethys from Schettino and Turco (2011). 47 48 132 We compared the models of Kneller et al. (2012); Schettino and Turco (2009); Seton et al. 49 50 51 133 (2012); Srivastava and Roest (1999) and Vissers et al. (2013) (Fig. 3 and 4). 52 53 134 (Figure 3 and 4 near here) 54 55 135 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 7 of 73 International Geology Review
1 2 3 136 2.1.1 Central Atlantic 4 5 6 137 7 8 138 Review of models proposed for early opening of Central Atlantic 9 10 11 139 Seafloor spreading between Africa and North America initiated in the Early Jurassic, 12 13 140 significantly earlier than between North America, Iberia and Eurasia. Consequently, relative 14 For Peer Review Only 15 141 motions between these two plates are significantly faster in the Jurassic than for the other plate 16 17 18 142 pairs, resulting in the kinematics of the Central Atlantic Ocean playing a dominant role in the 19 20 143 reconstructed rate and direction of motion in the oceanic basins of the western Tethys during the 21 22 144 Jurassic. 23 24 25 145 Proposed reconstructions for the early opening of the Central Atlantic vary in the timing for the 26 27 146 beginning of rifting, break up and seafloor spreading, as well as the direction and mechanism of 28 29 147 separation. A major source of disagreement surrounds the interpret and understanding of the 30 31 32 148 origin and nature of major magnetic anomalies of the East Coast Magnetic Anomaly (ECMA) 33 34 149 and Black Spur magnetic Anomaly (BSMA) on the North American margin, and the (proposed) 35 36 37 150 African Black Spur Magnetic Anomaly (ABSMA) and West Africa Coast Magnetic Anomaly 38 39 151 (WACMA) on the African margin (Fig. 2) . Jurassic reconstructions depend heavily on whether 40 41 152 or not these anomalies are related to seafloor spreading (Schettino and Turco, 2009) or formed 42 43 44 153 on continental and/or transitional crust. The lack of clear fracture zones extending into the crust 45 46 154 older than middle Upper Jurassic time (Matthews et al., 2011) allows for significant deviations 47 48 155 between the direction of relative motion during early separation in the Central Atlantic Ocean in 49 50 51 156 different tectonic reconstructions (Fig. 3). These uncertainties affect how we reconstruct the first 52 53 157 300 km of divergence between Africa and North America, from initiation of rifting to chron 25 54 55 158 (154 Ma) (Fig. 3). After this time, there is a general agreement on Central Atlantic kinematics 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 8 of 73
1 2 3 159 due to the formation of highly reliable normal seafloor spreading magnetic anomalies and 4 5 6 160 fracture zone (Klitgord and Schouten, 1986; Müller et al., 2008). 7 8 161 In the reconstruction of Schettino and Turco (2009), seafloor spreading starts between the central 9 10 11 162 parts of ECMA and BSMA (Fig. 2 and 3). They proposed a ridge jump at 185 Ma, corresponding 12 13 163 to a change in the direction and rate of spreading (Fig. 3 and 4E F). A contrasting scenario by 14 For Peer Review Only 15 164 Labails et al. (2010) and Sahabi et al. (2004) argues that the ECMA and WACMA are conjugate 16 17 18 165 coastal anomalies. At 170 Ma significant change in rate and direction of spreading occurs, 19 20 166 changing from slow NNW SSE spreading to normal NW SE spreading (Labails et al., 2010) 21 22 167 (Fig. 4E & F and 5 ). 23 24 25 168 In the most recent study, Kneller et al. (2012) introduced a new deforming model for the 26 27 169 Mesozoic reconstruction of the Central Atlantic Ocean based on the restoration of extended 28 29 170 continental crust. In both scenarios A and B of Kneller et al. (2012) rifting starts at 240 Ma (Fig. 30 31 32 171 4E and F), however, Scenario A assigns a younger age for the initiation of E W relative motion 33 34 172 and a possible ridge jump. Scenario A leads to more evenly distributed plate motion rates for 35 36 37 173 early spreading history. 38 39 174 40 41 175 Revised model for early opening of Central Atlantic (Model 1; Table 3) 42 43 44 176 45 46 177 While alternative reconstructions for the Central Atlantic Ocean based on interpretations of 47 48 178 coastal magnetic signatures were discussed in the previous section, other criteria can be used to 49 50 51 179 discriminate between competing models. 52 53 180 A consequence of imbedding the Schettino and Turco (2009) reconstruction within a global 54 55 181 tectonic model of Seton et al. (2015) is that it results in early Late Triassic (230 Ma) rifting in the 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 9 of 73 International Geology Review
1 2 3 182 Gulf of Mexico, which contradicts the consensus age for onset of rifting in this region of Late 4 5 6 183 Triassic Early Jurassic (~200Ma) (Fig. 5D) (References within Table 1). This early Late Triassic 7 8 184 motion results in earlier break up and seafloor spreading (185 Ma) however, dating of syn rift 9 10 11 185 sedimentary and salt deposits indicates that the final break up in Gulf of Mexico happened later, 12 13 186 in the Callovian time (~166 Ma) (Fig. 5D) (References within Table 1). These observations cast 14 For Peer Review Only 15 187 doubt on the model proposed by Schettino and Turco (2009) for the early opening of the Central 16 17 18 188 Atlantic Ocean. 19 20 189 The Labails et al. (2010) model predicts an initial opening of the Central Atlantic prior to 170 21 22 190 Ma around a stage pole located within Iberia, just north of the Central Atlantic itself. Since 23 24 25 191 Eurasia is only slowly rifting from North America during this period, a consequence of this 26 27 192 model is that it predicts overall convergence between Eurasia and North Africa during this initial 28 29 193 period of Central Atlantic opening, and therefore convergence somewhere within the Western 30 31 32 194 Tethys domain (Fig. 5C 200 to 170Ma). While such motion cannot be discounted, it is difficult 33 34 195 to reconcile with current knowledge of Jurassic Western Tethys kinematics (Handy et al., 2010; 35 36 37 196 Schettino and Turco, 2011; Stampfli et al., 2002). 38 39 197 Unlike the Schettino and Turco (2009) and Labails et al. (2010) reconstructions, Kneller et al. 40 41 198 (2012) implemented continental extension and eliminates the igneous addition and exhumed 42 43 44 199 mantle along the conjugate margins. This method has been applied to other rifted margins around 45 46 200 the world and the results are thought to represent the best possible match of rifted margins 47 48 201 (Hosseinpour et al., 2013; Williams et al., 2011). Using these quantitative estimates, Kneller et 49 50 51 202 al. (2012) showed that both the Labails et al. (2010) and Schettino and Turco (2009) 52 53 203 reconstructions show significant gaps and overlaps in the central and southern parts of the 54 55 204 Central Atlantic Ocean that affect the viability of these models. 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 10 of 73
1 2 3 205 Here, we implement a model where rifting initiates at 240 Ma (Kneller et al., 2012). We use the 4 5 6 206 rotation from scenario A of Kneller et al. (2012) (Fig. 5A) in our preferred model however, to 7 8 207 avoid the complexity and early compression that changing the direction of motion before 164 Ma 9 10 11 208 imposes on the Ligurian and Piemont Oceans reconstruction, we simplified these rotations in 12 13 209 such a way as to assume a constant opening direction for Lower and Middle Jurassic times (Fig. 14 For Peer Review Only 15 210 5A). 16 17 18 211 (Figure 5 and 6 near here) 19 20 212 2.1.2 North Central Atlantic, Bay of Biscay and Pyrenees 21 22 213 23 24 25 214 Review of models proposed for early opening of North Central Atlantic, Bay of Biscay and 26 27 215 opening and closure of Pyrenees rift 28 29 216 30 31 32 217 Previous studies on the early opening of the North Central Atlantic between Newfoundland and 33 34 218 Iberia have proposed quantitative reconstructions (e.g. Srivastava and Verhoef, 1992; Olivet et 35 36 37 219 al, 1996; Sibuet et al, 2004; Vissers et al, 2012; Neres et al, 2013), however, many uncertainties 38 39 220 remain concerning the timing and the mechanism of rifting and early seafloor spreading. These 40 41 221 discrepancies affect the position of Iberia as it starts its independent rotation from Africa, 42 43 44 222 opening of the Bay of Biscay, (Fig.6) as well as the timing and amount of this rotation. These 45 46 223 uncertainties largely stem from the lack of agreement over seafloor spreading magnetic anomaly 47 48 224 constraints (Fig. 3), paleomagnetic data, and the origin and nature of transitional crust of the west 49 50 51 225 Iberian and Newfoundland margin (Bronner et al., 2011; Norton et al., 2007; Shillington et al., 52 53 226 2006). 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 11 of 73 International Geology Review
1 2 3 227 One end member scenario is heavily reliant on the interpretation of the seafloor spreading 4 5 6 228 magnetic anomalies, including anomalies as old as M20 within limited sections of the crust of 7 8 229 the Iberia and Newfoundland conjugate margins (Klitgord and Schouten, 1986; Sibuet et al., 9 10 11 230 2004; Srivastava et al., 2000; Vissers et al., 2013). These studies also interpret a margin wide M0 12 13 231 lineation to extend all the way from the Gibraltar fault zone to the Bay of Biscay and joins the 14 For Peer Review Only 15 232 loosely identified M0 anomalies in the Bay of Biscay on the Iberian margin and from the 16 17 18 233 Newfoundland fault to 46°N on the North American side (Fig. 3). Fitting these lineations 19 20 234 imposes tight constraints on the Iberia North America Eurasia configuration at M0 time and 21 22 235 places Iberia in a more southward position toward Eurasia (Sibuet et al., 2004; Vissers et al., 23 24 25 236 2013) resulting in the opening of wide basins north of Iberia (more than 500 km in eastern 26 27 237 Pyrenees; Table 1) with dominant extensional tectonics under the Pyrenees and Bay of Biscay 28 29 238 from Late Jurassic to M0 time (Sibuet et al., 2004). To close this wide oceanic basin, Sibuet et al. 30 31 32 239 (2004) suggest a double subduction system under the Pyrenees starting from 120 Ma with 33 34 240 northward subduction of the Neotethys (Valais Ocean) until 85 Ma, after which the Pyrenean rift 35 36 37 241 opens northward as a back arc basin related to Neotethys subduction. Closure of the Pyrenean 38 39 242 rift at Santonian time initiates a second northward subduction from 84 until 35 Ma. 40 41 243 An alternative end member scenario derives the North Central Atlantic Ocean kinematics mainly 42 43 44 244 from geological data from the Bay of Biscay and Pyrenees. Onshore geological data support the 45 46 245 dominance of transtensional motion along the North Pyrenees Fault (NPF) from Late Jurassic 47 48 246 time to 120 Ma (Olivet, 1996; Schettino and Turco, 2011). Offshore, extensive study of seismic 49 50 51 247 reflection and refraction profiles transecting the Iberia Newfoundland margins have been 52 53 248 interpreted to show that the crust hosting magnetic anomalies interpreted as M0 and older may 54 55 249 not be oceanic in nature, but rather exhumed mantle lithosphere resulting from hyperextension of 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 12 of 73
1 2 3 250 continental material (Dean et al., 2000; Whitmarsh and Miles, 1995). Bronner (2011) argues that 4 5 6 251 similar to the eastern part of the southwest Indian ridge (Bronner et al., 2014), the M0 anomaly 7 8 252 ((identified within the larger anomalous signal of the J anomaly) formed over exhumed 9 10 11 253 serpentinised mantle due to a magmatic pulse which triggered breakup at 112 Ma (based on 12 13 254 dating of magmatic rocks in IODP drill holes 897 and 899), followed by the diachronous 14 For Peer Review Only 15 255 initiation of seafloor spreading from south to north. This family of reconstructions rejects the 16 17 18 256 need to fit M0 and older magnetic lineations in favour of a later break up, which places Iberia in 19 20 257 a more northward position in the early Cretaceous and thus reduces the width of the Pyrenees rift 21 22 258 and the Valais Ocean (Neotethys) by proposing a strike slip/transtensional tectonic setting for 23 24 25 259 opening the of the Pyrenees rift and the Bay of Biscay (Table 1) (Handy et al., 2010; Schettino 26 27 260 and Turco, 2011; Stampfli et al., 2002). Tectono stratigraphic data from the Pyrenees prior to 28 29 261 Aptian time show the sediments formed in highly faulted and transtensionally fragmented 30 31 32 262 margins with massive mantle exhumation and crustal thinning along fault zones (Jammes et al., 33 34 263 2009), which supports this model. 35 36 37 264 The independent Early Cretaceous rotation of Iberia relative to North America due to the 38 39 265 opening of the Bay of Biscay (Schettino and Scotese, 2005; Skogseid, 2011; Vissers et al., 2013) 40 41 266 generates a compressional phase in northern and northwestern Iberia, which eventually results in 42 43 44 267 the closure of the Pyrenees rift and the Valais Ocean (westernmost extension of NeoTethys), and 45 46 268 the onset of the Pyrenees orogeny. This compressional phase, exists in both end member models 47 48 269 and lasts until Eocene time, terminating with the more recent extensional phase and back arc 49 50 51 270 basin opening in Mediterranean region. Rapid subsidence and deepening of basins together with 52 53 271 excess marine sedimentation and salt migration are considered as evidence for this extensional 54 55 272 phase (Jammes et al., 2009). The spreading direction remains mostly E W after 83 Ma in all 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 13 of 73 International Geology Review
