Origin and time evolution of subduction polarity reversal from plate kinematics of Southeast Asia

Christoph von Hagke1,2*, Mélody Philippon3, Jean-Philippe Avouac2, and Michael Gurnis4 1Institute of Structural Geology, Tectonics, and Geomechanics, RWTH Aachen, Lochnerstrasse 4-20, 52056 Aachen, Germany 2Division of Geological and Planetary Sciences, California Institute of Technology, 1200 East California Boulevard, Pasadena, California 91125, USA 3Géosciences Montpellier UMR CNRS 5243, Université des Antilles, Campus de Fouillole, 97100 Point à Pitre, France 4Seismological Laboratory, California Institute of Technology, 1200 East California Boulevard, Pasadena 91125, California, USA

ABSTRACT for polarity reversals, and may provide an alter- We present a regional model of plate geometry and kinematics of Southeast Asia since the native hypothesis on the structuring of the pre- Late Cretaceous, embedded in a global plate model. The model involves subduction polar- collisional Eurasian passive margin. This is ity reversals and sheds new light on the origin of the subduction polarity reversal currently of particular interest in the light of a polarity observed in Taiwan. We show that this subduction zone reversal is inherited from subduction reversal possibly occurring in Taiwan at pres- of the proto–South China Sea plate and owes its current location to triple junction migration ent (Suppe, 1984; Ustaszewski et al., 2012) (see and slab rollback. This analysis sheds new light on the plate tectonic context of the Taiwan the GSA Data Repository1). Here we present orogeny and questions the hypothesis that northern Taiwan can be considered an older, more a plate tectonic reconstruction of the Taiwan mature equivalent of southern Taiwan. area since the Late Cretaceous, embedded in a global plate model. INTRODUCTION To understand better how subduction polar- Subduction zones are major drivers of plate ity reversals originate and evolve, we investi- POLARITY REVERSALS IN TIME AND motions (e.g., Conrad and Lithgow-Bertelloni, gate the plate tectonic evolution of Southeast SPACE 2002; Stadler et al., 2010) and govern much of Asia, focusing on the Taiwan orogeny, where We reconstruct the Philippine Sea plate start- Earth’s topography; they influence the architec- two active subduction zones of opposite ver- ing from previous models (Hall, 2002; Seton et ture of mountain belts and location of volcanic gence meet (Fig. 1B). Subduction polarity is al., 2012; Zahirovic et al., 2013), using GPlates arcs (Dewey et al., 1973). Seismic tomography known to have reversed in time there as well: software (www.gplates.org), which is based on and geologic evidence suggest that subduction the eastern Eurasian margin was the upper plate data from spreading. It allows calculation and zones change polarity as the overriding and to the westward-subducting Pacific plate in the real-time visualization of global plate tectonic subducting plates switch their roles either in Late Cretaceous, whereas it currently forms the time (at a given location) or space, along the lower plate and subducts eastward beneath the 1 GSA Data Repository item 2016213, the recon- struction approach, Figure DR1 (comparing different strike of a convergent plate boundary (at a given Philippine Sea plate (e.g., Hall, 2002). The kine- models) and Movie DR2 (new reconstructions), is time). Polarity reversals have been recognized in matics of this reversal raise the possibility of available online at www.geosociety.org/pubs/ft2016. the geological record at a number of locations gaining insights in the mechanisms responsible htm, or on request from [email protected]. (Fig. 1) and may have happened particularly in response to collisions of volcanic arcs with ocean-continent subduction zones or with sub- ducting ridges (Brown et al., 2011). It has been proposed that polarity reversals can be related to spontaneous flipping along a transform fault, propagating slab tear and breakoff, collision of two subduction zones, or propagating slab tear and rollback (e.g., Brown et al., 2011; Clift et al., 2003). These concepts are mostly derived from reasoning based on two-dimensional lith- osphere-scale cross sections or from present-day plate configurations. Here we assess subduction zone reversal in the context of a time-evolving plate tectonic model, taking global plate move- ments into account, as regional reconstructions tend to push inconsistencies outside the area Figure 1. A: Compilation of areas where subduction polarity reversals have been conjectured of interest in regional reconstructions (as noted in ancient orogens. B: Geodynamic map of Southeast Asia showing Taiwan at the junction of by Hall, 2002). two oppositely dipping subduction zones. Taiwan is the result of collision between the Phil- ippine Sea plate and the Eurasian plate. C: Structural map of Taiwan and plate configuration. Coastal Ranges are the part of the Luzon arc that has been accreted to Eurasia. Topographic maps are from GeoMapApp (http://www.geomapapp.org/; see Ryan et al., 2009). *E-mail: [email protected]

