Search

Large igneous provinces (LIPs) have been linked to both surface and deep mantle processes related to supercontinent formation. During the formation, tenure, and breakup of Pangea, the most recent supercontinent, there is a noted contemporaneous increase in the number of emplacement events of both continental and oceanic LIPs. There is currently no clear consensus on the origin of LIPs, but the most widely recognized hypothesis relates their formation to crustal emplacement of hot plume material originating in the deep mantle. The interaction of subducted slabs with the lowermost mantle thermal boundary and subsequent return-flow is a key control on plume generation. This mechanism has been explored for LIPs below the interior of a supercontinent (e.g., continental LIPs). However, a number of LIPs related to Pangea formed at the supercontinent’s exterior (e.g., Ontong Java Plateau in the Pacific Ocean), with no consensus on their formation mechanism. In this research, we consider the dynamics of global-scale supercontinent processes resultant from numerical models of mantle convection, and analyse whether circum-supercontinent subduction could generate both interior (continental) and exterior (oceanic) deep-mantle plumes. Our 2-D and 3-D numerical models show that subduction related to the supercontinent cycle can reproduce the location and timing of the Ontong Java Plateau, Caribbean LIP, and potentially the Shatsky Rise, when relating these LIPs to a deep mantle exterior plume. The findings here highlight the importance of taking into consideration mantle dynamics in every stage of the supercontinent cycle.

The late Neoproterozoic–Cambrian interval is characterized by global-scale orogenesis, rapid continental growth and profound changes in Earth systems. Orogenic activity involved continental collisions spanning more than 100 myr, culminating in Gondwana amalgamation. Avalonia is an example of arc magmatism and accretionary tectonics as subduction zones re-located to Gondwana's periphery in the aftermath of those collisions, and its evolution provides significant constraints for global reconstructions. Comprising late Neoproterozoic ( c. 650–570 Ma) arc-related magmatic and metasedimentary rocks, Avalonia is defined as a composite terrane by its latest Ediacaran–Ordovician overstep sequence: a distinctive, siliciclastic-dominated cover bearing ‘Acado-Baltic’ fauna. This definition implies that Neoproterozoic Avalonia may consist of several terranes, and so precise palaeomagnetic or provenance determination in one locality need not apply to all. On the basis of detrital zircon and Nd isotopic data, Avalonia and other lithotectonically related terranes, such as Cadomia, have long been thought to have resided along the Amazonian–West African margin of Gondwana between c. 650 and 500 Ma – Avalonia connected to Amazonia, and Cadomia to West Africa. These interpretations have constrained Paleozoic reconstructions, many of which imply that the departure of several peri-Gondwanan terranes led to the Early Paleozoic development of the Rheic Ocean whose subsequent demise in the Late Paleozoic led to Pangaea's amalgamation. Since these ideas were proposed, several new lines of evidence have challenged the Amazonian affinity of Avalonia. First, there is evidence that some Avalonian terranes may have been ‘peri-Baltican’ during the Neoproterozoic. Baltica was originally excluded as a potential source for Avalonia because, unlike Amazonia, arc-related Neoproterozoic rocks were not documented. However, subsequent recognition of Ediacaran arc-related sequences in the Timanides of northeastern Baltica invalidates this assumption. Second, detailed palaeontological and lithostratigraphic studies have been interpreted to reflect an insular Avalonia, well removed from either Gondwana or Baltica during the Ediacaran and early Cambrian. Third, recent palaeomagnetic data have raised the possibility of an ocean (Clymene Ocean) between Amazonia and West Africa in the late Neoproterozoic, thereby challenging conventional reconstructions that show the ‘peri-Gondwanan’ terranes as a contiguous belt straddling the suture zone between these cratons. In this contribution, we critically re-evaluate the provenance of the so-called ‘peri-Gondwanan’ terranes, the contiguity of the so-called ‘Avalonian–Cadomian’ belt and the validity of the various plate tectonic models based on the traditional interpretation of these terranes. In addition, we draw attention to critical uncertainties and the challenges that lie ahead.

Diachronous opening of the Rheic Ocean and separation of Avalonian–Cadomian terranes from Gondwana began with a change from an active to passive margin in the late Ediacaran to early Cambrian. During the Cambrian, extension within these terranes was recorded by magmatism and by the development of sedimentary basins. However, the timing, style, and kinematics of this transformation still remain poorly understood and plate-scale models vary significantly. To address this issue, the Přibram–Jince basin in the Bohemian Massif was chosen as a case study since it preserves an excellent record of Cambrian rifting. Here, after Cadomian subduction ceased at ∼527 Ma, extension initiated and a thick pile of continental siliciclastics was deposited in the basin between ∼515 Ma and ∼499 Ma, interrupted by marine transgression at ∼506–503 Ma. Our field, paleocurrent, and rock-magnetic data suggest that the source areas were located to the ∼ESE and to the ∼SW of the basin during the deposition of the lower and upper formations, respectively. Sediment sources changed accordingly from distant metamorphic basement (Gondwana?) and Cadomian volcanic arcs and an accretionary wedge underlying the basin. This redirection marked a change in the tectonic evolution of the basin from orthogonal to dextral oblique extension that enlarged the basin into a pull-apart structure. Integrating this depositional and tectonic record into a large-scale picture, we suggest that strike-slip movements along the former Avalonian–Cadomian belt controlled the diachronous opening of the Rheic Ocean. An inherited suture in the Avalonian ribbon terrane facilitated complete rifting and rift–drift transition while the Cadomian terranes remained attached to Gondwana. The kinematics of this event remains controversial. Either it was opposite along the westerly (sinistral) and easterly (dextral) segments of the belt, which may be explained by interaction with an intervening spreading center, or it was the same dextral transtension.