1 2 3 273 North Central Atlantic reconstructions due to the existence of well constrained magnetic 4 5 6 274 anomalies (Gaina et al., 2002). 7 8 275 9 10 11 276 Revised model for early opening of North Central Atlantic, Bay of Biscay and opening and 12 13 277 closure of Pyrenees rift 14 For Peer Review Only 15 278 Reconstructing the North Central Atlantic based on fitting M series anomalies between M25 M3 16 17 18 279 is controversial (Srivastava and Verhoef, 1992; Vissers et al., 2013) as the pairing of these 19 20 280 anomalies to the north in Bay of Biscay leads to opening of a significant gap, interpreted as an 21 22 281 oceanic domain under the Pyrenees between Iberia and Eurasia which is inconsistent with 23 24 25 282 geological data along Iberia Eurasia plate boundaries (see previous section). Recent tomographic 26 27 283 models (Souriau et al., 2008) do not show velocity anomalies that could be attributed to a wide 28 29 284 oceanic domain subducted beneath northern Iberia (Refer to section 4.2.3 for further 30 31 32 285 explanation), Paleomagnetic data provides constraints on rotation of Iberia, but do not extend to 33 34 286 ages older than 160 Ma and may not be reliable for 130 and 120 Ma due to the deviation from 35 36 37 287 global apparent polar wander (Neres et al., 2012), limiting their usefulness to discriminate 38 39 288 between alternative reconstructions. 40 41 289 Our preferred model for the early opening of the North Central Atlantic Ocean has been 42 43 44 290 primarily constrained by Iberia Eurasia kinematics. Our model is in agreement with Olivet 45 46 291 (1996) for Iberia and in contrast to the model presented in Sibuet et al. (2004). In our model, the 47 48 292 notable relative motion between Iberia and North America starts at 200 Ma, as in Skogseid 49 50 51 293 (2011) and Vissers et al. (2013). This age is consistent with our preferred model for the north 52 53 294 reaching extensional phase that opened the Central Atlantic at 240 Ma and the North Atlantic 54 55 295 Ocean at 155 Ma (Table 1). It is also supported by reports of major unconformities in 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 14 of 73
1 2 3 296 sedimentary basins on both margins around that time [Table 1; Srivastava and Verhoef (1992)]. 4 5 6 297 This phase of rifting and stretching is ultra slow as the amount of extension between Iberia and 7 8 298 North America remains less than 50 km from 200 to 155 Ma and the North Pyrenees Fault shows 9 10 11 299 less than 30 km left lateral strike slip motion while the Gibraltar Fault accommodates most of 12 13 300 plate kinematics between Africa and Iberia/Eurasia during this time (Schettino and Turco, 2011) 14 For Peer Review Only 15 301 (Fig. 6). 16 17 18 302 At 155 Ma, there is acceleration in rifting which leads to rapid mantle exhumation (Manatschal 19 20 303 and Bernoulli, 1999; Sutra et al., 2013; Tucholke and Sibuet, 2007). The total extension between 21 22 304 Newfoundland and Iberia predicted by our model from 200 to 120 Ma is about 500 km – this 23 24 25 305 value compares well with the 340 410 km of extension estimated from structural restoration of 26 27 306 seismic profiles between Iberia and the Flemish Cap by Sutra et al. (2013), combined with 28 29 307 additional extension in the Orphan Basin between Flemish Cap and North America (Sibuet et al, 30 31 32 308 2007). The direction of motion remains mostly E W during the main phase of rifting. Our 33 34 309 reconstruction of Iberia Newfoundland yields transtensional motion along the North Pyrenees 35 36 37 310 Fault between Eurasia and Iberia with more than 300 km of strike slip motion and a maximum 38 39 311 opening of 200 km under the Pyrenees rift and ~400 km in the Valais Ocean between Eurasia 40 41 312 and Briançonnais microplate, which was attached to Iberia at that time (Fig. 6). These kinematic 42 43 44 313 constraints are consistent with geological and seismic studies suggesting the similar motion and 45 46 314 amount of opening for both basins (Choukroune and Seguret, 1973; Jammes et al., 2009; Olivet, 47 48 315 1996). Our reconstruction uses the poles of rotation of Schettino and Scotese (2005) for M0 time 49 50 51 316 (120 Ma) onward. Iberia starts rotating counterclockwise against Eurasia at this time. Our 52 53 317 reconstruction predicts ~25° of rotation which leads to opening and seafloor spreading in the Bay 54 55 318 of Biscay and closure of the Pyrenees rift and Valais Ocean to the east (Fig. 6). 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 15 of 73 International Geology Review
1 2 3 319 4 5 6 320 2.1.3 North Atlantic, Labrador Sea - Baffin Bay and Northeast Atlantic 7 8 321 9 10 11 322 The northern branches of the Atlantic Ocean opened in a three plate system of North America, 12 13 323 Greenland and Eurasia in the late Jurassic time. The rifting occurred between Eurasia 14 For Peer Review Only 15 324 (Porcupine Rockall) North America, Labrador and Baffin Island West Greenland (Labrador Sea 16 17 18 325 and Baffin Bay) and Eurasia (Norwegian margin) East Greenland (North East Atlantic) (Seton 19 20 326 et al., 2012). Early rift basins developed along the conjugate margins in Late Jurassic Early 21 22 327 Cretaceous (Table 1). Rifting continued by diachronous northward propagation of the Iberia 23 24 25 328 Newfoundland ridge followed by the onset of spreading in North Atlantic, Labrador Sea and 26 27 329 Baffin Island and ultimately in the NE Atlantic. Within all these regions, the oldest seafloor 28 29 330 spreading is younger than 83 Ma. Before this time, the motions are less well constrained, and 30 31 32 331 also likely to be much slower than the relative motions of Africa North America Iberia during 33 34 332 the Jurassic and Early Cretaceous. 35 36 37 333 Our Iberia reconstruction (Model 1; Table 3) is compatible with the younger age for the initiation 38 39 334 of rifting in the North Atlantic Ocean (Bronner et al., 2011) . Here, we modified the Seton et al. 40 41 335 (2012) Eurasia North America reconstruction for the opening of the North and Northeast 42 43 44 336 Atlantic Oceans so that it uses Srivastava and Roest (1989) poles of rotation for chron 25 (155 45 46 337 Ma) as the fit reconstruction followed by poles of rotation of Gaina et al. (2002) for the rest of 47 48 338 relative motion (Fig. 3 and 4A B). We used the fit pole of rotation (120 Ma) of Hosseinpour et 49 50 51 339 al. (2013) for opening of the Labrador Sea and Baffin Bay followed by the reconstruction of 52 53 340 Oakey and Chalmers (2012) at the beginning of seafloor spreading (60 Ma). 54 55 341 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 16 of 73
1 2 3 342 2.1.4 Atlas rift and orogeny 4 5 6 343 7 8 344 The break up of Pangaea and the initial Atlantic opening in the early Jurassic led to the formation 9 10 11 345 of the Atlas rift and horizontal motion between Morocco and Northwest Africa (Fig. 2). This 12 13 346 continental rift remained inactive up to Oligocene time when convergence between Africa and 14 For Peer Review Only 15 347 Eurasia inverted the extensional tectonic rift into the anticline structures of the High Atlas 16 17 18 348 Mountains (Beauchamp et al., 1999; Beauchamp et al., 1996; De Lamotte et al., 2009). 19 20 349 In spite of uncertainties about the extent of the rift system, there is a general agreement, based on 21 22 350 balancing of geological cross sections, seismic reflection and well data, that the basin was small 23 24 25 351 and the shortening was minor (Table 1). This could imply that although Morocco and Northwest 26 27 352 Africa can be treated as two separate plates from Jurassic time, the relative motion between them 28 29 353 is trivial and can be ignored in global tectonic reconstructions. The quantitative global 30 31 32 354 reconstruction that are compared in this study (Models 1 and 2, see section 3 and Table 3) all 33 34 355 follow this approach. Nevertheless, the recent study of Schettino and Turco (2009) claims to 35 36 37 356 determine a significant (170 km) dextral offset between Morocco and Northwest Africa during 38 39 357 Atlantic rifting, which remains rather speculative in the absence of reliable geological data to 40 41 358 support it. This also further questions the early kinematics of the Central Atlantic Ocean 42 43 44 359 implemented in their model (Labails and Roest, 2010). 45 46 360 47 48 361 49 50 51 362 2.2 Tethyan Domain 52 53 363 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 17 of 73 International Geology Review
1 2 3 364 The kinematics of the Atlantic domain are important for framing the tectonic events in the 4 5 6 365 western margins of the Tethys (in particular the Pyrenees, Western Alps and Apennines). 7 8 366 However, developing a more complete model for the whole western Tethys tectonic evolution 9 10 11 367 relies on many additional lines of geological evidence. The closure of the Western Neotethys and 12 13 368 the back arc basins of the Paleotethys delimits the eastern parts of the western Tethys region and 14 For Peer Review Only 15 369 influences the formation of the Eastern Alps and Carpathians, Dinarides, Balkanides, Hellenides 16 17 18 370 and Taurides (Stampfli et al., 2002) (Fig.2). However, the remnants of the closure of these back 19 20 371 arc basins have been overprinted by the recent Alpine orogeny, obscuring much of the older part 21 22 372 of the tectonic history of this region. There are still conflicting ideas about whether the 23 24 25 373 westernmost branch of the Neotethys opened as a single ocean (Schettino and Turco, 2011; 26 27 374 Schmid et al., 2008), or was divided into northern and southern branches (Stampfli et al., 2002). 28 29 375 Furthermore, the subduction history of the Western Neotethys and Vardar Ocean is still poorly 30 31 32 376 known. Alternative reconstructions for the Vardar Ocean argue that the subduction either starts 33 34 377 and remains on the southern Eurasian margin (Seton et al., 2012), or develops as an intra oceanic 35 36 37 378 arc (Schettino and Turco, 2011). The timing of obduction of this oceanic arc on the African 38 39 379 (Adriatic) margin and the relocation of Vardar subduction to the Eurasian margin are also still 40 41 380 not well understood. Further to the south, closure of the Western Neotethys raises the same 42 43 44 381 controversies, including whether subduction remains on the Eurasian (Iran) margin (Norton, 45 46 382 1999; Seton et al., 2012) or retreats westward either as a consequence of opening of the Semail 47 48 383 back arc basin on the Eurasian margin (Stampfli et al., 2002; Stampfli and Borel, 2004) or a 49 50 51 384 short lived intra oceanic subduction zone (De Lamotte et al., 2009; Dercourt et al., 1993). 52 53 385 In the following sections, we briefly discuss the evolution of Tethyan oceanic basins and the 54 55 386 implications of these events on the tectonic history of Western Tethys. 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 18 of 73
1 2 3 387 4 5 6 388 2.2.1 Northern branch of Neotethys (Vardar Ocean) 7 8 389 9 10 11 390 Subduction of the PaleoTethys was associated with the opening of several generations of back 12 13 391 arc basins along the Eurasian margins between 240 170 Ma. It has been suggested that the 14 For Peer Review Only 15 392 Eastern Vardar Ocean opened along the Eurasian margin as a result of slab rollback of SE 16 17 18 393 dipping intra oceanic subduction (Schmid et al., 2008; Stampfli et al., 2002; Stampfli and Borel, 19 20 394 2004). Stampfli et al. (2002) and Stampfli and Borel (2004) as well as Channell and Kozur 21 22 395 (1997) treat the Pindos Basin as a southern extension of the Vardar Ocean (Model 3; Fig. 2). 23 24 25 396 Such a separate southern branch has not, however, been recognised by Schmid et al. (2008) and 26 27 397 Schettino and Turco (2011) (Models 1a, 1b and 2). 28 29 398 Different tectonic scenarios have been suggested for closure of the Vardar Ocean. One 30 31 32 399 possibility is that N dipping Western Neotethys subduction extends northward to the Vardar 33 34 400 Ocean in the Early Middle Jurassic and remains on the Eurasian margin for the entire period of 35 36 37 401 subduction (Norton, 1999; Seton et al., 2012). However, this reconstruction fails to satisfactorily 38 39 402 explain the complexity of the subduction history suggested by available geological evidence, 40 41 403 notably the Vardar ophiolites on the African (Adriatic) margin (Bortolotti et al., 2008; Schmid et 42 43 44 404 al., 2008; Sharp and Robertson, 2006). In other reconstructions, Vardar subduction starts in the 45 46 405 Early Middle Jurassic as an arcuate NE to SE dipping intra oceanic arc and retreats westward 47 48 406 (Schettino and Turco, 2011; Stampfli et al., 2002; Stampfli and Borel, 2004; van Hinsbergen et 49 50 51 407 al., 2005) (Models 1a, 1b and 3; Figs.7B and C). Collision of the Vardar arc with the Adriatic 52 53 408 margin and the consequent obduction of Vardar oceanic remnants onto the Panonian Carpathian 54 55 409 blocks occurs in Late Jurassic early Cretaceous time (Table 2) and is recorded along the 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 19 of 73 International Geology Review
1 2 3 410 Carpathian, Dinaride, Hellenide and Tauride platforms as the Vardar ophiolite belts [Schettino 4 5 6 411 and Turco (2011); Schmid et al. (2008); Schmid et al. (2004) and references therein]. 7 8 412 Uncertainty remains over the Vardar subduction history after the complete obduction of the 9 10 11 413 oceanic arc onto the Adriatic margin. Schettino and Turco (2011), following Sengör and Hsü 12 13 414 (1984), propose a northward jump of subduction to southern Eurasian margin after 120 Ma (Fig. 14 For Peer Review Only 15 415 7D) and describe the Eurasian margin as passive from the Late Jurassic to Early Cretaceous 16 17 18 416 (Models 1a and b). Alternatively, engör (1990) suggests that northward migration of Semail 19 20 417 subduction between Iran and Arabia under the Sanandaj Sirjan zone initiated N dipping 21 22 418 subduction under the Rhodope block in the Early Cretaceous (Model 3). 23 24 25 419 The most recent back arc basins of the Neotethys (e.g. Lycian; Fig. 2) have been suggested to 26 27 420 open and subduct within remnants of the Vardar Ocean in the Late Cretaceous (Schmid et al., 28 29 421 2008; Stampfli et al., 2002). Stampfli et al. (2002) argues that subduction reached the so called 30 31 32 422 Pindos Ocean during this time, creating the Pindos ophiolites (Model 3; Fig. 2). From 84 to 69 33 34 423 Ma, the Vardar Ocean closes and the Menderes Taurides block collides with Eurasia (Schettino 35 36 37 424 and Turco, 2011; Stampfli et al., 2002). 38 39 425 40 41 426 2.2.2 Southern Neotethys: Ionian Sea and Intra-Tauride Ocean (Eastern Mediterranean) 42 43 44 427 45 46 428 Reconstructions, that support an older opening time for these basins, consider them as a southern 47 48 429 branch of the Neotethys, opened between mainland Gondwana and smaller blocks that rifted 49 50 51 430 away from it (Stampfli and Borel, 2004). The timing of active spreading has been variously 52 53 431 interpreted between Jurassic to Paleogene (Table 1). 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 20 of 73