GEOLOGY, August 2016; v. 44; no. 8; p. 659–662 | Data Repository item 2016213 | doi:10.1130/G37821.1 | Published online 8 July 2016 ©GEOLOGY 2016 Geological | Volume Society 44 | ofNumber America. 8 For| www.gsapubs.org permission to copy, contact [email protected]. 659 reconstructions (Boyden et al., 2011). Here we use rigid plate motions, which introduce errors associated with plate deformation; however, it is an appropriate tool to trace polarity rever- sals through time. Our model can be seen as a locally refined version of the model of Seton et al. (2012) adjusted to fit geological constraints on the evolution of the Taiwan area (see the Data Repository). We start our reconstructions at a time when the oceanic crust of the now extinct Izanagi plate (a conjugate to the Pacific plate) was subducted westward beneath Eurasia, i.e., opposite to present-day subduction polarity south of Taiwan. Westward subduction beneath Eurasia occurred during the entire Mesozoic, and during the late Mesozoic the Eurasian margin under- went widespread diffuse continental extension, putatively driven by eastward rollback of the Izanagi slab (Zhou and Li, 2000). Widespread tectonic subsidence reached as far east as the East China Sea at 65 Ma (Yang et al., 2004), while also affecting the Taiwan region (Lin et al., 2003). Extension in south and east China resulted in opening of the proto–South China Sea, and subsequent seafloor spreading (Zahi- rovic et al., 2013). The extent of this oceanic plate is unconstrained; however, this does not influence the large-scale plate tectonic frame- work. During the early Cenozoic, Southeast Asia extruded eastward, due to collision of the Indian plate and the Eurasian margin (e.g., Tap- ponnier et al., 1982), extension along the Eur- asian margin (Houseman and England, 1993), or a combination of both (Hall, 2002; Morley, 2012). This eastward extrusion modified the kinematics of the eastern Eurasian margin, and coincided with widespread extension (Jolivet et al., 1994). At 48 Ma, the Philippine Sea plate located east of the Borneo subduction zone and south of the proto–South China Sea (Fig. 2A) started to rotate and moved northward. Due to this northward movement, the young and buoy- ant Philippine Sea plate reached the southern edge of the relatively old proto–South China Sea ca. 48 Ma (Fig. 2A). The juxtaposition of relatively old crust of the proto–South China Sea against relatively young crust of the Philippine Sea (Fig. 2A) eventually resulted in subduction initiation, possibly at the location of a preex- isting strike-slip fault. The proto–South China Sea started to subduct eastward (Fig. 2B). Such Figure 2. Tectonic reconstructions of Southeast Asia. Subduction polarity reversal induced major tectonic events at regional scale related by the motion (northward migration combined with rotation) of the Philippine Sea plate (PSP) to changes in plate motions are well-known in three stages. A: 48 Ma. The Philippine Sea plate is the upper plate of the Borneo subduc- candidates for subduction initiation or reversal tion zone. At its southern tip it is juxtaposed against young, buoyant crust of the Philippine (Gurnis et al., 2004). Sea plate. Initially connected by a strike-slip zone, subduction initiates in a northwestward direction. B: 37 Ma. The proto–South China Sea is consumed below the Philippine Sea plate. During consumption of the proto–South This motion, together with rotation of the plate, causes migration of the polarity reversal. At China Sea, the Philippine Sea plate rotated 35 Ma, the South China Sea starts spreading. C: 20 Ma. Rotation of the Philippine Sea plate clockwise, as suggested by the abandonment has stopped, and polarity reversal is no longer moving northward. Spreading between the of the east-west–oriented spreading ridge in Kyushu Ridge and the West Mariana Ridge is accommodated along a strike-slip fault. D: 17 favor of a NNE-SSW–trending ridge (Des- Ma. Subduction initiates along the strike-slip zone, and the Okinawa Trough extends west- ward. E: 6.5 Ma. Collision starts in central Taiwan; the Pacific plate rolls back rapidly. F: At champs and Lallemand, 2002). From 48 to 37 present day, the rollback of the subduction has reached the Eurasian margin, and interacts Ma, the Philippine Sea plate continued moving with mountain building in Taiwan.

