Problems in Western Gondwana Geology

Home , Alexander du Toit, Asterotheca, Beaufort Group, Bokkeveld Group, Cape Fold Belt, Choiyoi Group

PROBLEMS IN WESTERN GONDWANA GEOLOGY

- I Workshop -

“South America - Africa correlations: du Toit revisited”

th th Gramado-RS-Brazil, August 27 to 29 , 2007

EXTENDED ABSTRACTS

Edited by Roberto Iannuzzi and Daiana R. Boardman

PROBLEMS IN WESTERN GONDWANA GEOLOGY

- I Workshop -

“South America - Africa correlations: du Toit revisited”

Gramado-RS-Brazil, August 27th to 29th, 2007

ORGANIZING COMMITTEE

Coordinators: Roberto Iannuzzi (CIGO-UFRGS) Farid Chemale Jr. (IG-UFRGS) José Carlos Frantz (IG-UFRGS)

Technical Support: Daiana Rockenbach Boardman (PPGeo-UFRGS) Cristina Félix (PPGeo-UFRGS) Graciela Pereira Tybusch (PPGeo-UFRGS)

Treasurer: Farid Chemale Jr. (IG-UFRGS)

Scientific Committee: Edison José Milani (CENPES/PETROBRAS) Victor Ramos (UBA, Argentina) Maarteen de Wit (UCT, África do Sul)

Editors: Roberto Iannuzzi (CIGO-UFRGS) Daiana Rockenbach Boardman (PPGeo-UFRGS)

SPONSORED BY

Centro de Investigações do Gondwana (CIGO-UFRGS) Instituto de Geociências da Universidade Federal do Rio Grande do Sul (IG-UFRGS) Programa de Pós-Graduação em Geociências (PPGeo-UFRGS) Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) Petróleo Brasileiro S.A. (PETROBRAS)

I PROBLEMS IN WESTERN GONDWANA GEOLOGY - I Workshop - South America - Africa correlations: du Toit revisited Gramado-RS-Brazil, August 27th to 29th, 2007

PREFACE

Early in the 20th Century, pioneering correlations between the Paleozoic- Mesozoic basins of South America and southern Africa were used by Alexander du Toit to support the initial concepts of continental drift and the proposal of a united Gondwana continent. Du Toit found the bio- and lithostratigraphy of the South American rock sequences of the Paraná Basin in Brazil and of the distant mountains of Sierra de la Ventana in Argentina to be remarkably similar to those that he had himself mapped out carefully for many years in the Cape-Karoo Basin and its flanking Cape Fold Belt mountains in southern Africa. Many geologists have remarked on the similarities since, often incrementally improving on the correlations synthesized by du Toit in his 1937 book Our Wandering Continents.

Considering the 70th anniversary of du Toit’s seminal book, the Centro de Investigações do Gondwana (CIGO) da UFRGS (Gondwana Investigation Center of the Rio Grande do Sul Federal University) and RGEOTEC (Petrobras-Universities Network for Tectonic Studies) have organized a 3-day workshop devoted to discussions on SW Gondwana state-of-the-art correlations. The central idea is to promote a forum where du Toit’s stratigraphic and structural framework of Africa’s and South America’s pre-drift configuration will be discussed, pioneering ideas up-dated, and relevant new lines of international research explored. To make this challenging task possible, a selected group of experienced researchers from Brazil, Argentina, Uruguay and South Africa were invited to participate in this workshop. As a result, this Workshop welcomes about 40 Gondwana specialists from many countries in the city of Gramado, at the Rio Grande do Sul State, in southern Brazil.

This workshop had a famous precedent in the “Problems in Brazilian Gondwana Geology”, edited by J.J. Bigarella, R.D. Becker and I.D. Pinto in August 1967. That book represented a Brazilian contribution to the First Gondwana Symposium, held in Mar del Plata, Argentina, and concerned entirely with Gondwana Geology exposed in Brazil, containing the articles of many contributors. It was the first time that a synthesis of geological knowledge produced by Brazilian specialists on Gondwana’s rocks was presented in the international scenario. In that occasion, the CNPq (Brazilian Research Council), the Geology Institute of Paraná Federal University, Commission for the Geological Chart of Paraná and the CIGO-UFRGS sponsored the publication. For this reason, the current workshop is also celebrating the forty years of the “Problems in Brazilian Gondwana Geology”.