1 2 3 432 Alternative reconstructions for the tectonic development of the Eastern Mediterranean (Table 1) 4 5 6 433 suggest that the older crust of the Ionian Sea links Atlantic kinematics to Neotethys via a 7 8 434 continental rift between the Taurus plate and Africa. This rift develops into seafloor spreading 9 10 11 435 about 100 Myr after rifting ceases (Model 1a and 1b; Fig. 7E). Other studies reject the existence 12 13 436 of this Southern Neotethys branch, and suggest a tight fit between Adria and Africa until Late 14 For Peer Review Only 15 437 Cretaceous time (Schmid et al., 2008). Interpretations for the time of initiation of subduction and 16 17 18 438 formation of the Hellenic Arc vary between Late Cretaceous and early Neogene (Table 2). There 19 20 439 is geological and geophysical evidence to support the theory of the Vardar slab migrating 21 22 440 southward and initiating subduction in the East Mediterranean Sea after collision between 23 24 25 441 Rhodope and Menderes Taurides block (Faccenna et al., 2003; Jolivet et al., 2003) (Models 2 26 27 442 and 3) Alternatively, it has been argued that a compressional phase following the closure of 28 29 443 Vardar Ocean initiates new subduction in the Mediterranean in Early Palaeocene time (Schettino 30 31 32 444 and Turco, 2011) (Models 1a and 1b). 33 34 445 35 36 37 446 2.2.3 Piemont, Ligurian and Valais Oceans 38 39 447 40 41 448 The Piemont Ligurian and Valais Oceans (Alpine Tethys) are the western extension of the 42 43 44 449 Neotethys, which opened as a result of fragmentation of northern Gondwana and the separation 45 46 450 of Eurasia from the African plate. Due to the total consumption of any oceanic seafloor formed 47 48 451 within these basins and the consequent lack of any preserved magnetic anomalies the timing of 49 50 51 452 rifting, seafloor spreading, and the direction and amount of extension and spreading, in these 52 53 453 oceanic bodies is unknown and instead derived from Central and North Atlantic kinematics (e.g. 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 21 of 73 International Geology Review
1 2 3 454 Vissers et al., 2013). Therefore, the kinematics of the Alpine Tethys is dependent on the 4 5 6 455 alternative interpretations of early Atlantic kinematics. 7 8 456 Closure of these oceanic basins and consequent collisional events, led to the formation of the 9 10 11 457 major orogenic belts and massive nappe stacking now exposed in the Alps, Dinarides, Pennines 12 13 458 and Apennines (Handy et al., 2010; Schmid et al., 2008; Stampfli and Borel, 2004). 14 For Peer Review Only 15 459 The earliest indications of extensional tectonics in the Alpine Tethys are documented from mafic 16 17 18 460 intrusions formed in Permo Triassic time (Stampfli et al., 2002) followed by rifting and thermal 19 20 461 subsidence of continental crust and seafloor spreading within the Liguro Piemont Ocean (Models 21 22 462 1a, 1b, 2 and 3; Figs 7A to C). The Valais rift opened between the Briançonnais microplate and 23 24 25 463 Eurasia when Iberia and Newfoundland began to detach from each other in Late Jurassic Early 26 27 464 Cretaceous time (Handy et al., 2010; Stampfli et al., 2002) (Models 1a, 1b, 2, 3 and 4; Figs 7A to 28 29 465 C). Table 1 summarises different times and widths suggested for this small oceanic basin. As 30 31 32 466 outlined in previous sections, Iberian kinematics and the opening of the North Atlantic play a 33 34 467 crucial role in reconstructing the plate motions in the Bay of Biscay, the Pyrenees and the Valais 35 36 37 468 Ocean. The maximum width, timing and direction of the opening of the Valais ocean, is highly 38 39 469 influenced by full fit positions of Iberia, the time of Iberia Newfoundland rift initiation, and 40 41 470 whether or not real seafloor spreading started at 120 Ma (chron M0). 42 43 44 471 45 46 472 Subduction and collision in the Alps 47 48 473 As the North Atlantic opened, the active plate boundaries between Eurasia and northwest Africa 49 50 51 474 migrates northwards from the Gibraltar fault and Ligurian spreading ridge to the North Pyrenean 52 53 475 fault and Valais spreading centre (Figs. 7C to E). Spreading ceases in the Ligurian Ocean around 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 22 of 73
1 2 3 476 130 Ma (Handy et al., 2010; Schettino and Turco, 2011; Stampfli and Borel, 2004) (Models 1a, 4 5 6 477 1b, 2 and 3). 7 8 478 Alternative scenarios have been proposed for the initiation of subduction in the Ligurian Ocean. 9 10 11 479 In both the Schettino and Turco (2011) and Handy et al. (2010) models (Models 1a and 1b), the 12 13 480 Ligurian Ocean began subducting under the eastern margin of Iberia from 130 Ma as a result of 14 For Peer Review Only 15 481 eastward escape of Iberia during the opening of the North Atlantic, although the kinematics, 16 17 18 482 location and polarity of this subduction are different in these two models (Table 2). Stampfli et 19 20 483 al. (2002) (Model 3) on the other hand, modelled a 30 Myrs delay in the propagation of 21 22 484 subduction to the Western Alps. 23 24 25 485 The fact that Atlantic and Tethyan kinematics are controlling the motion in the Ligurian Piemont 26 27 486 Ocean results in major differences in the timing of termination of spreading and the mechanism 28 29 487 and timing for the initiation of Alpine subduction. Schettino and Turco (2011) (Models 1a and 30 31 32 488 1b) reconstruct the initiation of Piemont Ocean subduction as a result of the opening of the North 33 34 489 Atlantic at 120 Ma when Eurasia started to move relative to North America and thus rotate 35 36 37 490 southward, leading to a cessation of spreading and initiation of Alpine subduction (Fig 7D). The 38 39 491 closure of Vardar Ocean does not play a major role in initiation of subduction of the Piemont 40 41 492 Ocean in this scenario because Vardar subduction terminates on Adriatic margin and jumps to 42 43 44 493 Eurasian margin prior to 120 Ma (Fig 7D). In contrast, Stampfli et al. (2002) (Model 3) and 45 46 494 Handy et al. (2010) believe that the beginning of subduction in the Piemont Ocean was triggered 47 48 495 or highly influenced by Neotethyan slab pull in the Vardar Ocean. In this scenario, the 49 50 51 496 accretionary wedge of the Vardar Ocean passes through the continental units of the Carpatians 52 53 497 and Austro Alpine and eventually initiates subduction in the Piemont Ocean. This Cretaceous 54 55 498 intra continental subduction in the Eastern Alps has been referred to as the Eo Alpine orogeny 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 23 of 73 International Geology Review
1 2 3 499 (Eastern Alpine subduction).. Massive nappe stacking, high pressure metamorphism and orogeny 4 5 6 500 are the geological expressions of this subduction (Faryad and Hoinkes, 2003; Handy et al., 2010; 7 8 501 Seidel et al., 1981). The metamorphic core of this event decreases in age northward in the 9 10 11 502 western Alps thus emphasizing the southward directed subduction (Models 1a, 1b and 3). The 12 13 503 Eo Alpine compressional phase continues up to Upper Late Cretaceous time when oceanic crust 14 For Peer Review Only 15 504 of the Piemont Ocean is involved in the subduction process (Handy et al., 2010; Stampfli et al., 16 17 18 505 2002; Stampfli et al., 1998). 19 20 506 Subduction in the Valais Ocean initiates as a result of the rotation of Iberia and opening of the 21 22 507 Bay of Biscay. The exact timing and polarity of subduction is debated, but it is generally agreed 23 24 25 508 that the main acceleration of subduction occurred in Eocene time (Figs 7G and H). Collision in 26 27 509 the Alps began when Eurasian continental crust started subducting and thrusting in the Western 28 29 510 Alps (Figs 7H and I). It has been suggested that the break off of the Alpine slab in shallow 30 31 32 511 mantle occurs around this time (45 30 Ma, controversial along the Alps) due to significant 33 34 512 decrease in convergence rate (Bijwaard et al., 1998; Blanckenburg and Davies, 1995; Handy et 35 36 37 513 al., 2010; Shephard et al., 2014; Sinclair, 1997). 38 39 514 Whether the polarity of Ligurian subduction remains the same [Penninic prism, NW (Model 1a, 40 41 515 1b and 2) or SE dipping] or flips in Oligocene time (Apennenic prism; Model 3) is still under 42 43 44 516 debate. Evidence for back thrusting in the north and south Apennines has been invoked to 45 46 517 support this change in polarity (Handy et al., 2010), although it has been noted that the 47 48 518 continuous nature of sedimentation in the Appennenic belt and the opening of Liguro Provençal 49 50 51 519 back arc basin support a NW dipping subduction zone (Schettino and Turco (2011) and 52 53 520 references therein). Subsequent slab roll back causes massive extension and rifting of the 54 55 521 Liguro Provençal back arc basin, leading to the burial of the Valais Ocean suture zone (Stampfli 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 24 of 73
1 2 3 522 et al., 2002). This convergent boundary continued to the Ionian Ocean and opens the most recent 4 5 6 523 Tyrrhenian back arc basin (Figs 7H and I). 7 8 524 A limited amount of deformation has been reported from the south Iberic margin, in the form of 9 10 11 525 local inversion and strike slip motions (Schettino and Turco, 2011; Stampfli et al., 2002), but the 12 13 526 south Ligurian Ocean remains mostly passive from Cretaceous until Eocene/Oligocene time 14 For Peer Review Only 15 527 (Figs 7D to G). At this time subduction initiates on the south Iberic margin, followed by opening 16 17 18 528 of the Liguro provençal and Alboran back arc basins (Models 1a, 1b, 2 and 3; Figs 7H and I). 19 20 529 This subduction is described as a southward continuation of the Apennenic slab under the Betic 21 22 530 area (Spakman and Wortel, 2004). 23 24 25 531 (Figure 7 near here) 26 27 532 28 29 533 3. Models for the Western Tethys within a global kinematic framework 30 31 32 534 33 34 535 In this study, we present our preferred plate reconstruction model for the evolution of the western 35 36 37 536 Tethys in the framework of Atlantic and Tethyan kinematics (Fig. 7) (Model 1a, see Table 3). 38 39 537 This reconstruction uses the model of Schettino and Turco (2011) and Schettino and Turco 40 41 538 (2006) for the western Tethys; this study for the kinematics of the Atlantic domain and Seton et 42 43 44 539 al. (2012) for imbedding into a global framework (Fig. 7) , we rely most heavily on the model 45 46 540 of Schettino and Turco (2011) for the Western Tethys model; this choice is largely driven 47 48 49 541 by the availability of poles of rotation accompanying this model, in contrast to other models 50 51 542 discussed above, and in our assessment of seismic tomography we consider a range of 52 53 543 scenarios previously described. The absolute reference frame used in this model is based on 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 25 of 73 International Geology Review
1 2 3 544 moving Indo Atlantic hotspots from 100 Ma to the present day (O’Neill et al., 2006) and the 4 5 6 545 TPW corrected reference frame (Steinberger and Torsvik, 2008) for times older than 100 Ma. 7 8 546 We compare different aspects of Model 1a to three alternative models for the evolution of the 9 10 11 547 western Tethys: Seton et al. (2012) (Model 2); Stampfli et al. (2002) and Stampfli and Borel 12 13 548 (2004) (Model 3); and Sibuet et al. (2004) (Model 4). The tectonic reconstructions for all models 14 For Peer Review Only 15 549 are discussed in detail in their relevant publications and the main features of each model have 16 17 18 550 been highlighted and compared in Section 2 and Tables 1 and 2 of this study. 19 20 551 As it is difficult to assess the validity of competing models from the geological record alone, the 21 22 552 main focus of our study is to explore the relationships between the evolution of the Western 23 24 25 553 Tethys and deep mantle processes by comparing plate reconstructions to seismic tomography. As 26 27 554 such, we focus on episodes of larger scale, subduction related events in plate Model 1a. This 28 29 555 model covers the last 200 million years, as we are primarily interested in capturing the older 30 31 32 556 (Jurassic and Cretaceous) poorly constrained record of subduction and the main cycles of back 33 34 557 arc basin opening and closure. As the absolute motion of the plates is subject to increasing 35 36 37 558 uncertainty back in time, we explore the consequences of alternative reference frames for the 38 39 559 absolute plate motions within our study area by creating an alternative model [Model 1b that 40 41 560 uses Van der Meer et al. (2010)] subduction reference frame of Van der Meer et al. (2010). 42 43 44 561 Both Model 1a and 1b were defined within the new generation of global plate reconstruction 45 46 562 models with self consistent, dynamically evolving plate boundaries and polygons through time 47 48 563 (Gurnis et al., 2012; Seton et al., 2012). The plate boundaries allow us to directly compare the 49 50 51 564 timing and absolute location of subduction predicted by different reconstructions with 52 53 565 predictions of past subduction from seismic tomography [e.g. Shephard et al. (2013)]. 54 55 566 (Figure 8 and Table 3 near here) 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 26 of 73
1 2 3 567 4 5 6 568 4. Insights from Seismic tomography 7 8 569 9 10 11 570 Deep mantle thermal and chemical structure can be revealed and interpreted through seismic 12 13 571 tomography (Nolet, 1987). This technique is based on measuring travel times of P and S seismic 14 For Peer Review Only 15 572 waves through the mantle, and compiling information from large numbers of different ray paths 16 17 18 573 into 3D images of earth’s velocity structure. 19 20 574 High velocity structures are typically associated with subducted slab material and slower 21 22 575 velocities to mantle upwellings. Once subducted slabs enter the mantle, the time taken for them 23 24 25 576 to sink to the base of the lower mantle may be more than 200 Myr (Butterworth et al., 2014; Van 26 27 577 der Meer et al., 2010). Therefore, seismic tomography models are used to test the proposed 28 29 578 tectonic reconstruction of Western Tethys from 200 Ma to the present to investigate how the 30 31 32 579 observed distribution of high velocity material in the mantle corresponds with reconstruction 33 34 580 model predictions for the location, timing and duration of subduction in this region. 35 36 37 581 38 39 582 40 41 583 4.1 Method of Analysis 42 43 44 584 45 46 585 Here, we compare one global S wave mantle seismic tomography model of GrandS (Grand, 47 48 586 2002) and two P wave models of MITP 08 (Li et al., 2008), and GypsumP (Simmons et al., 49 50 51 587 2010). We assess vertical cross sections across the predicted locations of subduction zones in 52 53 588 three domains: the Vardar Ocean, Alpine Tethys and Pyrenees. 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 27 of 73 International Geology Review