660 www.gsapubs.org | Volume 44 | Number 8 | GEOLOGY northward along the Eurasian margin, propa- roughly coinciding with formation of the Luzon gating the polarity reversal and consuming the arc as an intraoceanic arc along the western proto–South China Sea (Figs. 2A and 2B). At boundary of the Philippine Sea plate (Sibuet 37 Ma, the proto–South China Sea had been and Hsu, 2004). almost entirely consumed, and the subduction During the Miocene, opening of the Oki- zone became jammed by the middle Miocene nawa Trough continued and reached its most (Clift et al., 2008; Hinz et al., 1989). The Min- extensive phase in the Pliocene (Yamaji, 2003; doro ophiolites and the basement of Sabah, as Fig. 2E). Possibly related to this increase in roll- well as parts of the Lupar Line suture in west- back velocities, the Pacific trench propagated ern Sarawak or the Huatung Basin offshore Tai- westward, and subduction of the Philippine Sea wan, have been suggested to be remnants of this plate along the former strike-slip fault north of subduction zone (Deschamps et al., 2000; Hall, Shikoku Basin initiated (Figs. 2E and 3). Roll- 2002; Hutchison, 2005; Zahirovic et al., 2013). back of the Philippine Sea plate and contempo- Eastward consumption of the proto–South China raneous northwestward movement of the Luzon Sea is consistent with previous reconstructions arc resulted in the present-day configuration; i.e., (e.g., Hall, 2002; Zahirovic et al., 2013). In oblique collision of the Luzon arc with Eurasia tomographic images, flat-lying high-amplitude and consequently a mature steady-state orog- anomalies at a depth of 500–600 km under the eny in central Taiwan, and ongoing subduction Figure 3. Three-dimensional sketches show- ing the subduction initiation at the northern South China Sea, northwest Borneo, and the of oceanic Eurasian crust in southern Taiwan Philippine Sea plate (PSP) along a strike-slip Luzon arc have been interpreted as the proto– (Malavieille et al., 2002; Figs. 2E and 2F). zone and its evolution due to slab rollback. South China Sea slab (Zahirovic et al., 2013). Previous plate tectonic reconstructions of A: Initial situation at 17 Ma. S.C.—spreading After consumption of the proto–South China the area differ from our model (for compila- center. B: The Philippine Sea plate starts to subduct below the Eurasian margin, where Sea, the Eurasian margin is subducted in an east- tion of models, see the Data Repository), due to oceanic crust is opposed to continental Eur- ward direction beneath the Philippine Sea plate. uncertainties mentioned here, as well as lack of asian crust. C: The Okinawa Trough extends Due to northward movement of the Philippine data in some areas. Opportunities for testing the westward due to rapid rollback of the Philip- Sea plate together with the Australian plate and model will be afforded through detailed imaging pine Sea plate. accompanying clockwise rotation of the Philip- of the proto–South China Sea slab at depth, as pine Sea plate, the polarity reversal continued well as comprehensive mapping and dating of moving northward at the triple junction between its proposed remnants. the orogeny propagating southward due to the Eurasia, the Philippine Sea plate, and the Pacific oblique collision of the Luzon arc with Eur- plate between 37 and 21 Ma (Fig. 2C). This REGIONAL AND GLOBAL asia (e.g., Simoes and Avouac, 2006), as Suppe northward movement coincides with extension IMPLICATIONS (1984) initially conjectured. The difference affecting Southeast Asia and formation of the Our reconstruction has major implications between northern and southern Taiwan results South China Sea (e.g., Lin et al., 2003; Seton for Southeast Asia geodynamics and Taiwan in from the late interaction of the Ryukyu subduc- et al., 2012). The South China Sea reached its particular. The reconstructions show that forma- tion zone with the prestructured passive Eurasian maximum extent at 30 Ma, followed by a reor- tion and consumption of the proto–South China margin