Earth's long-term climate has been profoundly influenced by the episodic assembly and breakup of supercontinents at intervals of ca. 500 m.y. This reflects the cycle's impact on global sea level and atmospheric CO

The status of Pannotia as an Ediacaran supercontinent, or even its mere existence as a coherent large landmass, is controversial. The effect of its hypothesized amalgamation is generally ignored in mantle convection models claiming the transition from Rodinia to Pangaea represents a single supercontinent cycle. We apply three geodynamic scenarios to Pannotia amalgamation that are tested using regional geology. Scenarios involving quasi-stationary mantle convection patterns are not supported by the geological record. A scenario involving feedback between the supercontinent cycle and global mantle convection patterns predicts upwellings beneath the Gondwanan portion of Pannotia and the arrival of plumes along the entire Gondwanan (but not Laurentian) margin beginning at c. 0.6 Ga. Such a scenario is compatible with regional geology, but the candidates for plume magmatism we propose require testing by detailed geochemical and isotopic studies. If verified, this scenario could provide geodynamic explanations for the origins of the late Neoproterozoic and Early Palaeozoic Iapetus and Rheic oceans and the terranes that were repeatedly detached from their margins.

Controversy about the status of Pannotia (Laurentia + Baltica + Gondwana) as an Ediacaran supercontinent centers on palaeomagnetic data (which is permissive not conclusive) and geochronology (which implies breakup commenced before full assembly). But evidence of past supercontinent assembly is not limited to these two criteria and can be found in many other phenomena that accompany the process. Irrespective of whether Pannotia qualifies as a supercontinent, a key unanswered question is whether the legacy of its amalgamation influenced global mantle convection patterns because such patterns are generally ignored in models claiming the transition from Rodinia to Pangaea represents a single supercontinent cycle. We contend that the proxy signals of assembly and breakup in the Ediacaran are unmistakable and indicate profound changes in mantle circulation. These changes correlate with a wealth of geologic data for Pan-African collisional orogenesis, reflecting the amalgamation of the Gondwana, and for tectonothermal activity along the Gondwanan portion of Pannotia’s periphery. Collisional orogenesis necessitates subduction of oceanic lithosphere between the converging continental blocks. By analogy with the amalgamation of Pangea, the subducted oceanic lithosphere should have congregated to form a “slab graveyard” along the core-mantle boundary that would have generated a superplume beneath the Gondwanan component of Pannotia, the effects of which can be seen along its margins. We suggest that such dramatic changes in mantle convection patterns can indeed be recognized, they provide insights into the processes responsible for the opening of the Iapetus and Rheic oceans, and a potential explanation for some of the enigmatic tectonothermal events that characterize the Late Neoproterozoic-Early Paleozoic tectonic evolution of the margin of Gondwana.

A supercontinent is generally considered to reflect the assembly of all, or most, of the Earth's continental lithosphere. Previous studies have used geological, atmospheric and biogenic ‘geomarkers’ to supplement supercontinent identification. However, there is no formal definition of how much continental material is required to be assembled, or indeed which geomarkers need to be present. Pannotia is a hypothesized landmass that existed in the interval c. 0.65–0.54 Ga and was comprised of Gondwana, Laurentia, Baltica and possibly Siberia. Although Pannotia was considerably smaller than Pangaea (and also fleeting in its existence), the presence of geomarkers in the geological record support its identification as a supercontinent. Using 3D mantle convection models, we simulate the evolution of the mantle in response to the convergence leading to amalgamation of Rodinia and Pangaea. We then compare this supercontinent ‘fingerprint’ to Pannotian activity. For the first time, we show that Pannotian continental convergence could have generated a mantle signature in keeping with that of a simulated supercontinent. As a result, we posit that any formal identification of a supercontinent must take into consideration the thermal evolution of the mantle associated with convergence leading to continental amalgamation, rather than simply the size of the connected continental landmass.

The West Kunlun Orogenic Belt (WKOB) is a key area for evaluating the evolution of the Proto-Tethys Ocean. To constrain this history, we present geochemical and zircon U-Pb geochronological data from a suite of metasedimentary rocks within this orogenic belt. Detrital zircons ages are clustered between 1000 and 400 Ma, with major peaks at ca. 510 Ma and 460 Ma, and minor peaks at ca. 840 Ma and 650 Ma. The age groups younger than 1000 Ma broadly overlap magmatic activity at ca. 540–420 Ma and minor magmatism at 920–540 Ma in the WKOB. The age data also reveal Early Paleozoic metamorphic basement rocks in the WKOB that were probably sourced from Kunlun terranes and blocks of Gondwanan affinity rather than the Tarim Craton. Compiled geochemical data suggest a protracted subduction setting in the WKOB in the Early Paleozoic. eHf(t) values show increasing trends at 900–740 Ma and 740–600 Ma and a distinct decreasing trend at 540–420 Ma, consistent with retreating subduction during the Neoproterozoic and advancing subduction in the Early Paleozoic. The retreating-advancing subduction cycle in the WKOB was possibly related to the opening and closure of the Proto-Tethys Ocean along the margin of Gondwana.