1 2 3 589 We infer the approximate location and depth of subducted slabs in the Tethyan realm by 4 5 6 590 assuming that oceanic lithosphere sinks vertically into the mantle (Grand et al., 1997; Van der 7 8 591 Meer et al., 2010). The rate of this vertical sinking is higher in the upper mantle and decreases 9 10 11 592 when the slab penetrates into the more viscous lower mantle (Lithgow ‐Bertelloni and Richards, 12 13 593 1998). Slab stagnation in the mantle transitional zone (above 660 km depth) is an factor that 14 For Peer Review Only 15 594 affects the shape and kinematics of slab sinking and thus the correlation between plate 16 17 18 595 reconstruction and tomographic models (Forte and Mitrovica, 2001). Various studies on slab 19 20 596 stagnation have suggested that this is a common phenomenon in the circum Pacific and 21 22 597 Mediterranean areas (Capitanio et al., 2009; Fukao et al., 2009; Yoshioka and Naganoda, 2010). 23 24 25 598 These issues, along with the possibility of complex mantle flow and lateral slab migration, can 26 27 599 be properly explored within numerical convection models constrained assimilating data from 28 29 30 600 paleogeographic reconstructions (Bower et al., 2013; Hassan et al., 2015), which is beyond the 31 32 601 scope of this study. Thus, with these limitations in mind, we here consider a rate of 1.2 cm/yr as 33 34 602 an average sinking rate, as has been suggested for the Aegean (Vardar) slab by Van der Meer et 35 36 37 603 al. (2010), and consistent with average sinking rates inferred from geodynamic modelling 38 39 604 (Hassan et al., 2015). 40 41 605 Another factor with implications for linking past plate motions with seismic tomography is the 42 43 44 606 choice of absolute reference frame. While the assumed sinking rate influences what depth level 45 46 607 within seismic tomography we relate to reconstructed subduction zones of a given age, our 47 48 49 608 choice of absolute reference frame influences the geographic location (in longitude/latitude 50 51 609 space) of the reconstructed subduction zones. Previous studies have typically relied on reference 52 53 610 frames derived from some combination of hotspot trails and paleomagnetic data (O'Neill et al., 54 55 56 611 2005; Torsvik et al., 2008a). However, Van der Meer et al. (2010) argue that significant 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 28 of 73
1 2 3 612 longitudinal shifts (compared to a hybrid hotspot/paleomagnetic reference frame) are required to 4 5 6 613 match tomographic signatures for three slabs (Farallon, Aegean Tethys and Mongol Okhotsk), 7 8 614 particularly for times older than ~80 Ma where constraints from hotspot trails are sparse or 9 10 11 615 absent (Fig. 8). This factor will be further discussed in the section 5. 12 13 616 It should be noted that another limiting factor in the interpretation of subduction history from 14 For Peer Review Only 15 617 seismic tomography images are the uncertainties in these images themselves. Lack of global 16 17 18 618 seismic waves coverage can lead to interpolation artefacts in areas of poor data coverage, which 19 20 619 can get influenced by adjacent well constrained regions and not necessarily corresponding with 21 22 620 the true nature of deep earth in these regions (Anderson, 2007). Limitations in the resolution and 23 24 25 621 robustness of seismic tomography images can be qualitatively assessed to some degree by 26 27 622 comparing different models and identifying which features are consistently imaged across a 28 29 623 range of models, which we do here as part of our evaluation of mantle structure in selected cross 30 31 32 624 sections. 33 34 625 35 36 37 626 4.2 Western Tethys Profiles 38 39 627 40 41 628 According to our Model 1a, the subduction related events in the western Tethys can be divided 42 43 44 629 into two major categories: the older events that occur over the Jurassic and Cretaceous time 45 46 630 period (200 67 Ma) and involve older subduction in the Vardar and Meso Neotethys Oceans 47 48 631 (profile 5) and Alpine Tethys (Piemont and Ligurian Ocean) (profile 4); and the Tertiary events 49 50 51 632 that cover the closure of the Vardar Ocean (Hellenic subduction) (profile 1) and subduction in 52 53 633 the Alpine (Piemont and Ligurian Oceans), Pennenic (Valais Ocean), Apennenic (Calabrian arc) 54 55 634 (profile 2) and proposed subduction for the Pyrenees Bay of Biscay (profile 3). 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 29 of 73 International Geology Review
1 2 3 635 Cenozoic events have been extensively studied in the light of shallow mantle seismic 4 5 6 636 tomography (Bijwaard and Spakman, 2000; Handy et al., 2010; Piromallo and Morelli, 2003; 7 8 637 Spakman and Wortel, 2004; Wortel and Spakman, 2000). Many aspects of the first order 9 10 11 638 velocity structure can be robustly related to recent subduction, although there are still 12 13 639 disagreements about the detailed interpretation of high velocity bodies visible in the upper 14 For Peer Review Only 15 640 mantle beneath the Mediterranean region. Establishing links between tomography and Jurassic 16 17 18 641 and Cretaceous aged subduction events is more challenging, due to poorly constrained plate 19 20 642 reconstructions, greater uncertainty over the characteristics of the lower mantle and slab 21 22 643 behaviour as it sinks deeper into the mantle, and lower resolution and reliability of tomographic 23 24 25 644 models for the deeper regions of the mantle. The aim of the following analysis is to explore the 26 27 645 signatures in seismic tomography models in relation to different relative and absolute plate 28 29 646 reconstruction scenarios, with a focus on Jurassic and Cretaceous subduction within the 30 31 32 647 Neotethys and Alpine Tethys. A secondary aim is to discuss the younger events in some 33 34 648 controversial areas. 35 36 37 649 In the next section we discuss 5 tomography profiles across the region beneath North Africa, the 38 39 650 Mediterranean and southern Eurasia. We discuss the profiles in terms of what we expect to see in 40 41 651 each profile based on Model 1a and the series of assumptions outlined above, what is actually 42 43 44 652 observed; and how dependent the observed seismic anomalies are on the choice of tomography 45 46 653 model. 47 48 654 (Figure 9 near here) 49 50 51 655 52 53 656 4.2.1 Profile 1: Vardar (120-69 Ma) and eastern Mediterranean subduction (Hellenic Arc) (56 54 55 657 Ma-present time)(Fig. 9) 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 30 of 73
1 2 3 658 4 5 6 659 Expected: According to Model 1a, this profile is parallel to the direction of convergence 7 8 660 between Africa Eurasia in the western Tethys and thus it is expected to image the subducted 9 10 11 661 material of the Vardar Ocean and Hellenic Arc. The predicted maximum depth of high velocity 12 13 662 material is around 1500 km. As the Hellenic subduction is still continuing in the western part, we 14 For Peer Review Only 15 663 expect to see the attached slab in tomographic models. Fig. 9 shows the profile 1 and the vertical 16 17 18 664 slices of three different tomographic models along this profile. 19 20 665 21 22 666 Observed: Three anomalies are detected in the shallow and deep mantle across all tomographic 23 24 25 667 models. The slab materials from ongoing Cenozoic (~55 0 Ma) Mediterranean subduction 26 27 668 (Hellenic Arc) are visibly attached to the surface and dip up to the 660 km transition zone 28 29 669 (Anomaly A). This anomaly was investigated in previous studies (Bijwaard et al., 1998; 30 31 32 670 Capitanio et al., 2009; Faccenna et al., 2003; van Hinsbergen et al., 2005; Widiyantoro et al., 33 34 671 2004). A NE dipping anomaly (Anomaly B) is located in the mid mantle between 660 and 1500 35 36 37 672 km. According to models 1a, 2 and 3 this anomaly can be correlated with the slab material from 38 39 673 the Vardar and Aegean basins that subducted between 120 and 55 Ma. The southwestward 40 41 674 extension of anomaly B between 25°N and 37°N at a depth of 1000 to 2000 km (Anomaly C) 42 43 44 675 was the topic of many studies and different suggestions have been made to explain this 45 46 676 deflection, including that the distortion was due to slab folding in the transition zone (Faccenna 47 48 677 et al., 2003), consequent flushing of stagnated slab material as “slab avalanches” (Capitanio et 49 50 51 678 al., 2009), or material belonging to Jurassic Vardar subduction (van Hinsbergen et al., 2005). All 52 53 679 these studies consider the existence of only one subduction zone from Early Cretaceous to the 54 55 680 present that retreats southward at around 50 Ma and initiates the subduction in the Aegean region 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 31 of 73 International Geology Review
1 2 3 681 (Faccenna et al., 2003; van Hinsbergen et al., 2005; Wortel and Spakman, 2000). Model 1a in 4 5 6 682 contrast, predicts the formation of a new subduction zone in the Aegean region at ~70 Ma that 7 8 683 moves northward as a result of northward motion of Africa and Eurasia (Fig. 9). This scenario 9 10 11 684 therefore predicts the existence of slab material from both the Vardar and Aegean subduction in 12 13 685 mid mantle between 660 1500 km, which can be responsible for thickening and extension of the 14 For Peer Review Only 15 686 southern tip of the slab (Anomaly C). The deeper part of anomaly C is located below 1500 km 16 17 18 687 (more distinguishable in MITP 08 and GypsumP models) and has previously been attributed to 19 20 688 Jurassic subduction of the western Vardar Ocean by van Hinsbergen et al. (2005). However, 21 22 689 more than 12° westward offset between the Jurassic paleo subduction zone of the Vardar Ocean 23 24 25 690 reconstructed with model 1 and associated slab material in the deep mantle (Fig. 9) makes this 26 27 691 correlation less apparent. We tentatively suggest that anomaly C might represent the subducted 28 29 692 material from Late Triassic Early Jurassic oceanic bodies from the subduction of earlier 30 31 32 693 generations of Paleotethys back arc basins (e.g. Kure, Izmir Ancara and eastern Vardar oceans) 33 34 694 along the Eurasian margin (Schettino and Turco, 2011; Stampfli et al., 2002). 35 36 37 695 38 39 696 40 41 697 4.2.2 Profile 2: Alpine and Calabrian subduction (70 Ma- present time) (Fig. 10) 42 43 44 698 45 46 699 Expected: this longitudinal profile cuts through the western, central and eastern Mediterranean 47 48 700 regions (Fig. 10). According to models 1 to 3 the subducted material beneath this area belongs to 49 50 51 701 Cenozoic (70 Ma and younger) subduction events in the Alps, Apennines and Aegean regions. 52 53 702 We expect to find these materials at the surface to a maximum depth of 800 km. The model 54 55 703 predicts the existence of slab material from west dipping Ligurian Calabrian subduction in the 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 32 of 73
1 2 3 704 Apennines, and the north dipping Aegean slab in the eastern Mediterranean region on the far east 4 5 6 705 side of the profile. 7 8 706 9 10 11 707 Observed: A massive volume of shallow high velocity material is present in all the tomography 12 13 708 models that make it difficult to attribute each anomaly to a separate subduction zone. Possible 14 For Peer Review Only 15 709 distinct regions of upper mantle slab material are best distinguished in MIT P08 and Gypsum P 16 17 18 710 models. 19 20 711 Anomaly A is the westernmost high speed body that is present in all tomographic models. The 21 22 712 main part of this anomaly lies in the 660 km transitional zone although in MIT P08 and GrandS 23 24 25 713 models it extends down to 900 km depth. Based on Model 1 and 2, we interpret anomaly A as 26 27 714 the west dipping Late Cretaceous Eocene Ligurian Ocean subducted material under Iberia. 28 29 715 However, it is difficult to directly infer the polarity of this subduction due to the highly flattened 30 31 32 716 morphology of this anomaly. Anomaly B is located between the surface and the upper mid 33 34 717 mantle transitional zone under the Tyrrhenian back arc basin. This anomaly dips steeply to 600 35 36 37 718 km depth and then is highly flattened and stretched westward along the transition zone for more 38 39 719 than 10°. This anomaly corresponds well with the location of Calabrian slab predicted in the 40 41 720 tectonic Models 1 and 2. The prolongation of slab material is reflective of the rapid westward 42 43 44 721 retreat of the Calabrian subduction zone during Oligocene time. This slab is extensively 45 46 722 discussed in other studies and tomographic models (Bijwaard et al., 1998; Piromallo and Morelli, 47 48 723 2003; Spakman and Wortel, 2004). To the east lies the main body of Aegean subducted material 49 50 51 724 (anomaly C) from the surface to the upper mid mantle transitional zone. This anomaly correlates 52 53 725 with anomaly A that is visible along profile 1 (Fig. 9). 54 55 726 (Figure 10 near here) 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 33 of 73 International Geology Review
1 2 3 727 4 5 6 728 4.2.3 Profile 3: Pyrenees (Fig. 11) 7 8 729 9 10 11 730 Expected: This profile cuts through the Pyrenees orogeny and is parallel to Iberia Eurasia 12 13 731 compression. Model 1 predicts no oceanic crust formed under Pyrenees and hence no oceanic 14 For Peer Review Only 15 732 subducted material is expected to be present under the Pyrenees along this profile. 16 17 18 733 19 20 734 Observed: Anomaly A and B are observed in all models in the shallow mantle between the 21 22 735 surface and 400 km depth. Sibuet et al. (2004) (Model 4) attributes the southern anomaly A to 23 24 25 736 the Neotethys slab (120 85 Ma) and the northward anomaly B to subducted lower crust of 26 27 737 Pyrenees rift (85 25 Ma). This rift was assumed to open as a back arc basin during Neotethys 28 29 738 subduction. Considering the sinking rate used in our study, we expect to see the main body of the 30 31 32 739 slab material of these two subduction zones at ~1400 to 1000 and 1000 to 300 km, respectively. 33 34 740 None of our tomographic models show such high velocity bodies at associated depths with these 35 36 37 741 inferred subduction zones. This result is also consistent with the tomographic study of Spakman 38 39 742 and Wortel (2004) and Souriau et al. (2008) who also report no slab material under the Pyrenees 40 41 743 in global and regional tomographic models. Anomaly C that lies horizontal in the transitional 42 43 44 744 zone corresponds to the Ligurian (Betic) subduction zone and was previously discussed as 45 46 745 anomaly A along profile 2 (Fig. 10). 47 48 746 (Figure 11 near here) 49 50 51 747 52 53 748 4.2.4 Profile 4: Pre-Alpine (Mesozoic Alpine) subduction (130-70 Ma) (Fig. 12) 54 55 749 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 34 of 73