The present 3-day workshop is dedicated to summary presentations on the 3 major topics of Africa-South America correlations, to be covered in detail during the event: a) Precambrian provinces and geochronology; b) Paleozoic - Mesozoic (up until break-up) basins sedimentology, stratigraphy, paleontology and magmatism; c) Cape-la Ventana mountain system. By the end of this workshop, the organizers II expected that a summary of key points emerged from these discussions: what we know, what we do not know, and what might be the best strategies for new investigation to improve cross-Atlantic correlations and greater understanding of the processes that shaped Gondwana and then tore it apart. In this way, we will be building on the shoulders of Alex du Toit.

The 1st Workshop on “Problems in Western Gondwana Geology” is promoted by the Centro de Investigações do Gondwana (CIGO-UFRGS), Instituto de Geociências da Universidade Federal do Rio Grande do Sul (IG-UFRGS), Programa de Pós-Graduação em Geociências (PPGeo-UFRGS), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), and mainly funded by Petróleo Brasileiro S.A. (Petrobras) through the RGEOTEC – Rede de Estudos Geotectônicos.

Edison José Milani Roberto Iannuzzi Farid Chemale Jr. José Carlos Frantz

III

RGEOTEC - Rede de Estudos Geotectônicos Tectonic Studies Network

According to the New Brazilian Law on petroleum issues (1997), the companies operating in the Country have to invest in Research & Development a certain percentage of their gross income from production of giant fields. Petrobras, with a long tradition of partnership with Brazilian universities, took advantage of this favorable environment to implement a new concept of relationship. Instead of individually contracting thousands of projects, the decision was to group them in a thematic way: 37 agreements were signed, creating networks of specific knowledge in areas considered of strategic importance for the oil industry. For the Geosciences, it includes the Tectonics, Sedimentology & Stratigraphy, Geophysics and Organic Geochemistrty networks.

The implementation of this new concept of relationship includes the total investment by Petrobras of about US$500 million in the period 2006-2008, supported by a growing national oil and gas production (close to 2 million barrels/day in 2007) and certainly favored by the high prices of petroleum in the international market these times. For the total investments, the Geosciences networks alone are responsible for about US$60 million.

The RGEOTEC – Tectonic Studies Network started to work in 2005, and includes the following Brazilian universities and research institutions: UFRGS, UFPR, USP, UNICAMP, UNESP, INPE, UFRJ, UFF, UERJ, UnB, ON, UFRN, UFOP and UFMG. Initial investments were mostly on creating/complementing universities’ infrastructure on geochronology equipment. Centered in the ON (National Observatory), a US$7 million pool of geophysical equipment (seismic refraction, magnetotelluric and gravimetry) is being established to become available for tectonic studies by early 2008.

Promoting the growing of scientific knowledge is one of the main objectives of the RGEOTEC, and the du Toit workshop stands exactly in the focus of action of this network concept. Thru this new policy, Petrobras reinforces its strong compromise with the making of science applied to oil industry and with the Coutry’s sustainable development.

V PROBLEMS IN WESTERN GONDWANA GEOLOGY - I Workshop - South America - Africa correlations: du Toit revisited Gramado-RS-Brazil, August 27th to 29th, 2007