The geology of SW England has long been interpreted to reflect Variscan collisional processes associated with the closure of the Rhenohercynian Ocean and the formation of Pangea. The Cornish peninsula is composed largely of Early Devonian to Late Carboniferous volcanosedimentary successions that were deposited in pre- and syn-collisional basins and were subsequently metamorphosed and deformed during the Variscan orogeny. Voluminous Early Permian granitic magmatism (Cornubian Batholith) is broadly coeval with the emplacement of ca. 280–295 Ma lamprophyric dykes and flows. Although these lamprophyres are well mapped and documented, the processes responsible for their genesis and their relationship with regional Variscan tectonic events are less understood. Pre- to syn-collisional basalts have intra-continental alkalic affinities, and have REE profiles consistent with derivation from the spinel–garnet lherzolite boundary. eNd values for the basalts range from + 0.37 to + 5.2 and TDM ages from 595 Ma to 705 Ma. The lamprophyres are extremely enriched in light rare earth elements, large iron lithophile elements, and are depleted in heavy rare earth elements, suggesting a deep, garnet lherzolite source that was previously metasomatised. They display eNd values ranging from − 1.4 to + 1.4, initial Sr values of ca. 0.706, and TDM ages from 671 Ma to 1031 Ma, suggesting that metasomatism occurred in the Neoproterozoic. Lamprophyres and coeval granite batholiths of similar chemistry to those in Cornwall occur in other regions of the Variscan orogen, including Iberia and Bohemia. By using new geochemical and isotopic data to constrain the evolution of the mantle beneath SW England and the processes associated with the formation of these post-collisional rocks, we may be able to gain a more complete understanding of mantle processes during the waning stages of supercontinent formation.

Trabajo presentado en GSA Annual Meeting (The Geological Society of America), celebrado en Indianapolis, Indiana (Estados Unidos), del 4 al 7 de noviembre de 2018

The spatial and temporal association of post-collisional granites and lamprophyre dykes is a common but enigmatic relationship in many orogenic belts, including the Variscan orogenic belt of SW England. The geology of SW England has long been interpreted to reflect orogenic processes associated with the closure of the Rheic Ocean and the formation of Pangaea. The SW England peninsula is composed largely of Early Devonian to Carboniferous volcano-sedimentary successions deposited in synrift and subsequent syncollisional basins that underwent deformation and low-grade regional metamorphism during the Variscan orogeny. Voluminous Early Permian granitic magmatism (Cornubian Batholith) is considered to be broadly coeval with the emplacement of lamprophyric dykes and lamprophyric and basaltic lava flows, largely on the basis of geochronological data from lamprophyric lavas in Devon. Although published geochronological data for Cornish lamprophyre dykes are consistent with this interpretation, these data are limited largely to imprecise K–Ar whole-rock and biotite analyses, hindering the understanding of the processes responsible for their genesis and their relationship to granitic magmatism and regional Variscan tectonics. 40 Ar/ 39 Ar geochronological data for four previously undated lamprophyre dykes from Cornwall, combined with published data, suggest that lamprophyre magmatism occurred between c . 293.6 and c . 285.4 Ma, supporting previous inferences that their emplacement was coeval with the Cornubian Batholith. These data provide insights into (1) the relative timing between the lamprophyres and basalts, the Cornubian batholith and post-collisional magmatism elsewhere in the European Variscides, and (2) the post-collisional processes responsible for the generation and emplacement of lamprophyres, basalts and granitoids. Supplementary data: Complete datasets, photomicrographs and photographs of sample locations are available online at http://www.geolsoc.org.uk/SUP18838.