1 2 3 750 Expected: Our Model 1 plate reconstruction predicts the existence of slab material related to the 4 5 6 751 entire sequence of Alpine subduction along this profile. According to this model, oceanic 7 8 752 subduction starts as early as 130 Ma in the Piemont Ocean and continues until around 40 Ma 9 10 11 753 followed by collision, slab break up and continental subduction. We expect to see the deepest 12 13 754 anomaly at ~1500 km depth while the shallow continental part of Alpine subduction should still 14 For Peer Review Only 15 755 be attached to the surface in the northern section of the profile. 16 17 18 756 19 20 757 Observed: Two separate slab like bodies are recognised under the Mediterranean area in all four 21 22 758 models (anomalies A and B). Both of these anomalies are still attached to the surface while the 23 24 25 759 deepest parts of Anomaly B can be traced as laying horizontal in the upper mid mantle 26 27 760 transitional zone. The deeper high velocity volumes (anomalies C and D) are located between 28 29 761 ~1000 1500 km, corresponding to a subduction event that happened between ~130 and 80 Ma. 30 31 32 762 The main body of the southernmost anomaly D is located under southern Libya (25°N) and is 33 34 763 identifiable in all tomographic models while the extension of anomaly C is not as clear. In the 35 36 37 764 MITP 08 model it forms a separate anomalous body under the Mediterranean Sea while in three 38 39 765 other models the boundary between this anomaly and anomalies B and D is obscure. Model 1 40 41 766 predicts the paleo location of mid Late Cretaceous Alpine subduction zones between 130 and 80 42 43 44 767 Ma to be located above anomaly D and C, respectively. Based on this model, we suggest that 45 46 768 anomaly C and D belong to a single subduction event that starts closing the Mesozoic Alpine 47 48 769 Tethys from the Cretaceous until the Cenozoic. The shallowing of these anomalies toward the 49 50 51 770 north is also in agreement with northward migration of Piemont subduction zone. Anomaly D 52 53 771 was also referred to by Van der Meer et al. (2010) as the Egypt (Eg) slab and was attributed to an 54 55 772 eastward subduction on the eastern margin of Menderes Taurides block in the Vardar domain 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 35 of 73 International Geology Review
1 2 3 773 between 120 and 80Ma (Stampfli and Borel, 2004) (Model 3). It should be noted that Model 1 4 5 6 774 does not support the existence of such a subduction zone. Furthermore, the polarity and paleo 7 8 775 location of this subduction zone from 120 80 Ma in Van der Meer et al. (2010) does not match 9 10 11 776 the referred trench in Stampfli and Borel (2004) that adds to uncertainties in connecting this high 12 13 777 velocity volume to the so called subduction zone. In the upper mantle, we attribute anomaly B to 14 For Peer Review Only 15 778 Cenozoic subduction of the Piemont Ocean. This high velocity mass is well identified and 16 17 18 779 studied in shallow mantle tomography models of the Mediterranean area (Piromallo and 19 20 780 Faccenna, 2004; Piromallo and Morelli, 2003; Spakman and Wortel, 2004). The shallow part of 21 22 781 this anomaly represents the Cenozoic oceanic subduction of Piemont Ocean while the hanging 23 24 25 782 slab can be attributed to post break up continental subduction. Ultimately, anomaly A represents 26 27 783 the shallowest high speed anomaly in the upper mantle. The location of this anomaly is in 28 29 784 agreement with the Calabrian slab that was already identified and discussed under the name of 30 31 32 785 anomaly B along Profile 2 and Fig. 10. 33 34 786 (Figure 12 near here) 35 36 37 787 38 39 788 4.2.5 Profile 5: Intra-oceanic Vardar (170-120 Ma) and western Neotethys subduction (100-70 40 41 789 Ma) (Fig. 13) 42 43 44 790 45 46 791 Expected: Model 1a predicts the presence of slab material associated with Vardar Ocean 47 48 792 subduction from Mid Jurassic to Late Jurassic/Early Cretaceous (170 to 130 Ma) within this 49 50 51 793 cross section. This northeast dipping intra oceanic subduction is progressively retreating 52 53 794 eastward and consuming the oceanic lithosphere between the island arc and the Adriatic margin. 54 55 795 The total amount of oceanic subduction inferred from the tectonic model is around 300 km over 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 36 of 73
1 2 3 796 25 Myr. According to slab sinking rates and considering uncertainties in these rates, material 4 5 6 797 from this subduction should be observed in the lower mantle under present day north central 7 8 798 Africa at depths between 1400 to 2200 km. The second feature predicted by Model 1 is 9 10 11 799 subducted material from the intra oceanic subduction zone of the western Neotethys between 12 13 800 Africa/Arabia and Iran (Eurasia) from 100 to 80 Ma. According to this model, these slab 14 For Peer Review Only 15 801 materials should be observable between 1000 1500 km depth below the west bank of Red Sea. 16 17 18 802 19 20 803 Observed: The Grand S and Gypsum P tomographic models show three distinct high velocity 21 22 804 anomalies in the mantle (A, B and C in Fig. 13), whereas the shallow anomaly is not recognised 23 24 25 805 in the MIT 08 model as it is imaged in a more northward location in this model (Fig. 12, 26 27 806 anomaly D). Anomaly A is visible under Northwest Africa between 1500 2300 km depth while 28 29 807 anomaly B is located between 20°E 35°E at depths between 800 1400 km. Anomaly C is 30 31 32 808 situated below 1000 km depth and extended deeper than 2500 km. All the anomalies are highly 33 34 809 flattened and extended longitudinally. 35 36 37 810 Anomaly B is located beneath the reconstructed location of Vardar intra oceanic subduction 38 39 811 from 170 to 145 Ma in Model 1. However, the shallow depth of this anomaly is difficult to 40 41 812 reconcile with the age inferred from the tectonic model for Vardar subduction. The age of about 42 43 44 813 70 120 Ma is what is inferred from the sinking rate for subducting slab related to this anomaly. 45 46 814 Anomaly A on the other hand is located at depths consistent with the plate reconstructions for the 47 48 815 early Vardar slab material, but it is located far to the west, offset from the associated paleo 49 50 51 816 subduction zone in Model 1 by more than 15°. Similar to Van der Meer et al. (2010), we refer to 52 53 817 this high velocity feature as the remnants of western Vardar intra oceanic subduction zone from 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 37 of 73 International Geology Review
1 2 3 818 170 to 140 Ma. Nonetheless, as noted by Van der Meer et al. (2010), we are unable to explain the 4 5 6 819 significant lateral offset between predicted and observed anomalous high velocity structures. 7 8 820 Anomaly B corresponds to the Egypt (Eg) slab in the tomographic study of Van der Meer et al. 9 10 11 821 (2010) between 1325 1175 km depths. As discussed in profile 4 we attribute this high velocity 12 13 822 material to Mesozoic subduction in Alpine Tethys (Anomaly D in figure 12). 14 For Peer Review Only 15 823 Anomaly C is observed beneath Arabia at depths between 1000 to more than 2000 km. The 16 17 18 824 depth range for this slab material corresponds to subduction aged between 160 to 80 Ma. This 19 20 825 feature is also mentioned as the Mesopotamia slab (Me) (Van der Meer et al., 2010), Eg SA slab 21 22 826 (Hafkenscheid et al., 2006) and western part of Zone III of Neotethys slabs (Van der Voo et al., 23 24 25 827 1999). However, there is no general consensus about the location and origin of the subduction 26 27 828 zone responsible for this slab material. Here we associate the eastern part of this anomaly with 28 29 829 Neotethys northeastward dipping subduction under the Eurasian margin (Model 1 and 2). The 30 31 32 830 western part of this anomaly can correlate with the intra Oceanic western Neotethys subduction 33 34 831 zone (Model 1 and Model 3) (Hafkenscheid et al., 2006). However, we are unable to 35 36 37 832 discriminate between these models to explain the origin of the high velocity material in mid deep 38 39 833 mantle. 40 41 834 (Figure 13 and 14 near here) 42 43 44 835 45 46 836 5. Discussion 47 48 837 49 50 51 838 Coupling the plate reconstruction from Model 1 with different tomographic models mentioned in 52 53 839 section 4.1, results in a satisfactory match between the Cenozoic subduction events in western 54 55 840 Tethys region and observed tomographic high velocity material. However, this match gets less 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 38 of 73
1 2 3 841 adequate as we go back in time to the Mesozoic era. The considerable lateral offset between the 4 5 6 842 deep mantle feature regarded as the Vardar subducted material and the paleo longitude of the 7 8 843 Vardar intra oceanic subduction zone is one of these uncertainties. 9 10 11 844 The same uncertainties extend to the Neotethys Ocean. Our tomographic study shows that a 12 13 845 short lived intra oceanic arc in the Western Tethys Ocean that is incorporated in tectonic Model 14 For Peer Review Only 15 846 1 can be responsible for high velocity material under Northwest Africa and the Arabian 16 17 18 847 Peninsula. Nevertheless, the possibility of opening a Cretaceous aged back arc basin on the 19 20 848 Eurasian margin (e.g. Semail Ocean Hafkenscheid et al. (2006); Stampfli and Borel (2004)) 21 22 849 (Model 3) as a western counterpart of the Bangong Nujiang back arc Ocean in the Eastern 23 24 25 850 Tethys Ocean (Gibbons et al., 2015) can not dismissed. The westward retreat of the Neotethys 26 27 851 subduction zone resulting from this opening could result in arc continent collision and obduction 28 29 852 of Tethyan oceanic crust on the Arabian margin (Semail Oman ophiolites). 30 31 32 853 The existence of oceanic subduction in the Alpine or pre Alpine Tethys basins (Channell and 33 34 854 Kozur, 1997; Rosenbaum and Lister, 2005) suggested by Model 1 is not certain. There are 35 36 37 855 suggestions that the Mesozoic phase of Alpine convergence especially in the Eastern Alps and 38 39 856 Carpathians was mostly continental (Handy et al., 2010; Piromallo and Faccenna, 2004; Stüwe 40 41 857 and Schuster, 2010) (Model 3). In the latter tectonic scenario, the tomographic volumes regarded 42 43 44 858 as subducted Mesozoic Alpine Tethys material could be associated with Pindos subduction 45 46 859 (Robertson and Shallo, 2000; Stampfli and Borel, 2004) (Model 2) or could be the result of 47 48 860 subduction of younger generations of back arc basins that opened and closed within the oceanic 49 50 51 861 domain of Vardar Ocean from Late Cretaceous time (e.g. Sava basin; Schmid et al. (2008)). 52 53 862 We mentioned previously that part of the discrepancies between the reconstruction and 54 55 863 tomographic models may arise from uncertainties in the absolute position of the continents 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 39 of 73 International Geology Review
1 2 3 864 through time, or due to the limitations in the assumption of vertical slab sinking and sinking 4 5 6 865 rates. Differences between the absolute reference frames results in notable variations in plate 7 8 866 velocities and the paleo location of plate boundaries (Shephard et al., 2013). This problem is 9 10 11 867 more acute for older times due to less reliable data (e.g. lack of hotspot tracks, limited 12 13 868 paleomagnetic data and poorly imaged subduction zones in the deep mantle). To partly address 14 For Peer Review Only 15 869 this issue, we created an alternative plate model (Model 1b), which takes the relative plate 16 17 18 870 motion of Model 1a but uses an alternative absolute reference frame of Van der Meer et al. 19 20 871 (2010). The difference between these two reference frames is not very significant for Cenozoic 21 22 872 times but major differences exist for the Mesozoic (Fig. 8), which is the focused time period of 23 24 25 873 this study. Fig. 14 shows the displacement of subduction zones using these two alternative 26 27 874 reference frames. This figure also shows paleo subduction zones of the model 2 for the 28 29 875 associated time steps to better compare these tectonic scenarios. 30 31 32 876 Applying the subduction reference frame of van der Meer (2010) results in a longitudinally 33 34 877 westward displacement of the subduction zones (in Model 1b) of about 850, 540, 700 and 500 35 36 37 878 km for times 160, 111, 84 and 40 Ma, respectively. The maximum displacement affects the 38 39 879 Vardar subduction system and partially accounts for the lateral offset between the Vardar 40 41 880 subduction zone and associated high velocity anomaly (Anomaly A in Fig 13. Also see previous 42 43 44 881 section). The influence of this westward shift is more pronounced in north south striking 45 46 882 subduction zones and thus the more east west trending pre Alpine, Alpine and Hellenic 47 48 883 subduction zone are less sensitive to this change in absolute reference frame. The Late 49 50 51 884 Cretaceous trench location of Western Neotethys in the Van der Meer et al. (2010) reference 52 53 885 frame is also not dramatically different due to the more oblique orientation of this subduction 54 55 886 zone. The geographic fit between the tectonic and tomographic model for the Cenozoic is better 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 40 of 73
1 2 3 887 in the absolute reference frame of Model 1a, supporting the notion that hotspot based reference 4 5 6 888 frames are a better predictor of absolute plate motions for Cenozoic times. 7 8 889 9 10 11 890 12 13 891 6. Conclusion 14 For Peer Review Only 15 892 16 17 18 893 We have constructed a plate tectonic model of the Western Tethys that incorporates a new model 19 20 894 for the early opening of the Central and North Atlantic conducted in this study and the 21 22 895 reconstructions of Schettino and Turco (2011) for the blocks and basins in the western Tethys, 23 24 25 896 imbedded within the global plate reconstruction of Seton et al. (2012). 26 27 897 Our modifications of the early kinematics in the Atlantic results in a revised motion of Iberia 28 29 898 relative to Europe and Africa. We support a transtensional tectonic regime along the North 30 31 32 899 Pyrenean Fault for Iberia Eurasia boundary from 200 to 120 Ma. As a consequence of such 33 34 900 motion, our model does not support the interpretation of oceanic magnetic lineations in the Bay 35 36 37 901 of Biscay as true M0 magnetic isochrons. 38 39 902 Comparing the new plate reconstruction of the Western Tethys to different P and S seismic wave 40 41 903 tomographic models, we found a good match between the plate kinematics and tomography 42 43 44 904 models for Late Cretaceous Cenozoic subduction of eastern Vardar Ocean eastern 45 46 905 Mediterranean (Hellenic Arc) and Alpine subduction. We were unable to recognize an oceanic 47 48 906 slab under the Pyrenees, as suggested by Sibuet et al. (2004), which reinforces our Iberian plate 49 50 51 907 reconstruction regarding the absence of oceanic crust under the Pyrenees. The possible 52 53 908 association between the high seismic velocity material under North Central Africa and the 54 55 909 Mesozoic subduction of the Alpine basins implemented in the reconstruction suggests the 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 41 of 73 International Geology Review