EXTENDED ABSTRACTS

Biostratigraphic correlations in the Western Gondwana Paleozoic- Mesozoic; an holistic overview 1 J.M. Anderson A summary of the geochronology and Precambrian crustal architecture of southern Africa, and possibilities for correlations with South America 3 R. Armstrong Precambrian tectonic domains of São Paulo, Paraná and Santa Catarina States, South Brazil: identification of major neoproterozoic suture zones 9 M.A.S. Basei, O. Siga Jr., C.R. Passarelli Some structural and stratigraphic enigmas of the eastern Cape Fold Belt, South Africa 15 P. Booth Tectonic evolution of Neoproterozoic to Eopaleozoic belts in the Southern Brazil and Southern Africa 22 F. Chemale Late Carboniferous to Permian sequence stratigraphy in the main Karoo Basin of South Africa and its application in Southwestern Gondwanaland 26 D.I. Cole Chrono-, chemical-, seismic- , electrical- and tectono-stratigraphy across parts of the Cape Fold Belt – Karoo Basin of South Africa: New foundations for correlations across the South Atlantic 34 M. de Wit Tectonic and climatic induced changes in depositional styles of the Mesozoic sedimentary record of southern Paraná Basin, Brazil 42 U. Faccini Stratigraphy and Sedimentology of the Late Paleozoic Glacial Record of the Paraná Basin: Brazil 46 A.B. França and F.F. Vesely Correlation of Neoproterozoic terranes between SE-Brazil and Africa: comparative tectonic evolution and open questions) 55 M. Heilbron, C.M. Valeriano, C.C.G. Tassinari, J. Almeida, M. Tupinambá, O. Siga Jr. VI Examples of climatic, tectonic and eustatic controls on the stratigraphic signatures of the Early Permian succession in the Paraná Basin 62 M. Holz, J. Küchle, P.D. Reis, J. Casagrande Biostratigraphic versus Chronologic frameworks in the Early Permian from Paraná Basin: looking forward a possible consensus. 72 R. Iannuzzi The Late Paleozoic Gondwanide Orogen: the Sierra de la Ventana Transpressional Foldbelt 78 M.S. Japas Paleoenvironmental evolution and biostratigraphy of Late Paleozoic Andean Gondwana basins 86 C.O. Limarino and S.N. Césari Late Triassic-Early Jurassic western Gondwana tetrapod correlations C. Marsicano 94 The Paraná Basin: a multi-cycle sedimentary and magmatic intracratonic province of W Gondwana 99 E.J. Milani Cyclostratigraphy during the Carboniferous glaciations in central western Argentina: glacial ageism and tectonic framework 108 P.J. Pazos Silurian and Devonian World–Wide Flooding Events in the Paraná Basin 115 E. Pereira, S. Bergamaschi, R. Rodrigues, M.S.P. Souza Paleontological characterization of Supersequences in the Paraná Basin 121 L.P. Quadros Early correlations between Cape Town and Ventania Systems: Keidel’s pioneer work and his influence on Wegener’s Continental drift and Du Toit’s ideas 128 V.A. Ramos The Ventania System: Tectonic constraints in its Paleozoic evolution and the Patagonia accretion 132 V.A. Ramos The stratigraphic nature of the Ordovician Graafwater/Peninsula formational contact in the lower Table Mountain Group: Diastem or discordance? 137 J. Rogers and Z. Dlamini The Passa Dois Group (Paraná Basin, Permian): investigations in progress 151 R. Rohn Karoo tetrapod biostratigraphy: relevance to understanding VII Gondwanan development 158 B. Rubidge Stratigraphic architecture of eolian strata of the Botucatu and Serra Geral formations (latest Jurassic – earliest Cretaceous), Paraná Basin, Brazil 168 C.M.S. Scherer The search for the P/Tr and Tr/J boundaries in South Brazil, Argentina and Uruguay 170 C.L. Schultz Environmental impact of the End-Permian Extinction on the Karoo Basin of South Africa 172 R.M.H. Smith An integrative analysis on advances and perspectives of the Pennsylvanian and Permian palynostratigraphy in the Paraná/Chacoparaná Basin (Brazil, Argentina and Uruguay) 181 P.A. Souza, M.M. Vergel, Á. Beri Serra Geral Magmatism in the Paraná Basin – A new stratigraphic proposal, chemical stratigraphy and geological structures 189 W. Wildner, L.A. Hartmann, R.C. Lopes

1 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Biostratigraphic correlations in the Western Gondwana Paleozoic- Mesozoic; an holistic overview

