https://doi.org/10.1130/G46255.1 Manuscript received 24 October 2018 Revised manuscript received 15 May 2019 Manuscript accepted 16 May 2019 © 2019 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license. Published online 6 June 2019 First direct evidence for a contiguous Gondwana shelf to the south of the Rheic Ocean Rolf L. Romer1* and Uwe Kroner2 1German Research Centre for Geosciences (GFZ), Telegrafenberg, D-14473 Potsdam, Germany 2Department of Geology, Technical University Bergakademie Freiberg, Bernhard-v.-Cotta Strasse 2, 09599 Freiberg, Germany ABSTRACT ter point has seen major modifications: Franke Sea-level rise after the Hirnantian glaciation resulted in the global inundation of continen- et al. (2017, p. 257)—favoring models involving tal shelf areas and the widespread formation of early Silurian black shales. Black shales that several microplates separated by oceanic litho- were deposited on shelves receiving drainage from earlier glaciated areas have high uranium sphere—stated “only the Rheic Ocean between (U) contents because large-scale glacial erosion brought rocks with leachable U to the surface. Avalonia and peri-Gondwana was wide enough In contrast, black shales receiving drainage from non-glaciated areas that had lost leachable to be unambiguously recorded by biogeogra- U earlier have low U contents. Early Silurian U-rich shales formed only on shelf areas that phy and palaeomagnetism”. The limitation of had not been separated from earlier-glaciated mainland Gondwana by oceanic lithosphere. paleo magnetic constraints was additionally Therefore, early Silurian U-rich black shales within the Variscan orogen provide direct evi- highlighted by Franke et al. (2017, p. 277): “it dence that these areas had not been separated from mainland Gondwana, but were part of is worth remembering that Palaeozoic palaeo- the same, contiguous shelf. This implies that the Rheic Ocean was the only pre-Silurian ocean magnetic poles from Armorica, Iberia, Saxo- that opened during the early Paleozoic extension of the peri-Gondwana shelf. thuringia and Bohemia, are few and of variable quality, and there are time gaps with no data”. INTRODUCTION The Rheic Ocean is the only ocean common to These statements support the view that current The assembly of western Pangea included all models of this type. Other proposed oceans paleomagnetic data cannot be taken as evidence the collision between Laurentia and the East commonly are of limited extent and are not against a contiguous shelf (Kroner and Romer, European Craton to form Laurussia (Caledonian necessarily present in all models (Fig. 1B). 2010, 2013). orogeny) followed by the collision of Gondwana The second group of models includes two plates The late Cambrian to Early Ordovician ex- with Laurussia in Europe (Variscan orogeny) (Gondwana and Laurussia) (e.g., Robardet, tension in peri-Gondwana eventually leading and the final closure of the remaining Rheic 2003) and one single divergent plate boundary, to the separation of Avalonia resulted in the Ocean between western Africa and North Amer- i.e., the mid-ocean ridge of the Rheic Ocean. In redistribution of chemically distinctive sand- ica (Alleghanian orogeny). Avalonia, which is these models, post-Silurian collision resulted in stones. The occurrence of such redeposited the most important Gondwana-derived terrane long-lasting subduction-accretion processes of sandstones implies that the corresponding pre- to the north of the Rheic Ocean, had separated the segmented shelf of both plates, culminating Variscan basement blocks had not separated from Gondwana before the late Cambrian and in intraplate continental subduction during the from mainland Gondwana before the Early docked to the East European Craton before col- late stages of the Variscan collision (Kroner and Ordovician (Noblet and Lefort, 1990). More- liding with Laurentia (Matte, 2001; Kroner and Romer, 2010, 2013). over, there are numerous additional indirect Romer, 2013; Franke et al., 2017; Stephan et al., arguments favoring a contiguous pre-Silurian 2019a). There is a long-lived disagreement about THE DEBATE: CONTIGUOUS SHELF shelf to the south of the Rheic Ocean. Ordovi- the number of pre-Silurian oceans and, thus, in- VERSUS SEVERAL MICROPLATES cian to Devonian fossil assemblages suggest dependently moving blocks of continental crust Models of the Variscan orogen involving that the Variscan massifs to the south of the (microplates) between mainland Gondwana and the presence of several independently mov- Rheic Ocean and mainland Gondwana were mainland Laurussia (Fig. 1). ing microplates separated by pre-Silurian oce- part of the same shelf (Robardet, 2003). The Basically there are two end-member groups anic lithosphere use essentially the same data age distribution of detrital zircon in Cambrian of models for the Variscan orogen. The first as models for a contiguous shelf. The debate to Devonian sedimentary rocks demonstrates group of models involves two plates (Gondwana originates from the contrasting weighting of that