1 2 3 910 possibility of the Alpine compressional phase initiating as an oceanic subduction alongside with 4 5 6 911 the Eo Alpine continental subduction. A final prediction that the tectonic model makes concerns 7 8 912 the paleo location of Vardar intra oceanic subduction under Central Africa. The deep mantle 9 10 11 913 high velocity material in tomographic models under NW Africa may be related to this 12 13 914 subduction. However, our tectonic reconstruction predicts this material to be farther eastward 14 For Peer Review Only 15 915 than what is predicted when examining correlations with seismic tomography. We found that 16 17 18 916 using an alternative reference frame (VDM) only partially accounts for this misfit and does not 19 20 917 solve the discrepancy entirely. 21 22 918 Our comparison of the plate tectonic models to seismic tomography is an important first step to 23 24 25 919 understanding the tectonic evolution of the western Tethys evolution and assessing the validity 26 27 920 of competing tectonic scenarios. We envisage that coupled plate tectonic geodynamic models, 28 29 921 with imposed plate motions and subduction scenarios studied in this paper as boundary 30 31 32 922 conditions, will shed light on the ongoing debates regarding the broad scale subduction history 33 34 923 of Western Tethys and the early history of Vardar and Alpine subduction in particular. 35 36 37 924 38 39 925 40 41 926 Acknowledgements 42 43 44 927 We would like to thank support from the Australian Research Council via grant FL0992245, 45 46 928 (MH, SW, RDM) and FT130101564 (MS). MH also thanks for the CSIRO OCE Ph.D 47 48 929 scholarship. The rotation and dynamic polygon files of our preferred model (Model 1a) have 49 50 51 930 been published and is publically available as part of (Müller et al., 2016) paper. 52 53 931 54 55 932 References 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 42 of 73
1 2 3 933 4 5 6 934 Anderson, D.L., 2007. Is there convincing tomographic evidence for whole mantle 7 935 convection. 8 936 Beauchamp, W., Allmendinger, R.W., Barazangi, M., Demnati, A., El Alji, M. and Dahmani, M., 9 937 1999. Inversion tectonics and the evolution of the High Atlas Mountains, Morocco, 10 11 938 based on a geological ‐geophysical transect. Tectonics, 18(2): 163-184. 12 939 Beauchamp, W., Barazangi, M., Demnati, A. and El Alji, M., 1996. Intracontinental rifting and 13 940 inversion: Missour basin and Atlas mountains, Morocco. AAPG bulletin, 80(9): 1459- 14 941 1481. For Peer Review Only 15 16 942 Bijwaard, H. and Spakman, W., 2000. Non-linear global P-wave tomography by iterated 17 943 linearized inversion. Geophysical Journal International, 141(1): 71-82. 18 944 Bijwaard, H., Spakman, W. and Engdahl, E.R., 1998. Closing the gap between regional and 19 945 global travel time tomography. Journal of Geophysical Research: Solid Earth (1978– 20 946 2012), 103(B12): 30055-30078. 21 22 947 Blanckenburg, F. and Davies, J.H., 1995. Slab breakoff: a model for syncollisional 23 948 magmatism and tectonics in the Alps. Tectonics, 14(1): 120-131. 24 949 Bortolotti, V., Chiari, M., Marcucci, M., Photiades, A., Principi, G. and Saccani, E., 2008. New 25 950 geochemical and age data on the ophiolites from the Othrys area (Greece): 26 951 implication for the Triassic evolution of the Vardar Ocean. Ofioliti, 33(2): 135-151. 27 28 952 Bower, D.J., Gurnis, M. and Seton, M., 2013. Lower mantle structure from 29 953 paleogeographically constrained dynamic Earth models. Geochemistry, Geophysics, 30 954 Geosystems, 14(1): 44-63. 31 955 Bronner, A., Sauter, D., Manatschal, G., Péron-Pinvidic, G. and Munschy, M., 2011. Magmatic 32 33 956 breakup as an explanation for magnetic anomalies at magma-poor rifted margins. 34 957 Nature Geoscience, 4(8): 549-553. 35 958 Bronner, A., Sauter, D., Munschy, M., Carlut, J., Searle, R., Cannat, M. and Manatschal, G., 36 959 2014. Magnetic signature of large exhumed mantle domains of the Southwest Indian 37 960 Ridge–results from a deep-tow geophysical survey over 0 to 11 Ma old seafloor. 38 39 961 Solid Earth, 5(1): 339-354. 40 962 Butterworth, N., Talsma, A., Müller, R., Seton, M., Bunge, H.-P., Schuberth, B., Shephard, G. 41 963 and Heine, C., 2014. Geological, tomographic, kinematic and geodynamic constraints 42 964 on the dynamics of sinking slabs. Journal of Geodynamics, 73: 1-13. 43 965 Capitanio, F., Faccenna, C. and Funiciello, R., 2009. The opening of Sirte basin: Result of slab 44 45 966 avalanching? Earth and Planetary Science Letters, 285(1): 210-216. 46 967 Channell, J. and Kozur, H., 1997. How many oceans? Meliata, Vardar and Pindos oceans in 47 968 Mesozoic Alpine paleogeography. Geology, 25(2): 183-186. 48 969 Choukroune, P. and Seguret, M., 1973. Tectonics of the Pyrenees: role of compression and 49 970 gravity. Gravity and Tectonics: 141-156. 50 51 971 De Lamotte, D.F., Leturmy, P., Missenard, Y., Khomsi, S., Ruiz, G., Saddiqi, O., Guillocheau, F. 52 972 and Michard, A., 2009. Mesozoic and Cenozoic vertical movements in the Atlas 53 973 system (Algeria, Morocco, Tunisia): an overview. Tectonophysics, 475(1): 9-28. 54 974 Dean, S., Minshull, T., Whitmarsh, R. and Louden, K., 2000. Deep structure of the ocean ‐ 55 56 975 continent transition in the southern Iberia Abyssal Plain from seismic refraction 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 43 of 73 International Geology Review
1 2 3 ‐ ′ 4 976 profiles: The IAM 9 transect at 40 20 N. Journal of Geophysical Research: Solid 5 977 Earth (1978 –2012), 105(B3): 5859-5885. 6 978 Dercourt, J., Ricou, L.-E. and Vrielynck, B., 1993. Atlas Tethys Palaeoenvironmental Maps: 7 979 Explanatory Notes. Liste Des Articles, Communications, Cartes Et Thèses. 8 9 980 Programme GS-Téthys, 1986-1993. Gauthier-Villars. 10 981 Faccenna, C. and Becker, T.W., 2010. Shaping mobile belts by small-scale convection. 11 982 Nature, 465(7298): 602-605. 12 983 Faccenna, C., Jolivet, L., Piromallo, C. and Morelli, A., 2003. Subduction and the depth of 13 984 convection in the Mediterranean mantle. Journal of Geophysical Research, 108(B2): 14 For Peer Review Only 15 985 2099. 16 986 Faccenna, C., Piromallo, C., Crespo ‐Blanc, A., Jolivet, L. and Rossetti, F., 2004. Lateral slab 17 987 deformation and the origin of the western Mediterranean arcs. Tectonics, 23(1). 18 988 Faryad, S. and Hoinkes, G., 2003. P–T gradient of Eo-Alpine metamorphism within the 19 20 989 Austroalpine basement units east of the Tauern Window (Austria). Mineralogy and 21 990 Petrology, 77(1-2): 129-159. 22 991 Forte, A.M. and Mitrovica, J.X., 2001. Deep-mantle high-viscosity flow and thermochemical 23 992 structure inferred from seismic and geodynamic data. Nature, 410(6832): 1049- 24 993 1056. 25 26 994 Fukao, Y., Obayashi, M. and Nakakuki, T., 2009. Stagnant slab: a review. Annual Review of 27 995 Earth and Planetary Sciences, 37: 19-46. 28 996 Gaina, C., Roest, W. and Müller, R., 2002. Late Cretaceous–Cenozoic deformation of 29 997 northeast Asia. Earth and Planetary Science Letters, 197(3): 273-286. 30 998 Gibbons, A., Zahirovic, S., Müller, R., Whittaker, J. and Yatheesh, V., 2015. A tectonic model 31 32 999 reconciling evidence for the collisions between India, Eurasia and intra-oceanic arcs 33 1000 of the central-eastern Tethys. Gondwana Research. 34 1001 Grand, S.P., 2002. Mantle shear–wave tomography and the fate of subducted slabs. 35 1002 Philosophical Transactions of the Royal Society of London. Series A: Mathematical, 36 1003 Physical and Engineering Sciences, 360(1800): 2475-2491. 37 38 1004 Grand, S.P., van der Hilst, R.D. and Widiyantoro, S., 1997. Global seismic tomography: A 39 1005 snapshot of convection in the Earth. GSA today, 7(4): 1-7. 40 1006 Gurnis, M., Turner, M., Zahirovic, S., DiCaprio, L., Spasojevic, S., Müller, R.D., Boyden, J., 41 1007 Seton, M., Manea, V.C. and Bower, D.J., 2012. Plate tectonic reconstructions with 42 1008 continuously closing plates. Computers & Geosciences, 38(1): 35-42. 43 44 1009 Hafkenscheid, E., Wortel, M. and Spakman, W., 2006. Subduction history of the Tethyan 45 1010 region derived from seismic tomography and tectonic reconstructions. Journal of 46 1011 Geophysical Research: Solid Earth (1978–2012), 111(B8). 47 1012 Handy, M.R., M Schmid, S., Bousquet, R., Kissling, E. and Bernoulli, D., 2010. Reconciling 48 1013 plate-tectonic reconstructions of Alpine Tethys with the geological–geophysical 49 50 1014 record of spreading and subduction in the Alps. Earth-Science Reviews, 102(3): 121- 51 1015 158. 52 1016 Hassan, R., Flament, N., Gurnis, M., Bower, D.J. and Müller, D., 2015. Provenance of plumes 53 1017 in global convection models. Geochemistry, Geophysics, Geosystems. 54 55 1018 Hosseinpour, M., Müller, R., Williams, S. and Whittaker, J., 2013. Full-fit reconstruction of 56 1019 the Labrador Sea and Baffin Bay. Solid Earth Discussions, 5(2). 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 44 of 73
1 2 3 1020 Jammes, S., Manatschal, G., Lavier, L. and Masini, E., 2009. Tectonosedimentary evolution 4 5 1021 related to extreme crustal thinning ahead of a propagating ocean: example of the 6 1022 western Pyrenees. Tectonics, 28(4). 7 1023 Jolivet, L., Faccenna, C., Goffé, B., Burov, E. and Agard, P., 2003. Subduction tectonics and 8 1024 exhumation of high-pressure metamorphic rocks in the Mediterranean orogens. 9 10 1025 American Journal of Science, 303(5): 353-409. 11 1026 Kissling, E., Schmid, S.M., Lippitsch, R., Ansorge, J. and Fügenschuh, B., 2006. Lithosphere 12 1027 structure and tectonic evolution of the Alpine arc: new evidence from high- 13 1028 resolution teleseismic tomography. Geological Society, London, Memoirs, 32(1): 14 1029 129-145.For Peer Review Only 15 16 1030 Klitgord, K. and Schouten, H., 1986. Plate kinematics of the central Atlantic. The Geology of 17 1031 North America, 1000: 351-378. 18 1032 Kneller, E.A., Johnson, C.A., Karner, G.D., Einhorn, J. and Queffelec, T.A., 2012. Inverse 19 1033 methods for modeling non-rigid plate kinematics: Application to mesozoic plate 20 1034 reconstructions of the Central Atlantic. Computers & Geosciences, 49: 217-230. 21 22 1035 Labails, C., Olivet, J.-L., Aslanian, D. and Roest, W.R., 2010. An alternative early opening 23 1036 scenario for the Central Atlantic Ocean. Earth and Planetary Science Letters, 297(3): 24 1037 355-368. 25 1038 Labails, C. and Roest, W.R., 2010. Comment on ‘Breakup of Pangaea and plate kinematics of 26 1039 the central Atlantic and Atlas regions’ by Antonio Schettino and Eugenio Turco. 27 28 1040 Geophysical Journal International, 183(1): 96-98. 29 1041 Li, C., van der Hilst, R.D., Engdahl, E.R. and Burdick, S., 2008. A new global model for P wave 30 1042 speed variations in Earth's mantle. Geochemistry, Geophysics, Geosystems, 9(5). 31 1043 Lithgow ‐Bertelloni, C. and Richards, M.A., 1998. The dynamics of Cenozoic and Mesozoic 32 33 1044 plate motions. Reviews of Geophysics, 36(1): 27-78. 34 1045 Manatschal, G. and Bernoulli, D., 1999. Architecture and tectonic evolution of nonvolcanic 35 1046 margins: Present ‐day Galicia and ancient Adria. Tectonics, 18(6): 1099-1119. 36 1047 Matthews, K.J., Müller, R.D., Wessel, P. and Whittaker, J.M., 2011. The tectonic fabric of the 37 38 1048 ocean basins. Journal of Geophysical Research: Solid Earth (1978–2012), 116(B12). 39 1049 Müller, R.D., Sdrolias, M., Gaina, C. and Roest, W.R., 2008. Age, spreading rates and 40 1050 spreading asymmetry of the world's ocean crust. Geochemistry, Geophysics, 41 1051 Geosystems, 9(4): Q04006, doi:10.1029/2007GC001743 42 1052 Müller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M., Shephard, 43 44 1053 G.E., Maloney, K., Barnett-Moore, N. and Hosseinpour, M., 2016. Ocean Basin 45 1054 Evolution and Global-Scale Plate Reorganization Events Since Pangea Breakup. 46 1055 Annual Review of Earth and Planetary Sciences, 44(1). 47 1056 Neres, M., Font, E., Miranda, J., Camps, P., Terrinha, P. and Mirão, J., 2012. Reconciling 48 1057 Cretaceous paleomagnetic and marine magnetic data for Iberia: New Iberian 49 50 1058 paleomagnetic poles. Journal of Geophysical Research: Solid Earth (1978–2012), 51 1059 117(B6). 52 1060 Nolet, G., 1987. Seismic wave propagation and seismic tomography, Seismic tomography. 53 1061 Springer, pp. 1-23. 54 1062 Norton, I., 1999. Global plate reconstruction model, technical report, ExxonMobil 55 56 1063 Exploration, Houston, Tex. 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 45 of 73 International Geology Review
1 2 3 1064 Norton, I., Lawver, L. and Gahagan, L., 2007. Plate motion of Iberia relative to Europe in the 4 5 1065 Cretaceous: Problems with the fit at M0 time, AGU Fall Meeting Abstracts, pp. 08. 6 1066 O'Neill, C., Müller, D. and Steinberger, B., 2005. On the uncertainties in hot spot 7 1067 reconstructions and the significance of moving hot spot reference frames. 8 1068 Geochemistry, Geophysics, Geosystems, 6(4). 9 10 1069 Oakey, G.N. and Chalmers, J.A., 2012. A new model for the Paleogene motion of Greenland 11 1070 relative to North America: Plate reconstructions of the Davis Strait and Nares Strait 12 1071 regions between Canada and Greenland. Journal of Geophysical Research: Solid 13 1072 Earth (1978–2012), 117(B10). 14 1073 Olivet, J., 1996.For Kinematics Peer of the Iberian Review plate. Bulletin des CentresOnly de Recherches 15 16 1074 Exploration-Production Elf Aquitaine, 20(1): 131-195. 17 1075 Piromallo, C. and Faccenna, C., 2004. How deep can we find the traces of Alpine subduction? 18 1076 Geophysical Research Letters, 31(6). 19 1077 Piromallo, C. and Morelli, A., 2003. P wave tomography of the mantle under the Alpine ‐ 20 – 21 1078 Mediterranean area. Journal of Geophysical Research: Solid Earth (1978 2012), 22 1079 108(B2). 23 1080 Robertson, A. and Shallo, M., 2000. Mesozoic–Tertiary tectonic evolution of Albania in its 24 1081 regional Eastern Mediterranean context. Tectonophysics, 316(3): 197-254. 25 1082 Rosenbaum, G. and Lister, G.S., 2005. The Western Alps from the Jurassic to Oligocene: 26 27 1083 spatio-temporal constraints and evolutionary reconstructions. Earth-Science 28 1084 Reviews, 69(3): 281-306. 29 1085 Sahabi, M., Aslanian, D. and Olivet, J.-L., 2004. A new starting point for the history of the 30 1086 central Atlantic. Comptes Rendus Geoscience, 336(12): 1041-1052. 31 1087 Schettino, A. and Scotese, C.R., 2005. Apparent polar wander paths for the major continents 32 33 1088 (200 Ma to the present day): a palaeomagnetic reference frame for global plate 34 1089 tectonic reconstructions. Geophysical Journal International, 163(2): 727-759. 35 1090 Schettino, A. and Turco, E., 2006. Plate kinematics of the Western Mediterranean region 36 1091 during the Oligocene and Early Miocene. Geophysical Journal International, 166(3): 37 38 1092 1398-1423. 39 1093 Schettino, A. and Turco, E., 2009. Breakup of Pangaea and plate kinematics of the central 40 1094 Atlantic and Atlas regions. Geophysical Journal International, 178(2): 1078-1097. 41 1095 Schettino, A. and Turco, E., 2011. Tectonic history of the western Tethys since the Late 42 1096 Triassic. Geological Society of America Bulletin, 123(1-2): 89-105. 43 44 1097 Schmid, S.M., Bernoulli, D., Fügenschuh, B., Matenco, L., Schefer, S., Schuster, R., Tischler, M. 45 1098 and Ustaszewski, K., 2008. The Alpine-Carpathian-Dinaridic orogenic system: 46 1099 correlation and evolution of tectonic units. Swiss Journal of Geosciences, 101(1): 47 1100 139-183. 48 1101 Schmid, S.M., Fügenschuh, B., Kissling, E. and Schuster, R., 2004. Tectonic map and overall 49 50 1102 architecture of the Alpine orogen. Eclogae Geologicae Helvetiae, 97(1): 93-117. 51 1103 Seidel, E., Okrusch, M., Kreuzer, H., Raschka, H. and Harre, W., 1981. Eo-alpine 52 1104 metamorphism in the uppermost unit of the Cretan nappe system—petrology and 53 1105 geochronology. Contributions to Mineralogy and Petrology, 76(3): 351-361. 54 1106 Sengör, A. and Hsü, K., 1984. The Cimmerides of eastern Asia: history of the eastern end of 55 56 1107 Paleo-Tethys. Mem. Soc. Geol. France, 147: 139-167. 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 46 of 73