John M. Anderson South African National Biodiversity Institute, Pretoria

We are moving ever closer to a world—whether that of Pangaea in the Paleozoic- Mesozoic or that of today—where a holistic viewpoint is fundamental. This Du Toit Workshop is clearly encouraging that vision. My presentation is likewise organized around this holistic view. Clearly the more accurate and closely resolved our correlations of geological strata (formations, members) between basins and between continents, the more reliable and holistic any palaeontological syntheses we might generate will become. We are, for instance, about to go to press with a book entitled ‘A Brief History of Gymnosperms’. In it we attempt a global review of the classification, biodiversity, phytogeography and ecology of the gymnosperms from their origin in the Late Devonian through to the present. Aside from considerations such as levels of sampling and robustness of taxonomy, any study of this nature can only be as good as the accuracy of the correlations on which it is based. We show a set of correlation charts of the megaflora-yielding formations of Gondwana (similar charts for Laurasia are not shown). Particularly interesting is how different basins across Gondwana are productive at different intervals (periods, epochs, stages). Emphasis in this regard is given to Western Gondwana: South America and Southern Africa. We then look at a series of phylogenetic charts set to the same scale based on the Standard Geological Time Scale. These cover the plants, insects and tetrapods, the three major terrestrial groups preserved in the fossil record. We finally look at a synthesis showing biodiversity trends in the unfolding gymnosperm evolution over the past 370 million years. Very informative patterns, related to the five previous global extinction events, for instance, emerge. Yet we need to remain

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 2 fully aware that these patterns, and any conclusions drawn from them, are dependent on our current best attempts at geological correlations. This workshop is highly welcomed as a platform for seeking synergy in improving these correlations—and hence improving our holistic view of Earth history.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 3 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

A summary of the geochronology and Precambrian crustal architecture of southern Africa, and possibilities for correlations with South America

Richard A. Armstrong1, Maarten de Wit2 ¹ Research School of Earth Sciences, College of Science, The Australian National University, Canberra 0200, ACT, Australia, [email protected], ² AEON and Department of Geological Sciences, University of Cape Town, South África, [email protected]

Like most large continents, the earliest preserved crustal history of southern Africa comprises an ancient, relatively small kernel of Archaean rocks of granite-gneiss-greenstone association. This earliest building block grew by accretion and amalgamation into a stable cratonic block with attached keel of relatively cool and resilient lithospheric mantle. The rock record and rate of development preserved in each craton will be unique, but in rare occasions there may be similarities with other old continental blocks, allowing at least some correlations to be made in terms of age, earth dynamics, evolution and tectonics processes. In these instances it might be possible to speculate about reconstructing ancient continental geometries and configurations, but the geological records are often too fragmentary or modified to allow this. Younger reconstructions become more feasible with a more complete (and usually less complicated) geological record and more reliable geochronological and palaeomagnetic data. Southern Africa has a long 3.6 Ga geological history which is to a large degree remarkably well-preserved, especially with respect to its earliest development. As a result it has some of the most intensely studied areas in the world, many of which have become models for Earth evolution. The Archaean and Palaeoproterozoic record is preserved in two large cratons, viz. the southern Kaapvaal craton and the adjoining Zimbabwe Craton further to the north and north east of the subcontinent. The oldest core of the subcontinent is found