The peri-Gondwanan Meguma terrane of southern Nova Scotia, Canada, is the only major lithotectonic element of the northern Appalachian orogen that has no clear correlatives elsewhere in the Appalachians and lacks firm linkages to the Caledonide and Variscan orogens of western and southern Europe. This characteristic is in contrast with its immediate peri-Gondwanan neighbor, Avalonia, which has features in common with portions of Carolinia in the southern Appalachians and has been traced from the Rhenohercynian Zone of southern Britain eastward around the Bohemian Massif to the Carpathians and western Pontides. At issue is the tendency in Europe to assign all peri-Gondwanan terranes lying outboard of the Rheic suture to Avalonia, characterized by relatively juvenile basement and detrital zircon ages that include Mesoproterozoic populations, and those inboard of the suture to Cadomia, characterized by a more evolved basement and detrital zircon ages that match Paleoproterozoic and older sources in the West African craton. Although the unexposed basements of Avalonia and Meguma are thought to be isotopically very similar, the Meguma sedimentary cover contains scarce Mesoproterozoic zircon and is dominated instead by Neoproterozoic and Paleoproterozoic populations like those of Cadomia. Hence, felsic magma produced by crustal melting in the Meguma terrane (e.g. the ca. 370 Ma South Mountain Batholith) is isotopically more juvenile (eNd = –5 to –1, TDM = 1.3 Ga) than the rocks it intruded (eNd= –12 to –7, TDM = 1.7 Ga). By contrast, felsic magma produced by crustal melting in Avalonia (eNd = –1 to +6, TDM = 0.7–1.2 Ga) is isotopically similar to its host rocks (eNd = –3 to +4, TDM = 0.9–1.4). The isotopic relationship shown by the Meguma terrane has also been recognized in the South Portuguese Zone of southern Spain, which is traditionally assigned to Avalonia. However, the Sierra Norte Batholith of the South Portuguese Zone (ca. 330 Ma; eNd = +1 to –3, TDM = 0.9–1.2 Ga) is on average more juvenile than the Late Devonian host rocks (eNd = –5 to –11) it intruded, suggesting instead an extension of the Meguma terrane into Europe. Available data for the Cornubian Batholith of SW England (ca. 275–295 Ma; eNd = –4 to –7, TDM = 1.3–1.8 Ga) and the Devonian–Carboniferous metasedimentary rocks it intruded (eNd = –8 to –11) suggests this may also be true of that part of the southern Britain (Rhenohercynian Zone) with which the South Portuguese Zone is traditionally correlated.SOMMAIRELe terrane peri-gondwanien de Meguma en Nouvelle-Ecosse au Canada, est le seul grand element lithotectonique de l’orogene des Appalaches du Nord qui n’ait pas de correspondant avere ailleurs dans les Appalaches et qui ne montre aucun lien sur avec les orogenes caledonienne et varisque de l’ouest et du sud de l’Europe. Cette situation contraste avec celle de son voisin peri-gondwanien immediat, l’Avalonie, qui partage certaines caracteristiques avec des portions de Carolinia des Appalaches du sud et qui a ete suivi a partir de la zone rhenohercynienne dans le sud de la Grande-Bretagne vers l’est autour du massif bohemien jusqu’aux Carpates et l’ouest de la chaine pontique. Ce qui est en question ici c’est la tendance en Europe a assigner l’Avalonie a tous les terranes peri-gondwaniens situes a l’exterieur de la suture rheique lesquels sont caracterises par un socle relativement juvenile et des âges de zircons detritiques qui comportent des populations mesoproterozoiques, et ceux situes a l’interieur de la suture a Cadomia, lesquels sont caracterises par un socle plus evolue et des âges de zircons detritiques qui correspondent a des sources du craton ouest africain paleoproterozoiques et plus anciennes. Bien que l’on estime que les socles non-exposes des terranes d’Avalonie et de Meguma soient tres similaires isotopiquement, le couvert sedimentaire de Meguma ne renferme que de rares zircons mesoproterozoiques, et ce sont plutot les populations de zircons neoproterozoiques et paleoproterozoiques qui dominent, comme c’est le cas pour Cadomia. Il en ressort que le magma felsique produit par la fusion de croute dans le terrane de Meguma (par ex. le batholite de South Mountain de 370 Ma env.) est isotopiquement plus jeune (eNd = –5 a –1, TDM = 1.3 Ga) que les roches qu’il recoupe (eNd= –12 a –7, TDM = 1.7 Ga). Par opposition, le magma felsique produit par la fusion de la croute dans le terrane d’Avalonie (eNd = –1 a +6, TDM = 0.7–1.2 Ga) est isotopiquement similaire aux roches de son encaissant (eNd = –3 a +4, TDM = 0.9–1.4). Le profil isotopique du terrane de Meguma, traditionnellement assignee a l’Avalonie, a aussi ete detecte dans la Zone sud-portugaise du sud de l’Espagne. Cependant, le batholite de Sierra Norte de la Zone sud-portugaise (ca. 330 Ma; eNd = +1 a –3, TDM = 0.9–1.2 Ga) est en moyenne plus jeune que l’encaissant du Devonien moyen (eNd = –5 a –11) qu’il recoupe, ce qui permet de penser a une extension du terrane de Meguma en Europe. Les donnees disponibles du batholite de Cornubian dans le S-O de l’Angleterre (ca. 275–295 Ma; eNd = –4 a –7, TDM = 1.3–1.8 Ga) et des roches metasedimentaires devono-carboniferes qu’il recoupe (eNd = –8 to –11) permet de penser qu’il pourrait en etre de meme de cette portion du sud de la Grande-Bretagne (Zone rhenohercynienne) avec laquelle la Zone sud-portugaise est traditionnellement correlee.

this volume is an outcome of a symposium entitled “amalgamation and Breakup of Pangea in the americas” held in Duncan’s honor in conjunction with the Cordilleran Section Meeting of the Geological Society of america on March 29–31, 2012 in Queretaro, Mexico. the symposium was co-sponsored by IGCP 597 (Origin and Evolution of Pangea) and IGCP 574 (Bending and Bent Orogens, and Continental Ribbons) and featured 16 contributions from friends, colleagues, current and former students, and one from his son, Fraser. Many of the papers presented at Queretaro are represented in this volume, supplemented by others from friends and colleagues who could not make that meeting but wished to contribute to this commemorative volume (Fig. 1). Duncan Keppie’s CV reveals a record of sustained excellence and productivity for nearly 40 years with the publication of more than 260 papers in refereed journals, over 40 book chapters, and a host of government publications, maps and memoirs, and the editorship of 7 books or journal special issues. But those of us who have known Duncan Keppie as a friend and scientific colleague know that these numbers only begin to tell the story. Duncan has been an excellent motivator and innovator and has an infectious enthusiasm for the earth sciences. His papers are not just a documentation of the evolution of various orogenic belts, although that in itself is a significant accomplishment. they also document fundamental processes, the principles, and practices of which have been applied to the structural evolution of orogenic belts throughout the world and continue to provide a theoretical framework for the understanding of the evolution and growth of continental crust. One of his many gifts, and a major strength in his research, is an unparalleled ability to grasp the essence of concepts and synthesize geological data from a very wide range of fields and from seemingly disparate pieces Duncan keppie: an appreciation and a celebration

The recognition that Earth history has been punctuated by supercontinents, the assembly and breakup of which have profoundly influenced the evolution of the geosphere, hydrosphere, atmosphere and biosphere, is arguably the most important development in Earth Science since the advent of plate tectonics. But whereas the widespread recognition of the importance of supercontinents is quite recent, the concept of a supercontinent cycle is not new and advocacy of episodicity in tectonic processes predates plate tectonics. In order to give current deliberations on the supercontinent cycle some historical perspective, we trace the development of ideas concerning long-term episodicity in tectonic processes from early views on episodic orogeny and continental crust formation, such as those embodied in the chelogenic cycle, through the first realization that such episodicity was the manifestation of the cyclic assembly and breakup of supercontinents, to the surge in interest in supercontinent reconstructions. We then chronicle some of the key contributions that led to the cycle's widespread recognition and the rapidly expanding developments of the past ten years.