1 2 3 1108 Seton, M., Flament, N., Whittaker, J., Müller, R.D., Gurnis, M. and Bower, D.J., 2015. Ridge 4 5 1109 subduction sparked reorganization of the Pacific plate ‐mantle system 60 ‐50 6 1110 million years ago. Geophysical Research Letters. 7 1111 Seton, M., Müller, R., Zahirovic, S., Gaina, C., Torsvik, T., Shephard, G., Talsma, A., Gurnis, M., 8 1112 Turner, M. and Maus, S., 2012. Global continental and ocean basin reconstructions 9 10 1113 since 200Ma. Earth-Science Reviews, 113(3): 212-270. 11 1114 Sharp, I.R. and Robertson, A.H., 2006. Tectonic-sedimentary evolution of the western 12 1115 margin of the Mesozoic Vardar Ocean: evidence from the Pelagonian and Almopias 13 1116 zones, northern Greece. Geological Society, London, Special Publications, 260(1): 14 1117 373-412.For Peer Review Only 15 16 1118 Shephard, G., Flament, N., Williams, S., Seton, M., Gurnis, M. and Müller, R., 2014. Circum ‐ 17 1119 Arctic mantle structure and long ‐wavelength topography since the Jurassic. 18 1120 Journal of Geophysical Research: Solid Earth, 119(10): 7889-7908. 19 20 1121 Shephard, G.E., Müller, R.D. and Seton, M., 2013. The tectonic evolution of the Arctic since 21 1122 Pangea breakup: Integrating constraints from surface geology and geophysics with 22 1123 mantle structure. Earth-Science Reviews, 124: 148-183. 23 1124 Shillington, D.J., Holbrook, W.S., Van Avendonk, H.J., Tucholke, B.E., Hopper, J.R., Louden, 24 1125 K.E., Larsen, H.C. and Nunes, G.T., 2006. Evidence for asymmetric nonvolcanic rifting 25 26 1126 and slow incipient oceanic accretion from seismic reflection data on the 27 1127 Newfoundland margin. Journal of Geophysical Research: Solid Earth (1978–2012), 28 1128 111(B9). 29 1129 Sibuet, J.C., Srivastava, S.P. and Spakman, W., 2004. Pyrenean orogeny and plate kinematics. 30 1130 Journal of Geophysical Research: Solid Earth (1978–2012), 109(B8). 31 32 1131 Simmons, N.A., Forte, A.M., Boschi, L. and Grand, S.P., 2010. GyPSuM: A joint tomographic 33 1132 model of mantle density and seismic wave speeds. Journal of Geophysical Research: 34 1133 Solid Earth (1978–2012), 115(B12). 35 1134 Sinclair, H., 1997. Flysch to molasse transition in peripheral foreland basins: The role of the 36 1135 passive margin versus slab breakoff. Geology, 25(12): 1123-1126. 37 38 1136 Skogseid, J., 2011. The Orphan Basin–a key to understanding the kinematic linkage 39 1137 between North and NE Atlantic Mesozoic rifting. CM 2010-Abstracts, 2. 40 1138 Souriau, A., Chevrot, S. and Olivera, C., 2008. A new tomographic image of the Pyrenean 41 1139 lithosphere from teleseismic data. Tectonophysics, 460(1): 206-214. 42 1140 Spakman, W., van der Lee, S. and van der Hilst, R., 1993. Travel-time tomography of the 43 44 1141 European-Mediterranean mantle down to 1400 km. Physics of the Earth and 45 1142 Planetary Interiors, 79(1): 3-74. 46 1143 Spakman, W. and Wortel, R., 2004. A tomographic view on western Mediterranean 47 1144 geodynamics, The TRANSMED atlas. The Mediterranean region from crust to mantle. 48 1145 Springer, pp. 31-52. 49 50 1146 Speranza, F., Minelli, L., Pignatelli, A. and Chiappini, M., 2012. The Ionian Sea: the oldest in 51 1147 situ ocean fragment of the world? Journal of Geophysical Research: Solid Earth 52 1148 (1978–2012), 117(B12). 53 1149 Srivastava, S. and Roest, W., 1989. Seafloor spreading history II–IV. East Coast Basin Atlas 54 55 1150 Series: Labrador Sea: 100-109. 56 1151 Srivastava, S. and Roest, W., 1999. Extent of oceanic crust in the Labrador Sea< sup> 57 1152 1. Marine and Petroleum Geology, 16(1): 65-84. 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 47 of 73 International Geology Review
1 2 3 1153 Srivastava, S., Sibuet, J.-C., Cande, S., Roest, W. and Reid, I.D., 2000. Magnetic evidence for 4 5 1154 slow seafloor spreading during the formation of the Newfoundland and Iberian 6 1155 margins. Earth and Planetary Science Letters, 182(1): 61-76. 7 1156 Srivastava, S. and Verhoef, J., 1992. Evolution of Mesozoic sedimentary basins around the 8 1157 North Central Atlantic: a preliminary plate kinematic solution. Geological Society, 9 10 1158 London, Special Publications, 62(1): 397-420. 11 1159 Stampfli, G., Borel, G., Marchant, R. and Mosar, J., 2002. Western Alps geological constraints 12 1160 on western Tethyan reconstructions. Journal of the Virtual Explorer, 8: 77-106. 13 1161 Stampfli, G., Mosar, J., Marquer, D., Marchant, R., Baudin, T. and Borel, G., 1998. Subduction 14 1162 and obductionFor processes Peer in the SwissReview Alps. Tectonop hysics,Only 296(1): 159-204. 15 16 1163 Stampfli, G.M. and Borel, G.D., 2004. The TRANSMED transects in space and time: 17 1164 constraints on the paleotectonic evolution of the Mediterranean domain, The 18 1165 TRANSMED Atlas. The Mediterranean region from crust to mantle. Springer, pp. 53- 19 1166 80. 20 1167 Stüwe, K. and Schuster, R., 2010. Initiation of subduction in the Alps: Continent or ocean? 21 22 1168 Geology, 38(2): 175-178. 23 1169 Sutra, E., Manatschal, G., Mohn, G. and Unternehr, P., 2013. Quantification and restoration of 24 1170 extensional deformation along the Western Iberia and Newfoundland rifted 25 1171 margins. Geochemistry, Geophysics, Geosystems, 14(8): 2575-2597. 26 1172 Torsvik, T.H., Muller, R.D., Van der Voo, R., Steinberger, B. and Gaina, C., 2008a. Global plate 27 28 1173 motion frames: Toward a unified model. Reviews of Geophysics, 46(3): G3004- 29 1174 G3004. 30 1175 Torsvik, T.H., Müller, R.D., Van der Voo, R., Steinberger, B. and Gaina, C., 2008b. Global plate 31 1176 motion frames: toward a unified model. Reviews of Geophysics, 46(3). 32 1177 Tucholke, B.E. and Sibuet, J.-C., 2007. Leg 210 synthesis: Tectonic, magmatic, and 33 34 1178 sedimentary evolution of the Newfoundland-Iberia rift, Proceedings of the Ocean 35 1179 Drilling Program, scientific results, pp. 1-56. 36 1180 Van der Meer, D.G., Spakman, W., van Hinsbergen, D.J., Amaru, M.L. and Torsvik, T.H., 2010. 37 1181 Towards absolute plate motions constrained by lower-mantle slab remnants. Nature 38 1182 Geoscience, 3(1): 36-40. 39 40 1183 Van der Voo, R., Spakman, W. and Bijwaard, H., 1999. Tethyan subducted slabs under India. 41 1184 Earth and Planetary Science Letters, 171(1): 7-20. 42 1185 van Hinsbergen, D.J.J., Hafkenscheid, E., Spakman, W., Meulenkamp, J. and Wortel, R., 2005. 43 1186 Nappe stacking resulting from subduction of oceanic and continental lithosphere 44 1187 below Greece. Geology, 33(4): 325-328. 45 46 1188 Vissers, R.L., van Hinsbergen, D.J., Meijer, P.T. and Piccardo, G.B., 2013. Kinematics of 47 1189 Jurassic ultra-slow spreading in the Piemonte Ligurian ocean. Earth and Planetary 48 1190 Science Letters, 380: 138-150. 49 1191 Whitmarsh, R.B. and Miles, P.R., 1995. Models of the development of the West Iberia rifted 50 ° ′ ‐ 51 1192 continental margin at 40 30 N deduced from surface and deep tow magnetic 52 1193 anomalies. Journal of Geophysical Research: Solid Earth (1978 –2012), 100(B3): 53 1194 3789-3806. 54 1195 Widiyantoro, S., Van Der Hilst, R. and Wenzel, F., 2004. Deformation of the Aegean slab in 55 56 1196 the mantle transition zone. International Journal of Tomography & Statistics D, 4: 1- 57 1197 14. 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 48 of 73
1 2 3 ‐ 4 1198 Williams, S.E., Whittaker, J.M. and Müller, R.D., 2011. Full fit, palinspastic reconstruction 5 1199 of the conjugate Australian ‐Antarctic margins. Tectonics, 30(6). 6 1200 Wortel, M. and Spakman, W., 2000. Subduction and slab detachment in the Mediterranean- 7 1201 Carpathian region. Science, 290(5498): 1910-1917. 8 9 1202 Yoshioka, S. and Naganoda, A., 2010. Effects of trench migration on fall of stagnant slabs 10 1203 into the lower mantle. Physics of the Earth and Planetary Interiors, 183(1): 321-329. 11 1204 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 49 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 Fig. 1. Schematic illustration of the Western Tethys region in the framework of surrounding 19 20 plates. The red arrows show the relative motion between plates as defined in the rotation file. 21 The black arrows are inferred motions. 22 23 24 25 26 27 28
29 30 31 32 33 34 35 36 37 38
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 50 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 Fig. 2. Location of various geographic, geological, tectonic and geophysical features in the 21 Western Tethys and surrounding areas presented in two different phases. A. Early Cretaceous 22 time showing the early stages of the Alpine compressional phase. B. Collision and formation 23 of orogenic belts are shown in Late Miocene time. Subduction zones (black), mid ocean 24 ridges (red) and magnetic anomalies (grey) are reconstructed using the tectonic model 1a 25 [modified from Schettino and Turco (2011)]. Model 2 [based on Seton et al. (2012)] has been 26 shown in green wherever it was different from model 1a. The purple shows the basins and 27 28 subduction zones according to model 3 [based on Stampfli et al. (2002)]. The Pyrenean 29 subduction zone (orange) is based on the model 4 (Sibuet et al., 2004). Models 2, 3 and 4 are 30 schematic. Blue, gold and pale blue represent oceanic and continental crust and rift zones 31 respectively. LS: Labrador Sea; NEAO: Northeast Atlantic Ocean; NAO: North Atlantic 32 Ocean; NCAO: North Central Atlantic ocean; CAO: Central Atlantic Ocean; ECMA: East 33 Coast magnetic Anomaly; BSMA: Black Spur Magnetic Anomaly; WACMA: West African 34 Coast Magnetic Anomaly; ABSMA: African Black Spur Magnetic Anomaly; BB: Bay of 35 Biscay; Py: Pyrenees; VaO: Valais Ocean; PO: Piemont Ocean; LO: Ligurian Ocean; Ar: 36 Atlas rift; IS: Ionian Sea; EMb: Eastern Mediterranean basin; PiO: Pindos Ocean; EVO: 37 38 Eastern Vardar Ocean; Ly: Lycian; Vo: Vardar Ophiolites; Iz An: Izmir Ancara; NT: 39 Neotethys; WNT: Western Neotethys; Al: Alboran; Ao: Atlas orogenic belt; Li Pr: Liguro 40 Provençal; Ty: Tyrrhenian Sea; Pe: Pennines; Ap: Apennines; Al: Alps; Ca: Carpathians; 41 Aus: Austroalpine; Di: Dinarides; Ha: Hellenic arc; Sa Si: Sanandaj Sirjan; Za: Zagreb; Be: 42 Belgrade; Mi: Milano. Refer to text for further explanations of each domain/ feature. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 51 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Fig. 3. Flowlines of different tectonic models suggested for opening of various branches of 36 the Atlantic Ocean are plotted on the magnetic anomaly map of Maus et al. (2009). Boxes 37 labelled A and B are the conjugate margins of the North Atlantic; C and D are the conjugate 38 margins of North Central Atlantic; and E and F are the conjugate margins of Central Atlantics 39 40 (see Fig. 4). Black lines show the flowlines of (Schettino and Turco, 2009); Green (Seton et 41 al., 2012): [(203 170Ma):(Labails et al., 2010), (170 0.0Ma): (Müller et al., 1997)]; Dark 42 violet (Vissers et al., 2013); Pale violet (Scenario A. Kneller et al. (2012)); Red (Srivastava 43 and Verhoef, 1992); Orange (Scenario B. Kneller et al. (2012)); Blue This study. Thick red 44 and dark blue lines represent mid ocean ridges and fracture zones respectively. ECMA: East 45 Coast Magnetic Anomaly; BSMA: Black Spur Magnetic Anomaly; WACMA: West Africa 46 Coastal Magnetic Anomaly; ABSMA: African Black Spur Magnetic Anomaly. Refer to text 47 for more information about each model. 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 52 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Fig. 4. Late Triassic to Early Mid Jurassic opening of Atlantic Ocean plotted on the marine 51 gravity map of Sandwell et al. (2014) . The segments A to F were marked on Fig. 3. The 52 flowlines are the same as Fig. 3 and divided to 5 Myr increments to show the direction and 53 rate of opening. The time for full fit reconstruction and major changes for each model are 54 also depicted. 4A and B North Atlantic, North America and Eurasian margins, respectively. 55 56 4C and D North Central Atlantic, Newfoundland and Iberian margins, respectively. 4E and 57 F Central Atlantic, North American and African margins, respectively. Key is the same as 58 for Fig. 3. 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 53 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 For Peer Review Only 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] 47 48 49 50 51 52 53 54 55 56 57 58 59 60 International Geology Review Page 54 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 Fig. 6: Iberian kinematics relative to Africa and Eurasia from the beginning of rifting in 28 North Central Atlantic Ocean in Early Jurassic until the Mid Late Cretaceous and onset of the 29 compressional phase between Iberia and Eurasia, as proposed by this study. Iberia moves 30 relative to Eurasia along the Gibraltar fault (GF) from 130 to 100 Ma and North Pyrenean 31 Fault (NPF) from 130 until 83 Ma when it starts its independent rotation relative to Eurasia. 32 The maximum width of the Pyrenean rift and Valais Ocean in this reconstruction are 200 and 33 34 400 km, respectively. The amount of Iberian rotation is calculated as 25°. North America is 35 fixed in this reconstruction. 36 37 38 39 40 41 42 43 44 45 Fig. 5: Reconstruction of various tectonic models proposed for early opening of Central, 46 North Central and North Atlantic and the implication of each model on plate 47 configuration in adjacent regions (e.g. Western Tethys and Gulf of Mexico; Refer to text 48 for more details). We plotted the motion of North America (blue)-Eurasia (green)-Iberia 49 (red) and Tisza (purple) relative to fix Africa (grey) for tree time steps using four 50 different reconstructions. The plate outlines of Model 1a (A-This study) are plotted on 51 the other reconstruction models for better comparison. 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 55 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Fig. 7. Tectonic reconstructions of Model 1a for the evolution of Western Tethys and 33 adjacent continental and oceanic plates from 200 to 20 Ma, with Africa remaining fixed for 34 entire time of reconstruction. The early Central Atlantic kinematics from 240 to 154 Ma is 35 reconstructed using our proposed model for this area. For Iberia Newfoundland relative 36 motion from 200 to 120 Ma we use the rotation derived from our alternative Iberian 37 reconstruction in this study. The opening of the Central North Atlantic Ocean after this time 38 was reconstructed with poles of rotation from Schettino and Scotese (2005). The opening of 39 40 the North Atlantic was reconstructed using Seton et al. (2012) poles of rotation from 155 Ma 41 till present time. The spreading and closure history of the Neotethys basin is based on Seton 42 et al. (2012) model from 200 to 100 Ma. We implemented a short lived intra oceanic 43 subduction in the Neotethys Ocean from 100 to 75 Ma to account for western Neotethys 44 subduction. Tectonic Model 1 uses the poles of rotation of Schettino and Turco (2006) and 45 Schettino and Turco (2011) for relative motion of small microplates inside western Tethys 46 region. NT: Neotethys; WVO: western Vardar Ocean; EVO: eastern Vardar Ocean; BB: Bay 47 of Biscay; Ca: Calabrian subduction. EUR: Eurasia; NAM: North America; AFR: Africa; 48 IBR: Iberia; NW AFR: Northwest Africa; NE AFR: Northeast Africa; ADR: Adria; CAR 49 50 TAU: Carpathian Taurides; KIR: Kir ehir; DIN: Dinarides; PAN: Panonian basin. 51 52 53 54 55 56 57