rather than from a decreasing degree of ganization of plate boundaries in Southeast Asia Sea played a key role for Southeast Asia geo- maturity from north to south. The Ryukyu sub- ca. 25 Ma that mostly affects the plate boundar- dynamics until the present day. First, this event duction zone does not play a major role during ies north of Australia (Hall, 2002; Seton et al., of seafloor spreading modified the buoyancy of the orogeny (Clift et al., 2003). In contrast, it 2012; Zahirovic et al., 2013). the passive margin. During this spreading, the is the Mesozoic polarity reversal that controls North of the Philippine Sea plate the Pacific Pacific plate rolled back and subducted below recent mountain building in Taiwan. plate was subducting northward below Eurasia. the proto–South China Sea. Due to northward At a larger scale, this study emphasizes the During the late Paleogene to early Miocene, the movement of the Philippine Sea plate, young, importance of global reconstructions for under- Pacific plate started rolling back, which may buoyant oceanic crust was juxtaposed against standing subduction polarity reversal and their be related to spreading initiation between the older oceanic crust of the proto–South China influence on present-day mountain belts. Polar- Kyushu Ridge and the West Mariana Ridge and Sea. This resulted in eastward subduction of the ity reversal, triggered by plate reorganization opening of the Shikoku and Parece Vela Basins proto–South China Sea. This reversed subduc- and buoyancy contrasts (Gurnis et al., 2004), between 32 and 31 Ma (Mrozowski and Hayes, tion polarity was able to propagate northward may migrate over large distances. Two-dimen- 1979), and to opening of the Okinawa Trough along the Eurasian margin until it reached the sional models such as spontaneous slab break- (Xu et al., 2014; Fig. 2C). At 20 Ma, the Philip- Taiwan area, where further propagation was off, or models only based on the present-day pine Sea plate continued its northward motion inhibited. At a later stage, the Ryukyu trench, plate configuration, need to be tested against and clockwise rotation at a rate of 1.5°/m.y. along which the Philippine Sea plate is sub- models of plate kinematic evolution through The absence of displacement toward the east ducted northward below the Eurasian margin, time, and against reconstructions that include and spreading east of the Kyushu Ridge dis- rolled back and reached the Taiwan area. This lithosphere consumed previously. Our results abled propagation of the polarity reversal farther interplay between the two diachronous but may be applied to areas where similar detail is northeast. Due to ongoing rollback of the Pacific related movements resulted in the present-day not possible because of later tectonic events, for plate and continuous opening of the Shikoku plate configuration of Taiwan. example, in the European Alps or Caledonian and Parece Vela Basins east of the Kyushu Ridge This implies that northern Taiwan might not orogenic belts of northern Europe and the east- (Seno and Maruyama, 1984), a left-lateral fault be an older equivalent of central and southern ern United States. developed at the northwestern tip of the Philip- Taiwan, as commonly assumed (e.g., Malavie- ACKNOWLEDGMENTS pine Sea plate (Mahony et al., 2011; Figs. 2C ille et al., 2002; Suppe, 1984) but questioned Von Hagke was funded by a California Institute of and 2D). Spreading east of the Kyushu Ridge by new thermochronological data (Lee et al., Technology Tectonics Observatory postdoctoral fel- ceased ca. 17 Ma (Mrozowski and Hayes, 1979), 2015). Only in central and southern Taiwan is lowship. Discussions with Sabin Zahirovic, Jonny