The supercontinent cycle, by which Earth history is seen as having been punctuated by the episodic assembly and breakup of supercontinents, has influenced the rock record more than any other geologic phenomena, and its recognition is arguably the most important advance in Earth Science since plate tectonics. It documents fundamental aspects of the planet's interior dynamics and has charted the course of Earth's tectonic, climatic and biogeochemical evolution for billions of years. But while the widespread realization of the importance of supercontinents in Earth history is a relatively recent development, the supercontinent cycle was first proposed thirty years ago and episodicity in tectonic processes was recognized long before plate tectonics provided a potential explanation for its occurrence. With interest in the supercontinent cycle gaining momentum and the literature expanding rapidly, it is instructive to recall the historical context from which the concept developed. Here we examine the supercontinent cycle from this perspective by tracing its development from the early recognition of long-term episodicity in tectonic processes, through the identification of tectonic cycles following the advent of plate tectonics, to the first realization that these phenomena were the manifestation of episodic supercontinent assembly and breakup.

The supercontinent cycle has had a profound effect on the Earth's evolution since the Late Archean but our understanding of the forces responsible for its operation remains elusive. Supercontinents appear to form by two end-member processes: extroversion, in which the oceanic lithosphere surrounding the supercontinent (exterior ocean) is preferentially subducted (e.g. Pannotia), and introversion in which the oceanic lithosphere formed between dispersing fragments of the previous supercontinent (interior ocean) is preferentially subducted (e.g. Pangea). Extroversion can be explained by “top–down” geodynamics, in which a supercontinent breaks up over a geoid high and amalgamates above a geoid low. Introversion, on the other hand, requires that the combined forces of slab-pull and ridge push (which operate in concert after supercontinent break-up) must be overcome in order to enable the previously dispersing continents to turn inward. Introversion may begin when subduction zones are initiated along boundaries between the interior and exterior oceans and become trapped within the interior ocean. We speculate that the reversal in continental motion required for introversion may be induced by slab avalanche events that trigger the rise of superplumes from the core-mantle boundary.

High pressure (HP) rocks in the Paleozoic Acatlan Complex of southern Mexico occur in a median belt extruded into the complex, which is the root zone of a klippe to the west. These HP rocks are predominantly clastic metasedimentary rocks and include a bimodal suite of amphibolites and granitoids. U–Pb LA-ICPMS zircon dating indicates that one of the amphibolites was intruded at or before 461 ± 2 Ma, and a megacrystic granitoid yielded an intrusive age of 488 + 8/− 7 Ma, which is within the 490–450 Ma range of other dated HP megacrystic granitoids. The amphibolites have a tholeiitic (transitional to alkalic), within-plate chemistry derived from a heterogeneous mantle by various degrees of partial melting. The host metasedimentary rocks contain detrital zircons derived both from the Acatlan granitoids and the ca. 1.0–1.3 Ga Oaxacan Complex, and are inferred to represent rifted passive margin sediments. An age of 348 + 4/− 1 Ma from zircons in a migmatized amphibolite is interpreted as the time of migmatization and dates decompression melting following peak pressure at depths of 45–55 km. Retrogression from 530 °C to 370 °C occurred between 343 and 335 Ma. The predominance of reasonably coherent, rifted passive margin rocks over a surface area of 10 × 90 km suggests removal of a large slice of the upper plate during flat slab subduction. This slice was subducted to depths of ca. 45–55 km where it was underplated and transferred to the overriding plate. The absence of Oaxacan Complex rocks in the HP belt suggests the Oaxacan basement was subducted to greater depths. Underplating was followed by extrusion facilitated by extension in the upper plate that produced Mississippian rift basins east of the median HP belt. Westward overthrusting of the HP rocks produced a klippe that overrode Mississippian clastic wedge sediments.

The Neoproterozoic-Early Cambrian evolution of peri-Gondwanan terranes (e.g. Avalonia, Carolinia, Cadomia) along the northern (Amazonia, West Africa) margin of Gondwana provides insights into the amalgamation of West Gondwana. The main phase of tectonothermal activity occurred between ca. 640–540 Ma and produced voluminous arc-related igneous and sedimentary successions related to subduction beneath the northern Gondwana margin. Subduction was not terminated by continental collision so that these terranes continued to face an open ocean into the Cambrian. Prior to the main phase of tectonothermal activity, Sm-Nd isotopic studies suggest that the basement of Avalonia, Carolinia and part of Cadomia was juvenile lithosphere generated between 0.8 and 1.1 Ga within the peri-Rodinian (Mirovoi) ocean. Vestiges of primitive 760–670 Ma arcs developed upon this lithosphere are preserved. Juvenile lithosphere generated between 0.8 and 1.1 Ga also underlies arcs formed in the Brazilide Ocean between the converging Congo/Sao Francisco and West Africa/Amazonia cratons (e.g. the Tocantins province of Brazil). Together, these juvenile arc assemblages with similar isotopic characteristics may reflect subduction in the Mirovoi and Brazilide oceans as a compensation for the ongoing breakup of Rodinia and the generation of the Paleopacific. Unlike the peri-Gondwanan terranes, however, arc magmatism in the Brazilide Ocean was terminated by continent-continent collisions and the resulting orogens became located within the interior of an amalgamated West Gondwana. Accretion of juvenile peri-Gondwanan terranes to the northern Gondwanan margin occurred in a piecemeal fashion between 650 and 600 Ma, after which subduction stepped outboard to produce the relatively mature and voluminous main arc phase along the periphery of West Gondwana. This accretionary event may be a far-field response to the breakup of Rodinia. The geodynamic relationship between the closure of the Brazilide Ocean, the collision between the Congo/Sao Francisco and Amazonia/West Africa cratons, and the tectonic evolution of the peri-Gondwanan terranes may be broadly analogous to the Mesozoic-Cenozoic closure of the Tethys Ocean, the collision between India and Asia beginning at ca. 50 Ma, and the tectonic evolution of the western Pacific Ocean.