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 4 in the 3.56 – 3.4 Ga greenstone and granite-gneiss sequences of the Ancient Gneiss Complex and the Barberton Mountain Land on the present-day eastern margin of the Kaapvaal Craton. Evidence for an even older history is fragmentary and based mainly on dating of detrital or xenocrystic zircons, or on Sm-Nd isotope studies. The oldest zircons thus far detected are ~3.94 Ma (Armstrong, unpubl. data), but other subpopulations at 3.7 – 3.6 Ga are more common. The earliest substantial terrestrial sediments in the Barberton Mountain Land, the Moodies Group, were largely fluvial-marine clastic sediments deposited some 3.2 Ga ago. The Kaapvaal Craton appears to have grown signiificantly over the next several hundred million years through processes of accretion and collision of arcs and micro-continents (Wit et al, 1992; Schmitz et al, 2004; Eglington and Armstrong, 2004) forming a broad, stable, true continental crustal platform on which large sedimentary basins were deposited over the next billion years and below which a cratonic keel developed allowing for the formation and preservation of some of the Earth’s most significant diamond resources. These sedimentary basins formed over extraordinarily long times (up to 600 Ma), and have enormous economic and scientific significance. This extended sedimentary record is nearly continuous across the craton, starting at ~3.07 Ga with the auriferous Witwatersrand Supergroup and its proposed (Beukes and Cairncross, 1991) easterly equivalent, the Mozaan Group of the Pongola Supergroup. Deposition of the Witwatersrand Supergroup was terminated by one of the many episodes of extensive volcanic activity (LIPs) that punctuated the history of the craton. The Ventersdorp Supergroup is a major volcano-sedimentary sequence that started 2.72 Ma ago as massive outpourings of ultramafic and mafic lavas under initial compressional and then extensional tectonic regimes. This was followed by an extended period of marine and shelf- facies sedimentation when almost the entire craton was submerged. The various parts of the Transvaal mega-basin preserve an almost unbroken geological history from ~2650 Ma to the intrusion of the Bushveld Complex 2055 Ma ago (Armstrong, Kamo and Harmer, in prep.). Included in this sequence are some important golden spikes in the mapping of the Earth’s evolution, one being the oxidation of the Earth’s atmosphere to the point where the first oxidised terrestrial red beds occurred. From a correlation point of view, much of the stratigraphy of the Transvaal Supergroup is almost exactly replicated in the Hamersley Province of the Pilbara Craton, Australia, suggesting that there was a physical connection between these important continental blocks in the Palaeoproterozoic. Similarities in age between the rich banded iron formations of these two provinces and those of the Quadrilátero Ferrífero, of Minas Gerais, Brazil could suggest an even wider connection between continents during the Palaeoproterozoic, although it is yet to be determined whether there were physical connections, or simply coeval developments at a critical time in Earth history. The emplacement of the Bushveld Complex (and its associated extrusive felsic