58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 56 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 Fig. 8 . Comparing two absolute reference frames in three different time steps for tectonic 28 Models 1a and 1b. Red lines with triangles denote the subduction zones in the Hotspot True 29 Polar Wonder reference frame (model 1a). Black lines with triangles denote those of the 30 subduction reference frame of Van der Meer et al. (2010) (model 1b). 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 57 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Fig. 9. Vertical cross section along line I II through the reconstructed location of the eastern 39 Vardar Ocean and Eastern Mediterranean domain. The vertical profiles show comparison 40 with two P waves and one S wave tomography models of GrandS (Grand, 2002), MITP 08 41 (Li et al., 2008) and GypsumP (Simmons et al., 2010) , as labelled on figure. A: Interpreted 42 Hellenic slab. B: Interpreted Eastern Vardar and Hellenic slab. C: Interpreted Triassic and 43 44 Jurassic Kure, Izmir Ancara and Vardar slabs. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 58 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Fig. 10. Vertical cross section along line I II through the reconstructed location of the 34 Cenozoic Ligurian (Betic) Calabrian and Hellenic subductions. The vertical profiles show 35 comparison with two P waves and one S wave tomography models of GrandS (Grand, 36 2002), MITP 08 (Li et al., 2008) and GypsumP(Simmons et al., 2010) , as labelled on figure. 37 38 A: Interpreted Ligurian slab. B: Interpreted Calabrian slab. C: Interpreted Hellenic slab. 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 59 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Fig. 11. Vertical cross section along line I II through Pyrenees and reconstructed location of 40 southern Ligurian subductions. The vertical profiles show comparison with two P waves and 41 one S wave tomography models of GrandS (Grand, 2002), MITP 08 (Li et al., 2008) and 42 GypsumP(Simmons et al., 2010) , as labelled on figure. A and B: Pyrenees slabs interpreted 43 by Sibuet et al. (2004). C: Interpreted Ligurian (Betic) slab. 44 45 46 47 48 49 50 51 52 53
54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 60 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Fig. 12. Vertical cross section along line I II through reconstructed location of Alpine 36 (Piemont) subduction. The vertical profiles show comparison with two P waves and one S 37 wave tomography models of GrandS (Grand, 2002), MITP 08 (Li et al., 2008) and 38 GypsumP(Simmons et al., 2010) , as labelled on figure. A: Interpreted Calabrian slab. B: 39 40 Interpreted Cenozoic Alpine slab. C and D: Interpreted Mesozoic Alpine (pre Alpine) slab. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 61 of 73 International Geology Review
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Fig. 13. Vertical cross section along line I II through reconstructed location of western 34 Vardar, Western Neotethys and Neotethys subductions. The vertical profiles show 35 comparison with two P waves and one S wave tomography models of GrandS (Grand, 36 2002), MITP 08 (Li et al., 2008) and GypsumP(Simmons et al., 2010) , as labelled on figure. 37 A: Interpreted Vardar slab. B: Interpreted Mesozoic Alpine (pre Alpine) slab. C: Interpreted 38 39 Western Neotethys slab. 40 41 42 43 44 45 46
47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 62 of 73
1 2 3 4 5 6 7 8 9 10 11 12 13 14 For Peer Review Only 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Fig. 14. Illustration of the paleo location of subduction zones for four reconstructed times 46 using two different absolute reference frames. Model 1a (magenta) and model 2 (green) are 47 plotted using hotspot True Polar Wander reference frame. Model 1b (black) is representing 48 49 the same plate boundaries plotted in the subduction reference frame. Associated horizontal 50 tomographic slices of model MITP 08 underlay the reconstruction model for each time step. 51 We used sinking rate of 1.2 cm/yr to constrain the depth age relationship for each time step. 52 The tomographic high velocity materials associated with each subduction zone are also 53 marked for further resolution. Grey lines represent present day coastlines. 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 63 of 73 International Geology Review
1 2 3 Table 1. Summary of the rifting and seafloor spreading history in the Atlantic domains and western 4 Tethys basins. see table 2 for abbreviations and references. 5 6 Rifting and Seafloor spreading Basin/ Tectonic 7 event Onset of 8 rifting Onset of SFS End of SFS Comments 9 ECMA and BSMA considered either as true 3 4, 5 , 2 3,5 MAs or Coastal anomalies . The possibility 10 240* ; 230 ; 3 1, 2, 3 4 5 200 and 190 of ridge jump being considered . A major Central Atlantic 203 ; 190 ; 3, 4, 5, 8 N/A 11 6,7,16 Ma change in direction and rate of spreading 12 175 Ma between 185-170Ma has been reported 2, 3, 4, 5 13 14 230 2 and 200 8, 9 185 2 and 166 9, Gulf of MexicoFor Peer10, 11 ReviewN/A Only N/A 15 Ma Ma 16 Atlas rift 200 2, 12, 13, 14 Ma N/A Maximum width 113 12 km 17 203 8; 200 17, *; Rifting and spreading are diachronous 3 18 180 ; 156 ; 8, 19 1, 3, (propagates northward). The final 18 26 145 ; 132 16 North Central 140 Ma . 18, 25 17, 26, separation is dated as late as 95 or 84 19 ; 120 N/A 26 Atlantic Break up time: 62 21, 26 Ma . Times for major tectonic events were 20 175 7, 8,16 ; 146 1, ; 112 Ma derived from magnetic 8, 16, 18, 19, 22 , Seismic 17, 20 23, 24, 26 26 16, 21 21 Ma , Drilling and strathigraphy data. , 1, 27, 22 North Atlantic ~160-155* 29, 37 33 110 Ma N/A N/A 23 Ma 24 33 1 32 1 *, Labrador Sea and ~150 ; 130 ; 90 ; 79 ; 60 19, 33 25 Baffin Bay 120*, 33, 36 Ma 30, 33 Ma 34 Ma N/A 26 , 1, 28, ~160-155* 1, 28, 31 27 Northeast Atlantic 31 Ma 56 Ma N/A N/A
28 61 59 59 61 163 ; 156 ; 124 ; 120 ; 3, 40 , 61 59 29 Bay of Biscay 120*, 3, 40 Ma 100*, 3, 40 Ma 83 Ma Width of the rift zone: 200* , 400 km 30 NPF 155- 120 40 ; 140-120 61 ; 147-83*, 3 Ma Total Motion: 300*; 520 46 km 31 Pyrenees rift 120*, 46, 59 Ma N/A Width: 200*; 600 59 , 32 8, 39, 40, 41 , 68 40 , 3, 42 33 Pindos Ocean 240 Ma Ma 174 Ma No Pindos Ocean* 34 200*, 1,3, 45 ; Vardar Ocean 240*, 3; 200 68 ; 174 8, 39, 40, 41 Ma 170 42 ; 155 40, N/A 35 63 ; 112 39 Ma 36 3 *, 3, 40 200 Ma; 37 230 ; 152- 40, 47 50 Jurassic ; Evidences for early rifting: Triassic 38 270*, 40, 47, 50 ; 145 Ma Early volcanism, normal faulting, and Oceanic Ionian Sea 230*, 3,40, 47, 49 ; Resumed Cretaceous 45 ; sedimentation in southeasters turkey 49 . 39 48 42 spreading: 118- 200 ; 100 Ma 46 Late Isostatic equilibrium, seismic velocities, 94 and 67-35 46, 47 40 Cretaceous elastic thickness 41 Ma 48 ; 35 46 Ma 42 270 40, 47 ; 230*, 43 Eastern 3, 49 ; 200 48 ; 100*, 3 Ma 67 3; 41 48 Ma N/A Mediterranean 42 44 100 Ma 40 , 3 , 3, 40, 45 250 ; 230* ; 180-170* , 3, 41, 46 Ligurian Ocean 220 46 Ma 46 Ma 130Ma* Width of stretched continental crust: 100- 46 150 40 to >300km 54 . Width of Liguro-Piemont , 3, 40 46 55 40 , 3 , 3, 40, 40 , 46 Ocean: 500* , 800 , 1500 km . 47 250 ; 230* ; 180-170* 155 ; 130* ; 40 46 48 Piemont Ocean 220 46 ; 200 60 Ma 46, 60 Ma 120 3 Ma Spreading rate: 0.6 , 2cm/yr . 49 156 59 ; 146 3, 46 ; 50 130 40 ; 130 3, 46 ; 120 40 ; 95-83 40, 46, 59, 60 Max with: 50 56 , 100 46 , 200 57 , 250-350 40 , Valais Ocean 60 40 60 59 , 3 51 127Ma ; 120 112 Ma Ma 350 and 430* Ma Ma 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] International Geology Review Page 64 of 73
1 2 3 Table 2 . Summary of the compressional events in the Western Tethys basins 4 5 Basin/ tectonic Closure, Subduction and collision Polarity of subduction 6 event 7 Closure time: 33 Ma 64, 65, 66. Amount of shortening between 13- Atlas Orogeny 1, 67 8 36km 9 Pindos Ocean Subduction: 86 40, 41 , 160 68 and 67 68 Ma 10 N-dipping subduction on Eurasian margin: 230-180 3 or 160 Ma 1. 11 E-SE-dipping intra-oceanic Subduction: 180 3, 174 Ma 40 Collision: 12 Vardar Ocean ~145 Ma 3, 39, 41, 42, 46, 68 . Obduction: 130-113 40, 41, 42 , 120 Ma 3. Re- N/A 13 initiation of N-NE-dipping subduction on Eurasian margin: 120 3, 40, 41 14 For Peer Review113 Ma . Only 15 Ionian Sea 30Ma 46 N/A 16 56 3, 83 48 . New subduction zone 3 versus southward migrating Mediterranean 17 Vardar arc (several HP methamorphic belts from North to N/A basin 51, 52 18 south, Seismic tomography) 19 Intra-Tauride Subduction: 75-67 3 N/A
20 South Ligurian 3, 40, 69 3, 40, 46 Subduction: 33 , Polarity: N 21 Ocean 3, 22 3, 45 40 69 Polarity: Always NW West Ligurian Subduction: 132 , 100 ; 84Ma . Location: Under Iberian 69 23 3 46 , SE and then NW at Ocean margin , Intra oceanic ; 70 46 24 42-49 or 35Ma 25 Subduction: 120 3,40 , 100 46 , 84 69 Ma; Obduction: 45 60 ; 40 3; 35 46 ; 26 Evidences of obduction: HP and UHP meta. Of ophiolites 60 . Piemont Ocean 3 46, 60, N/A Polarity: N-ward under Eurasia ; S-ward under Briançonnais 27 69 28 46 40 3, 59 60 Subduction: 100 , 95-83 ; 83 ; 56 ; evidences for 56Ma 3 40, 46 29 Polarity: N , S : N- subduction: Geochronology, Ophiolit dating. Obduction: 45- Valais Ocean 60 3 ward migrating for- 30 35 ; 40-33 ; Evidences: Radiometric dating of UP and UHP 60 deep basins 31 rocks in ICM 32 Eo-Alpine continental 40 46 46 33 ~130 to 95 , 140 to 84 Polarity: SE-E 34 subduction and orogeny 35 Pyrenees 36 subduction and 84-35 46 , 118-100 Ma 59 and 85-25 Ma 59 37 Orogeny Bay of Biscay 59 38 83-35 , 39 subduction 40 41 42 43 44 *This study; 1(Seton et al., 2012) ; 2(Kneller et al., 2012) ; 3(Schettino and Turco, 2009) ; 4(Labails et al., 2010) ; 5 6 7 8 9 45 (Sahabi et al., 2004); (Dunbar and Sawyer, 1989); (Klitgord and Schouten, 1986); (Vissers et al., 2013); (Keppie and Keppie, 2012); 10 (Pindell and Kennan, 2001); 11 (Marton and Buffler, 1999); 12 (Pindell, 1994); 13 (Beauchamp et 46 al., 1999); 14 (Beauchamp et al., 1996); 15 (De Lamotte et al., 2009); 16 (Srivastava and Verhoef, 1992); 17 (Skogseid, 47 2011); 18 (Srivastava et al., 1990); 19 (Srivastava et al., 2000); 20 (Srivastava and Tapscott, 1986); 21 (Tucholke and 48 Sibuet, 2007); 22 (Sibuet et al., 2007); 23 (Russell and Whitmarsh, 2003); 24 (Sutra et al., 2013); 25 (Schettino and 49 Scotese, 2005); 26 (Bronner et al., 2011); 27 (Welford et al., 2012); 28 (Faleide et al., 2008); 29 (Gaina et al., 2002); 50 30 (Oakey and Chalmers, 2012); 31 (Skogseid et al., 2000); 32 (Srivastava and Roest, 1989); 33 (Hosseinpour et al., 2013); 34 35 36 51 (Catalano et al., 2001) (Cole and Peachey, 1999); (Chian and Louden, 1994); (Srivastava and Roest, 1999); 37 (Müller et al., 1997); 38 (Stampfli et al., 1991); 39 (Stampfli et al., 1998); 40 (Stampfli et al., 2002); 41 (Stampfli and 52 Borel, 2004); 42 (Schmid et al., 2008); 43 (Tekin and Göncüoglu, 2007); 44 (Göncüoglu et al., 2006); 45 (Sengör and Hsü, 53 1984); 46 (Handy et al., 2010); 47 (Müller et al., 2008); 48 (De Lamotte et al., 2011); 49 (Sengör and Yilmaz, 1981); 54 50 (Catalano et al., 2001); 51 (Faccenna et al., 2003); 52 (Jolivet et al., 2003); 53 (Robertson et al., 2009); 54 (Lavier and 55 Manatschal, 2006); 55 (Savostin et al., 1986); 56 (Schmid et al., 2004); 57 (Rosenbaum et al., 2002); 58 (Treves, 1984); 56 59 (Sibuet et al., 2004); 60 (Rosenbaum and Lister, 2005); 61 (Malod and Mauffret, 1990); 62 (Olivet, 1996); 63 (van Hinsbergen et al., 2005); 64 (Beauchamp et al., 1999); 65 (Beauchamp et al., 1996); 66 (De Lamotte et al., 2009); 67 (Ruiz- 57 Martínez et al., 2012); 68 (Channell and Kozur, 1997); 69 (Wortmann et al., 2001); 70 (Lustrino et al., 2009) 58 MA: Magnetic Anomaly; NPF: North Pyrenean Fault; ICM: Internal Crystalline Massif 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected] Page 65 of 73 International Geology Review
1 2 3 4 5 Table 3 . Tectonic scenarios suggested for the evolution of different oceanic and continental basins in 6 the Western Tethys region. Model 1a and 1b are reconstructed in this study. We compared these 7 models to alternative scenarios of 2 to 4 in relevant domains. 8 9 Model Description 10 11 Western Tethys: Schettino and Turco (2011) and Schettino and Turco (2006) ; western Neotethys : 200 12 13 1a to 100Ma: Seton et al. (2012)-100 Ma to present day: De Lamotte et al. (2011); Atlantic Kinematics: 14 ForThis study Peer as described in sectionReview 2.1.4; Absolute reference Onlyframe: Hybrid-TPW 15 16 Tectonic reconstruction: The same as Model 1; Absolute reference frame: subduction reference frame 17 1b 18 of Van der Meer et al. (2010) 19 20 2 As described in Seton et al. (2012) ; see Table 1 and 2 21 3 As described in Stampfli et al. (2002) and Stampfli and Borel (2004) ; see Table 1 and 2 22 23 4 As described in Sibuet et al. (2004); see Table 1 and 2 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 URL: https://mc.manuscriptcentral.com/tigr E-mail: [email protected]