The Rheic Ocean was one of the most important oceans of the Paleozoic Era. It lay between Laurentia and Gondwana from the Early Ordovician and closed to produce the vast Ouachita-Alleghanian-Variscan orogen during the assembly of Pangea. Rifting began in the Cambrian as a continuation of Neoproterozoic orogenic activity and the ocean opened in the Early Ordovician with the separation of several Neoproterozoic arc terranes from the continental margin of northern Gondwana along the line of a former suture. The rapid rate of ocean opening suggests it was driven by slab pull in the outboard Iapetus Ocean. The ocean reached its greatest width with the closure of Iapetus and the accretion of the peri-Gondwanan arc terranes to Laurentia in the Silurian. Ocean closure began in the Devonian and continued through the Mississippian as Gondwana sutured to Laurussia to form Pangea. The ocean consequently plays a dominant role in the Appalachian-Ouachita orogeny of North America, in the basement geology of southern Europe, and in the Paleozoic sedimentary, structural and tectonothermal record from Middle America to the Middle East. Its closure brought the Paleozoic Era to an end.

The Rheic Ocean formed at ca. 500 Ma, when several peri-Gondwanan terranes (e.g. Avalonia and Carolinia) drifted from the northern margin of Gondwana, and were consumed during the Late Carboniferous collision between Laurussia and Gondwana, a key event in the formation of Pangea. Several mafic complexes ranging in age from ca. 400–330 Ma preserve many of the lithotectonic and/or chemical characteristics of ophiolites. They are characterized by anomalously high eNd values that are typically either between or above the widely accepted model depleted mantle curves. These data indicate derivation from a highly depleted (HD) mantle and imply that (i) the mantle source of these complexes displays time-integrated depletion in Nd relative to Sm, and (ii) depletion is the result of an earlier melting event in the mantle from which basalt was extracted. The extent of mantle depletion indicates that this melting event occurred in the Neoproterozoic, possibly up to 500 million years before the Rheic Ocean formed. If so, the mantle lithosphere that gave rise to the Rheic Ocean mafic complexes must have been captured from an adjacent, older oceanic tract. The transfer of this captured lithosphere to the upper plate enabled it to become preferentially preserved. Possible Mesozoic–Cenozoic analogues include the capture of the Caribbean plate or the Scotia plate from the Pacific to the Atlantic oceanic realm. Our model implies that virtually all of the oceanic lithosphere generated during the opening phase of the Rheic Ocean was consumed by subduction during Laurentia–Gondwana convergence.

The basement of most peri-Gondwanan terranes in Mexico, the Appalachians, the Caledonides, and the Variscides is buried beneath younger Ediacaran arc, Palaeozoic passive margin, and/or Mesozoic–Cenozoic platformal carbonates. However, it is exposed in the Oaxaquia terrane of Mexico (Oaxacan, Novillo, Huiznopala, and Guichicovi complexes), where it is characterized by ca. 1.0–1.3 billion year protolith ages and igneous rocks with depleted mantle model ages (T DM) of 1.35–1.77 billion years. The T DM ages represent a bulk average composition of the source and can be used as a tracer; these T DM ages overlap with those in ca. 546 Ma arc clasts from the 65.5 Ma Chicxulub bolide breccia, suggesting that the northern Maya block is also underlain by Oaxaquia-type basement. Similar T DM ages occur in Ediacaran arc rocks in Suwannee (Florida), NW Avalonia, Ganderia, Iberia, Armorica, and Bohemia, and in lower Palaeozoic plutons cutting adjacent Palaeozoic passive margin rocks (Acatlan Complex, Gander Group), sugge...

The Olympus massif of NE Greece comprises metamorphosed and deformed Triassic and Cretaceous–Eocene carbonates considered to represent part of the passive southwestern margin of Neotethys, which separated the Apulia plate of mainland Greece from southern Europe in the Mesozoic. Ocean closure in the Early Tertiary led to subduction of this passive margin, as a result of which the Olympus carbonates were tectonically overridden in the Eocene by a series of thrust sheets consisting of locally high P/T (blueschist) metamorphosed continental margin sediments, basement granitoid gneisses, and ophiolitic rocks. Following collision, uplift of the Olympus carbonates by ~ 6–8 km was accomplished by Early Neogene ductile and Late Neogene–Recent brittle normal faulting, which exposed the carbonates in the form of a structural window through the overriding thrust stack. The focus of this exhumation occurred on the eastern flank of the massif where a thick calc–mylonite with down-to-the-east sense-of-shear indicators and a prominent zone of NNW- to NW-trending brittle normal faults separate the carbonates from telescoped structurally overlying units. Published 40 Ar/ 39 Ar microcline ages that indicate a thermal perturbation close to the contact zone in excess of 100–150 °C at 16–23 Ma (Early Miocene) are taken to date the onset of ductile extension, and imply long-term uplift rates of ~ 0.4 to 1 km/m.y. Degradation of Pleistocene glacial features on the Olympus massif and the presence on its eastern piedmont of deeply incised streams, erosional terraces, and fault scarps that offset the sedimentary deposits of three cycles of Pleistocene glacial deposition (Units 1–3), testify to its continuing uplift. A prominent frontal fault offsets Unit 2 by > 150 m and several subsidiary NW-trending normal faults with a cumulative displacement of ~ 130 m offset a paleosol developed on Unit 1. Kinematic indicators indicate normal dip-slip movement. Cumulative fault displacement for Unit 1 exceeds 277 m, yielding a minimum uplift rate of ~ 1.3 mm/yr, given the oxygen isotope stage 7 (Mindel/Riss) age of 210 ka proposed for the Unit 1 paleosol. Cumulative offset of > 196 m for the paleosol developed on Unit 2 yields an uplift rate of ~ 1.6 mm/yr, given its proposed isotope stage 5e (Riss/Wurm) age of 125 ka. This estimate is supported by longitudinal valley profile data that document ~ 200 m of stream incision since Unit 2 deposition and show the offset as reflecting uplift of the Olympus massif rather than subsidence of the adjacent Gulf of Thermaikos. Assuming 1.8 Ma for the base of the Pleistocene, these rates yield a total uplift range for the Quaternary of 2.3–4.1 km, consistent with the present 2.6 km height of Olympus above the exposed frontal fault. Neogene uplift of Olympus was likely facilitated by extension accompanying the opening of the Aegean back-arc basin during the Miocene as a result of the southward retreat of the Hellenic subduction zone. For the Pliocene–Recent, uplift was likely facilitated by extension linked to the termination of dextral motion at the western end of the North Anatolian Fault associated with the westward tectonic escape of Anatolia .