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 5 rocks) marked the end of the Transvaal mega-basin, but significant deposits of terrestrial sediments and the earliest red beds were deposited immediately after cessation of the Bushveld volcanism. These sediments and subordinate volcanic rocks of the Waterberg, Soutpansberg and other sub-basins continued for some 200 million years and marked the end of a long period of intracratonic basin development, but also some significant tectonic activity on the western margins of the Kaapvaal Craton and the Zimbabwe Craton. The southeastern and western margins of the Kaapvaal Craton are flanked by two Mesoproterozoic metamorphic belts – the Natal terrane and the Namaqua terrane respectively. Even though they are separated by extensive Phanerozoic cover, early workers (Nicolaysen and Burger, 1965) suggested a link between the two terranes. Subsequent geophysical (De Beer and Meyer, 1984) and geochronological investigations in deep boreholes penetrating the younger cover (Eglington and Armstrong, 2003) have confirmed that this is a continuous belt, now recognised as the Namaqua-Natal Belt or Metamorphic Province. Out-board correlations are possible with similar lithologies in Antarctica and the Falkland/Malvinias Islands, and further extensions and/or correlations of similar age are found in Namibia, Botswana and Zambia. Although commonly referred to as “Kibaran” in age, the various components of the Namaqua-Natal metamorphic belt are significantly younger than the classic Kibaran rocks of central Africa. The Namaqua-Natal Belt has been subdivided by various researchers into several subdomains, based mainly on structural, tectonic and limited geochronological criteria. The past decade has seen a significant expansion in the geochronological database for the regions through a number of focussed U-Pb SHRIMP dating projects, and this has allowed correlations to be made both regionally and on a broader scale (see Eglington, 2006, and McCourt et al., 2006, for the most recent published summary of the geochronology and literature). The Natal belt is a composite arc terrane comprising juvenile granitoid gneisses, gneisses and intrusive rocks showing a limited range of emplacement ages between ~1235 – 1025 Ma (McCourt et al, 2006). Model Rb-Sr and Sm-Nd ages and direct dating of inherited zircons and detrital zircons from supracrustal rocks suggest only minor involvement of older crustal material, possibly with a maximum age of 1.4 Ga, although recent discovery of ~1.7 Ga detrital zircons may point to incorporation of some limited older crust. The Natal belt amalgamated during two accretionary events, the older (between ~1208 and ~1155 Ma) marking the time of accretion onto the southern margin Kaapvaal craton within the Tugela terrane (McCourt et al., 2006). The Namaqua Belt occupies a broad area of southern Namibia and northwestern South Africa stretching from the Kaapvaal Craton in the east to the Atlantic Ocean in the west. It is a complex area with multiply-deformed greenschist-to granulite-grade (low pressure, high temperature) gneisses, and supracrustal rocks and has been subdivided into a number of tectono-stratigraphic terranes, often divided by large transcurrent faults and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 6 shear zones. In contrast to the Natal Belt, the Namaqua Belt has an extended history ranging from the ~2.0 Ga Palaeoproterozoic rocks of the Richtersveld terrane to the extensive Mesoproterozoic metamorphism of the Namaqua orogeny (~1180-1030 Ma, with possible further subdivisions of different events within this time interval; Robb et al., 1999, Clifford et al., 2004). As is the case for the Natal section of the Namaqua-Natal Belt, SHRIMP geochronological studies have been able to unravel complex tectono-metamorphic histories for this region that can be used as fingerprints for possible correlations with other areas within Gondwana. The oldest rocks in the region are found in the low-grade Richtersveld terrane, situated between Namibia and South Africa. Here pre-Namaquan ~ 2 Ga volcanic and sedimentary rocks are intruded by the 1.8 Ga Orange River Group (Reid, 1979; Reid and Armstrong, unpubl. data). Similar ages have been found in the Gladkop Suite in the O’okiep region of western Namaqualand (Robb et al., 1999). The Kheis Terrane on the western margin of the Kaapvaal Craton comprises thick successions of volcanic and sedimentary rocks deposited along a passive continental margin, with sediments derived from the interior of the craton (van Niekerk, 2006). Deposition ages are between ~1.8 – 2.0 Ga, with the western margin of the craton affected by at least two tectonic events, the later collision event (the “Kheis Orogeny”) bracketed between 1290 Ma (Armstrong and Moen, unpubl. data) and 1170 Ma, just prior to the Namaqua orogeny, and not a Palaeoproterozoic event as previously assumed. The western limit of the Namaqua province is marked by the Gordonia sub-province, comprising mainly juvenile mafic igneous rocks (1.28 and 1.08 Ga, Cornel, et al. 1990; Bailie et al, 2007, in prep.), intrusive granitoids and metasediments. Detrital zircons studies of the sediments show no detritus was derived from the Kaapvaal Craton (van Niekerk, 2006). The Bushmanland sub-province has long been assigned an age of >1.6 Ga, based on a single Sm-Nd date and Pb-Pb model ages from the base metal deposits of the Aggeneys- Gamsberg region. Several recently completed geochronological studies have shown that the basement and the supracrustal rocks are significantly younger (e.g. Raith et al., 2003; Bailie et al., 2007a, b). New lithostratigraphic studies and detrital zircon dating have also shown that the metasediments can be correlated over a wide area and have maximum deposition ages between 1650 Ma and 1130 Ma (McClung, 2006). Isotopic studies and detrital and inherited zircon dating have shown that the crustal development of the region did not begin before ~ 2 Ga. The post-Namaquan history of southwestern Africa includes a number of events or features pertinent to Gondwana reconstructions. These include the Neoproterozoic Gariep Belt and the sediments and volcanic rocks of the Saldanian Belt and their intrusive granites. Absolute ages are essential for true and accurate correlations to be made, but provenance studies, e.g. based on detrital zircon ages, are also valuable in demonstrating the coeval development of basins and possible stratigraphic and tectonic correlations. Basei et al.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 7 (2005) used this technique to demonstrate that the sediments of the Rocha Group of the Punte del Este Terrane (Dom Feliciano Belt) in Uruguay and the Gariep Belt in southern Africa shared a similar provenance and were thus very likely to have formed in the same basin and can thus be directly correlated. The system of Pan-African orogenic belts that are found through much of Africa is represented in the southwest by the low-grade Saldanian Belt. It crops out as a number of poorly exposed inliers within the Cape Fold Belt (Rozendaal et al., 1999). Dating of some of these inliers has shed new light on the ages of these volcano-sedimentary sequences (Barnett, et al., 1999; Armstrong et al, 1998) and those of the I- and S-type granites of the Cape Granite Suite that intrude the belt (e.g. Armstrong, 1998, Da Silva, 2000, 2005, Scheepers and Armstrong, 2002). A detailed chronology of these granites shows an extended span of granitic magmatism in the Saldanian belt from ~552 – 519 Ma. Also of interest from these geochronological studies is the observation that the basement in the southern extremity of Africa is not Pan-African as previously assumed, but most likely Namaquan (1.0 -1.3 Ga) – an important point when cross-Atlantic correlations are made?