GEOLOGY | Volume 44 | Number 8 | www.gsapubs.org 661 Wu, Bruce Shyu, and John Suppe helped to clarify tectonics along the north-western continental mar- Simoes, M., and Avouac, J.P., 2006, Investigating the the picture. We thank Jonathan Lewis, Taras Gerya, gin of Sabah/Borneo: Geologische Rundschau, kinematics of mountain building in Taiwan from and anonymous reviewers for constructive comments. v. 78, p. 705–730, doi:10​ ​.1007​/BF01829317. the spatiotemporal evolution of the foreland ba- Houseman, G., and England, P., 1993, Crustal thicken- sin and western foothills: Journal of Geophysical REFERENCES CITED ing versus lateral expulsion in the Indian-Asian Research, v. 111, no. B10, B10401, doi:​10​.1029​ Boyden, J.A., Müller, R.D., Gurnis, M., Torsvik, T.H., continental collision: Journal of Geophysical /2005JB004209. Clark, J.A., Turner, M., Ivey-Law, H., Watson, Research, v. 98, p. 12,233–12,249, doi:​10​.1029​ Stadler, G., Gurnis, M., Burstedde, C., Wilcox, L.C., R.J., and Cannon, J.S., 2011, Next-generation /93JB00443. Alisic, L., and Ghattas, O., 2010, 2010, The dy- plate-tectonic reconstructions using GPlates, in Hutchison, C.S., 2005, Geology of north-west Borneo: namics of plate tectonics and mantle flow: From Keller, G.R., and Baru, C., eds., Geoinformatics: Amsterdam, Elsevier, 421 p. local to global scales: Science, v. 329, p. 1033– Cyberinfrastructure for the solid Earth sciences: Jolivet, L., Tamaki, K., and Fournier, M., 1994, Ja- 1038, doi:​10​.1126​/science​.1191223. Cambridge, UK, Cambridge University Press, p. pan Sea, opening history and mechanism: A syn- Suppe, J., 1984, Kinematics of arc-continent collision, 95–113, doi:​10​.1017​/CBO9780511976308​.008. thesis: Journal of Geophysical Research, v. 99, flipping of subduction, and back-arc spreading Brown, D., et al., 2011, Arc–continent collision: The p. 22,237–22,259, doi:​10​.1029​/93JB0346. near Taiwan: Memoir of the Geological Society making of an orogen, in Brown, D., and Ryan, P., Lee, Y.-H., Byrne, T., Wang, W.-H., Lo, W., Rau, R.- of China, v. 6, p. 21–33. eds., Arc-continent collision: Berlin, Springer, p. J., and Lu, H.-Y., 2015, Simultaneous mountain Tapponnier, P., Peltzer, G., Le Dain, A.Y., Armijo, 477–493, doi:​10​.1007​/978​-3​-540​-88558​-0_17. building in the Taiwan orogenic belt: Geology, R., and Cobbold, P., 1982, Propagating extru- Clift, P.D., Schouten, H., and Draut, A.E., 2003, A v. 43, p. 451–454, doi:​10​.1130​/G36373​.1. sion tectonics in Asia: New insights from sim- general model of arc-continent collision and sub- Lin, A.T., Watts, A.B., and Hesselbo, S.P., 2003, Ce- ple experiments with plasticine: Geology, v. 10, duction polarity reversal from Taiwan and the nozoic stratigraphy and subsidence history of the p. 611–616, doi:​10​.1130​/0091​-7613​(1982)10​ Irish Caledonides, in Larter, R.D., and Leat, P.T., South China Sea margin in the Taiwan region: <611:​PETIAN​>2​.0​.CO;2. eds., Intra-oceanic subduction systems: Tectonic Basin Research, v. 15, p. 453–478, doi:10​ ​.1046​ Ustaszewski, K., Wu, Y.-M., Suppe, J., Huang, H.-H., and magmatic processes: Geological Society of /j​.1365​-2117​.2003​.00215​.x. Chang, C.-H., and Carena, S., 2012, Crust-mantle London Special Publication 219, p. 81–98, doi:​ Mahony, S., Wallace, L., Miyoshi, M., Villamor, P., boundaries in the Taiwan-Luzon arc-continent 10​.1144​/GSL​.SP​.2003​.219​.01​.04. Sparks, R., and Hasenaka, T., 2011, Volcano- collision system determined from local earth- Clift, P., Lee, G.H., Anh Duc, N., Barckhausen, U., Van tectonic interactions during rapid plate-bound- quake tomography and 1D models: Implications Long, H., and Zhen, S., 2008, Seismic reflection ary evolution in the Kyushu region, SW Japan: for the mode of subduction polarity reversal: Tec- evidence for a Dangerous Grounds miniplate: No Geological Society of America Bulletin, v. 123, tonophysics, v. 578, p. 31–49, doi:10​ .1016​ /j​ .tecto​ ​ extrusion origin for the South China Sea: Tecton- p. 2201–2223, doi:​10​.1130​/B30408​.1. .2011​.12​.029. ics, v. 27, TC3008, doi:​10​.1029​/2007TC002216. Malavieille, J., et al., 2002, Arc-continent collision in Xu, J., Ben-Avraham, Z., Kelty, T., and Yu, H.-S., 2014, Conrad, C.P., and Lithgow-Bertelloni, C., 2002, How Taiwan: New marine observations and tectonic Origin of marginal basins of the NW Pacific and mantle slabs drive plate tectonics: Science, v. 298, evolution, in Byrne, T.B., and Liu, C.