The Rheic Ocean, which separated Laurussia from Gondwana following the closure of Iapetus, is arguably the most important ocean of the Palaeozoic. Its suture extends from Mexico to Turkey and its closure produced the climactic Variscan–Alleghanian–Ouachita orogeny that assembled the supercontinent, Pangaea. Following protracted Cambrian rifting that represented a continuum from Neoproterozoic orogenic processes, the Rheic Ocean opened in the Early Ordovician with the separation of several Neoproterozoic arc terranes from the continental margin of northern Gondwana. Separation occurred along the line of a former Neoproterozoic suture following the onset of subduction in the outboard Iapetus Ocean. The timing of rift–drift transition and drive for subsequent spreading was likely governed by slab pull, accounting for the rapid rate (8–10 cm/yr) at which the Rheic Ocean widened. During the Ordovician, the ocean broadened at the expense of Iapetus and attained its greatest width (~ 4000 km) in the Silurian, by which time Baltica had sutured to Laurentia and the Neoproterozoic arc terranes had accreted to Laurussia, closing Iapetus in the process. Closure of the Rheic Ocean began in the Devonian and was facilitated by northward subduction beneath southern Baltica and southward subduction beneath northwest Gondwana. Closure was largely complete by the Mississippian as Gondwana and Laurussia sutured to build Pangaea, North Africa colliding with southern Europe to create the Variscan orogen in the Devonian–Carboniferous, and West Africa and South America suturing to North America to form the Alleghanian and Ouachita orogens, respectively, during the Carboniferous–Permian. The Rheic Ocean consequently plays a dominant role in the basement geology of southern Europe, in the Appalachian–Ouachita orogeny of North America, and in the Palaeozoic sedimentary, structural and tectonothermal record from Middle America to the Middle East. With its closure, the ocean brought about the assembly of Pangaea and brought the Palaeozoic Era to an end.

Differences in the origin and evolution of the Iapetan and Rheic oceans profoundly affected the development of the Ouachitan–Appalachian–Variscan orogen. The Iapetus Ocean was initiated in the Late Neoproterozoic–Early Cambrian by the separation of large continental landmasses (Laurentia, Baltica, Gondwana), whereas the Rheic Ocean originated in the Late Cambrian–Early Ordovician as the result of the separation of several ribbon continents, collectively termed peri-Gondwanan terranes, from the northern margin of Gondwana. Compared to the classic passive margin developed along the Laurentian margin of Iapetus, passive margin successions to the Rheic Ocean are more limited in extent. During convergence, both margins of the Iapetus Ocean developed western Pacific-type arc–back-arc systems, whereas subduction of the Rheic Ocean was primarily directed beneath Laurentia and was Andean in style. Closure of Iapetus involved a number of arc–back-arc accretionary events and had occurred by the Middle Silurian as a result of Laurentia–Baltica collision (to form Laurussia) and the accretion of the peri-Gondwanan terranes. Rheic Ocean closure occurred in the Carboniferous by way of continent–continent collision between Laurussia and Gondwana, with Gondwana on the lower plate. Within Iapetus, the formation of ophiolites occurred early in the orogenic cycle, and was followed soon after by obduction, HP–LT metamorphism and foreland-directed thrusting that is intimately related to accretionary events. These features dominate the northern Appalachian orogen where HT–LP metamorphism is attributed to post-accretionary slab break-off events. In contrast, HP–LT metamorphism within the Rheic Ocean reflects A-type subduction during Laurussia–Gondwana collision, and is rapidly succeeded by HT–LP metamorphism attributed to orogenic collapse. Ophiolites are rare and were formed either very early (Early Ordovician) or very late (Devonian–Carboniferous) in the evolution of the ocean. The Early Ordovician ophiolites were not obducted until the Carboniferous, at least 100 million years after their protoliths were formed. Rheic Ocean closure resulted in the formation of large-scale bivergent structures within the orogen, with symmetry changing on either side of the Rheic suture. Comparison of the evolution of the Iapetus and Rheic oceans suggests that differences in rift configuration, passive margin development, and subduction zone geometry within the two oceans profoundly influenced the subsequent style of orogenesis within the Ouachita–Appalachian–Variscan orogen.