the different Variscan massifs were part of and Laurussia) and several independently mov- the various data sets, none of which is decisive. the Gondwana shelf (Stephan et al., 2019a, and ing microplates or terranes—now preserved in For instance, paleomagnetic data were widely references therein), with different zircon assem- the various Variscan massifs—between them, believed to provide key evidence for viewing blages on the western and eastern parts of the all of which were separated by oceanic litho- the Variscan massifs between mainland Gond- contiguous shelf. The presence of detrital zircon sphere (e.g., Matte, 2001; Franke et al., 2017). wana and mainland Laurussia as independently of eastern-shelf provenance in Ordovician and moving microplates that were separated by Silurian sediments of the western shelf requires *E-mail: romer@ gfz -potsdam.de wide oceans (e.g., Tait et al., 1997). The lat- a contiguous shelf during redeposition of sedi- CITATION: Romer, R.L., and Kroner, U., 2019, First direct evidence for a contiguous Gondwana shelf to the south of the Rheic Ocean: Geology, v. 47, p. 767– 770, https:// doi .org /10 .1130 /G46255.1 Geological Society of America | GEOLOGY | Volume 47 | Number 8 | www.gsapubs.org 767 Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/767/4793711/767.pdf by guest on 29 September 2021 ducing conditions, U sourced from terrestrial A Contiguous East European Craton A.S. 250 Ma weathering becomes preferentially sequestered Peri-Gondwana shelf S.P. in the proximal shelf, eventually resulting in a south of the Laurussia I drawdown of U supply to the distal shelf and Rheic Ocean the adjacent abyssal plain. Avalonia Paleozoic hot shales reflecting major global Rheic suture anoxic events are known for instance from Tornquist suture Baltica (middle Cambrian to Early Ordovi- North II III Iapetus suture American cian Alum Shale), northern Gondwana (early Craton Silurian Lower Graptolite Shale), and North Iberia Variscides Tethys America (mid-Devonian Chattanooga Shale). IV The most important feature of these shale de- Adria equator posits—apart from their high U content—is that coeval black shale deposits on other continental Gondwana shelves do not have particularly high U contents C.S. Variscan overprint (Fig. 2). This implies that continuous input of U is essential, i.e., the fertility of the drainage area West Menderes serving as the U source controls whether a black African Arabia Craton shale turns into a hot shale or not (Romer and Cuney, 2018). High U input from surface runoff 1000 km U-deposit districts: I -Teplá-Barrandian on one shelf cannot be trapped in black shale Unit / Saxo-Thuringian Zone; II-French deposited on another shelf, i.e., U source and U B Earlier proposed sutures south of the Rheic suture Massif Central; III - Armorica; IV-Central Varisca Iberian Zone / Ossa Morena Zone trap cannot be separated by oceanic lithosphere. n deformation front STZ RHZ post-Variscan U-mineralization Uranium is readily mobilized in the con- TBU MSZ Rheic suture Silurian “hot shales” tinental environment, in particular in a highly M Paleozoic (meta)sediments with oxygenated or CO -rich atmosphere by oxida- AM Hirnantian tillites and dropstones 2 tion of U-rich sedimentary rocks, oxidation of t C.S. Devonian Chattanooga Shale ine fron uraninite in granites, and leaching of metamict FMC Alp N S.P. Silurian shales of Poland Iberia A.S. Cambro-Ordovician Alum Shale OMZ silicate minerals (Romer and Cuney, 2018). 500 km SPZ CIZ AM- Armorican Massif; CIZ- Central Iberian Zone; Thus, near-surface rocks are readily depleted in FMC-French Massif Central;M-Moldanubian Zone Galicia-Southern-Brittany (Matte, 2001) MSZ- Moravo-Silesian Zone; OMZ- Ossa Morena U and, thereafter, do not represent a U source. Al Zone; RHZ- Rheno-Hercynian Zone; SPZ- South pine fr Saxo-Thuringian (Franke et al., 2017) For black shales to become hot, they have to ont Portuguese Zone; STZ- Saxo-Thuringian Zone; Galicia-Moldanubian (Franke et al., 2017) TBU- Teplá-Barrandian Unit receive drainage from an area that still is a U source, i.e., leachable U must not have been lost Figure 1. A: Pangea reconstruction for 250 Ma (Stephan et al., 2019b) showing distribution of Hirnantian tillites and dropstones, early Silurian hot shales, and post-Variscan uranium previously. Tectonic uplift followed by strong mineralization (as a proxy for hot Siluran shales in orogenically overprinted peri-Gondwana erosion is a process that brings rocks not de- crust; yellow dots). Note early Silurian hot shales are restricted to the Gondwana mainland pleted earlier in U to the surface and thereby and Variscan blocks to the south of the Rheic suture. Data sources: Thickpenny and Leggett establishes a U source. Major glaciation also (1987), Lüning et al. (2000), Le Heron et al. (2009), Romer and Cuney (2018), Porębski et al. removes rocks that have lost their leachable U (2019). B: Simplified tectonic map of Variscan orogen showing earlier proposed oceanic su- tures (compiled from Matte, 2001; Franke et al., 2017).