-H., eds., their plate tectonic reconstructions: Earth-Sci- p. 207–209, doi:​10​.1126​/science​.1074161. Geology and geophysics of an arc-continent col- ence Reviews, v. 130, p. 154–196, doi:10​ ​.1016​/j​ Deschamps, A., and Lallemand, S., 2002, The West lision, Taiwan: Geological Society of America .earscirev​.2013​.10​.002. Philippine Basin: An Eocene to early Oligocene Special Paper 358, p. 187–211, doi:​10​.1130​/0​ Yamaji, A., 2003, Slab rollback suggested by latest back arc basin opened between two opposed sub- -8137​-2358​-2​.187. Miocene to Pliocene forearc stress and migration duction zones: Journal of Geophysical Research, Morley, C.K., 2012, Late Cretaceous–Early Palaeo- of volcanic front in southern Kyushu, northern v. 107, 2322, 24 p., doi:10​ ​.1029​/2001JB001706. gene tectonic development of SE Asia: Earth- Ryukyu Arc: Tectonophysics, v. 364, p. 9–24, doi:​ Deschamps, A., Monié, P., Lallemand, S., Hsu, S.K., Science Reviews, v. 115, p. 37–75, doi:​10​.1016​ 10​.1016​/S0040​-1951​(03)00047​-7. and Yeh, K.Y., 2000, Evidence for Early Cre- /j​.earscirev​.2012​.08​.002. Yang, S., Hu, S., Cai, D., Feng, X., Chen, L., and Gao, taceous oceanic crust trapped in the Philippine Mrozowski, C.L., and Hayes, D.E., 1979, The evolu- L., 2004, Present-day heat flow, thermal history Sea plate: Earth and Planetary Science Letters, tion of the Parece Vela Basin, eastern Philippine and tectonic subsidence of the East China Sea v. 179, p. 503–516, doi:10​ .1016​ /S0012​ -821X​ (00)​ ​ Sea: Earth and Planetary Science Letters, v. 46, Basin: Marine and Petroleum Geology, v. 21, 00136​-9. p. 49–67, doi:​10​.1016​/0012​-821X​(79)90065​-7. p. 1095–1105, doi:​10​.1016​/j​.marpetgeo​.2004​ Dewey, J.F., Pitman, W.C., Ryan, W.B.F., and Bonnin, Ryan, W.B., Carbotte, S.M., Coplan, J.O., O’Hara, S., .05.007. J., 1973, Plate tectonics and the evolution of the Melkonian, A., Arko, R., Weissel, R.A., Ferrini, Zahirovic, S., Seton, M., and Mueller, R.D., 2013, The Alpine system: Geological Society of America V., Goodwillie, A., and Nitsche, F., 2009, Global Cretaceous and Cenozoic tectonic evolution of Bulletin, v. 84, p. 3137–3180, doi:10​ ​.1130​/0016​ multi-resolution topography synthesis: Geochem- Southeast Asia: Solid Earth Discussions, v. 5, -7606​(1973)84​<3137:​PTATEO>2​.0​.CO;2. istry, Geophysics, Geosystems, v. 10, Q03014, p. 1335–1422, doi:​10​.5194​/sed​-5​-1335​-2013. Gurnis, M., Hall, C., and Lavier, L., 2004, Evolv- doi:​10​.1029​/2008GC002332. Zhou, X.M., and Li, W.X., 2000, Origin of late Meso­ ing force balance during incipient subduction: Seno, T., and Maruyama, S., 1984, Paleogeographic zoic igneous rocks in southeastern China: Implica- Geochemistry, Geophysics, Geosystems, v. 5, reconstruction and origin of the Philippine Sea: tions for lithosphere subduction and underplating Q07001, doi:​10​.1029​/2003GC000681. Tectonophysics, v. 102, p. 53–84, doi:​10​.1016​ of mafic magmas: Tectonophysics, v. 326, p. 269– Hall, R., 2002, Cenozoic geological and plate tec- /0040​-1951​(84)90008​-8. 287, doi:10​ .1016​ /S0040​ -1951​ (00)​ 00120-7.​ tonic evolution of SE Asia and the SW Pacific: Seton, M., et al., 2012, Global continental and ocean Computer-based reconstructions, model and ani- basin reconstructions since 200 Ma: Earth-Sci- mations: Journal of Asian Earth Sciences, v. 20, ence Reviews, v. 113, p. 212–270, doi:10​ ​.1016​/j​ Manuscript received 22 February 2016 p. 353–431, doi:10​ .1016​ /S1367​ -9120​ (01)​ 00069-4.​ .earscirev​.2012​.03​.002. Revised manuscript received 9 June 2016 Hinz, K., Fritsch, J., Kempter, E., Mohammad, M.A.M., Sibuet, J.-C., and Hsu, S.-K., 2004, How was Taiwan Manuscript accepted 17 June 2016 Meyer, J., Mohamed, M.D., Vosberg, D.G.H., created?: Tectonophysics, v. 379, p. 159–181, doi:​ Weber, D.I.J., and Benavidez, M.J., 1989, Thrust 10​.1016​/j​.tecto​.2003​.10​.022. Printed in USA

662 www.gsapubs.org | Volume 44 | Number 8 | GEOLOGY