A Siluro-Devonian, high-pressure (HP) belt (the Iberian–Czech or IC belt) extends from Iberia (Cordoba–Coimbra Shear Zone) through Armorica and the Massif Central to the Bohemian Massif (Iberian–Czech [IC] belt), and includes arc and periarc rocks, MORB and supra-subduction ophiolites, and passive margin sequences. It has generally been interpreted to be either: (a) a suture of the Galician, South Brittany, Massif Central and Moldanubian ocean; (b) a nappe rooted to the north in the Rheic Ocean; or (c) the result of post-collisional strike-slip shuffling of the Gondwanan margin during which a slice of the Rheic Ocean was inserted into the Gondwanan margin. We agree with a Rheic Ocean origin, but propose that the IC belt represents a pre-collisional part of the Gondwanan continental–oceanic margin that was removed by subduction erosion, underwent HP metamorphism, and was then extruded into the overlying Gondwanan plate. This model explains: (i) the lack of paleolatitudinal and faunal differences across the IC belt, (ii) the juxtaposition of active margin tectonics within synchronous passive margin sequences, and (iii) the pre-collisional age of most of the HP metamorphism. The complete range of HP rocks in the IC belt, from blueschist to hot eclogites, indicates that previous correlations with B- and A-type subduction, respectively, cannot be sustained.

Repeated amalgamation and subsequent breakup of continental lithosphere have profoundly affected Earth9s evolution since the Archean. Following breakup, distinctive rift and passive margin sequences along the trailing edges of dispersing continents have been used to identify such margins in the geologic past. Using western North America as an analogue, we show that the leading edges of dispersing continents have isotopic characteristics that can likewise be used to identify these margins. For example, the Sm-Nd isotopic signatures of Late Neoproterozoic and early Paleozoic igneous rocks along the northern margin of Gondwana indicate derivation from 0.7 to 1.1 Ga old mantle lithosphere. This lithosphere originated in the Mirovoi Ocean surrounding Rodinia. It subsequently accreted to northern Gondwana in response to Rodinia breakup, and provided a source for subsequent magmatism. Accretion and subsequent recycling of oceanic mantle lithosphere should be common along the leading edges of dispersing continents following supercontinent breakup, providing an additional aid in paleocontinental reconstructions.

Repeated cycles of supercontinent amalgamation and dispersal have occurred since the Late Archean and have had a profound influence on the evolution of the Earth's crust, atmosphere, hydrosphere, and life. When a supercontinent breaks up, two geodynamically distinct tracts of oceanic lithosphere exist: relatively young interior ocean floor that develops between the dispersing continents, and relatively old exterior ocean floor, which surrounded the supercontinent before breakup. The geologic and Sm/Nd isotopic record suggests that supercontinents may form by two end-member mechanisms: introversion (e.g. Pangea), in which interior ocean floor is preferentially subducted, and extroversion (e.g. Pannotia), in which exterior ocean floor is preferentially subducted. The mechanisms responsible remain elusive. Top–down geodynamic models predict that supercontinents form by extroversion, explaining the formation of Pannotia in the latest Neoproterozoic, but not the formation of Pangea. Preliminary analysis indicates that the onset of subduction in the interior (Rheic) ocean in the early Paleozoic, which culminated in the amalgamation of Pangea, was coeval with a major change in the tectonic regime in the exterior (paleo-Pacific) ocean, suggesting a geodynamic linkage between these events. Sea level fall from the Late Ordovician to the Carboniferous suggests that the average elevation of the oceanic crust decreased in this time interval, implying that the average age of the oceanic lithosphere increased as the Rheic Ocean was contracting, and that subduction of relatively new Rheic Ocean lithosphere was favoured over the subduction of relatively old, paleo-Pacific lithosphere. A coeval increase in the rate of sea floor spreading is suggested by the relatively low initial 87Sr/86Sr in late Paleozoic ocean waters. We speculate that superplumes, perhaps driven by slab avalanche events, can occasionally overwhelm top–down geodynamics, imposing a geoid high over a pre-existing geoid low and causing the dispersing continents to reverse their directions to produce an introverted supercontinent.

Detrital zircon age populations from Palaeozoic sedimentary and metasedimentary rocks in Mexico support palinspastic linkages to the northwestern margin of Gondwana (Amazonia) during the late Proterozoic–Palaeozoic. Age data from: (1) the latest Cambrian-Pennsylvanian cover of the c. 1 GaOaxacan Complex of southernMexico; (2) the ?Cambro-Ordovician to Triassic Acatlan Complex of southern Mexico’s Mixteca terrane; and (3) the ?Silurian Granjeno Schist of northeastern Mexico’s Sierra Madre terrane, collectively suggest Precambrian provenances in: (1) the c. 500–650 Ma Brasiliano orogens and c. 600–950 Ma Goias magmatic arc of South America, the Pan-African Maya terrane of the Yucatan Peninsula, and/or the c. 550–600 Ma basement that potentially underlies parts of the Acatlan Complex; (2) the Oaxaquia terrane or other c. 1 Ga basement complexes of the northern Andes; and (3) c. 1.4–3.0 Ga cratonic provinces that most closely match those of Amazonia. Exhumation within the Acatlan Complex of c. 440–480 Ma granitoids prior to the Late Devonian–early Mississippian, and c. 290 Ma granitoids in the early Permian, likely provided additional sources in the Palaeozoic. The detrital age data support the broad correlation of Palaeozoic strata in the Mixteca and Sierra Madre terranes, and suggest that, rather than representing vestiges of Iapetus or earlier oceanic tracts as has previously been proposed, both were deposited along the southern, Gondwanan (Oaxaquia) margin of the Rheic Ocean and were accreted to Laurentia during the assembly of Pangaea in the late Palaeozoic.