References Armstrong, RA., de Wit, MJ., Reid, D., York, D. & Zartman, R. 1998. Cape Town’s Table Mountain reveals rapid Pan-African uplift of its basement rock. Journal of African Earth Sciences, v. 27-1A, p.10-11.

Bailie, R., Armstrong, R. and Reid, D. 2007. The Bushmanland Group supracrustal succession, Aggeneys, Bushmanland, South Africa: Provenance, age of deposition and metamorphism. S.A. Journal of Geol., v. 110, p. 59-86.

Bailie, R., Armstrong, R. and Reid, D. 2007. Composition and single zircon U-Pb emplacement and metamorphic ages of the Aggeneys Granite Suite, Bushmanland, South Africa. S.A. Journal of Geol.,v. 110, p. 87-110.

Basei, M.A.S., Frimmel, H.E., Nutman, A.P., Preciozzi, F. and Jacob, J. 2005. A connection between the Neoproterozoic Dom Feliciano (Brazil/Uruguay) and Gariep (Namibia/South Africa) orogenic belts – evidence from a reconnaissance study. Precambrian Res.,v. 139. p. 195-221.

Clifford, T.N., Barton, E.S. and Stern, R.A. 2004. U-Pb zircon calendar for Namaqua (Grenville) crustal events in the granulite facies terrane of the O’okiep Copper District of South Africa. J. Petrol., v. 45, p. 669-691.

Cornel, D.H., Kröner, A., Humphreys, H.C. and Griffin. G. 1990. Age of origin of the polymetamorphosed Copperton Formation, Namaqua-Natal Province, determined by single grain zircon Pb-Pb dating. S.A. Journal of Geol.,v. 93, p. 709-716.

Da Silva, L.C., Gresse, P.G., Scheepers, R., McNaughton, N.J., Hartmann, L.A. and Fletcher, I. 2000. U-Pb SHRIMP and Sm-Nd age constraints on the timing and sources of the Pan-African Cape Granite Suite, South Africa. Journal of African Earth Sciences, v. 30, p. 795-815.

Da Silva, L.C., McNaughton, N.J., Armstrong, R., Hartmann, L.A. and Fletcher, I.R. 2005. The neoproterozoic Mantiqueira Province and its African connections” a zircon-based U-Pb geochronologic subdivision for the Brasiliano/Pan-African systems of orogens. Precamb. Res., v. 136, p. 203-240.

De Beer, J.H. and Meyer, R. 1984. Geophysical characteristics of the Namaqua-Natal Belt and its boundaries. J. Geodyn., v. 1, p. 473-494.

Eglington, B.M. 2006. Evolution of the Namaqua-Natal Belt, southern Africa – A geochronological and isotope geochemical review. Journal of African Earth Sciences, v. 46, p. 93-111.

De Wit, M.J., ., Roering, C., Hart, R.J., Armstrong, R.A., de Ronde, C.E.J., Green, R.W.E, Tredoux, M., Pederdy, E., and Hart, R.A., 1992. Formation of an Archaean continent. Nature, v. 367, p. 553-562.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 8 Eglington, B.M and Armstrong, R.A. 2003. Geochronological and isotopic constraints on the Mesoproterozoic Namaqua-Natal Belt: evidence from deep borehole intersections in South Africa. Precamb. Res., v. 125, p. 179-189.

McClung, C.R. 2006. Basin analysis of the Mesoproterozoic Bushmanland Group on the Namaqua Metamorphic Province, South Africa. PhD Thesis (unpubl.), University of Johannesburg, p. 307.

McCourt, S., Armstrong, R.A., Grantham, G.H and Thomas, R.J. 2006. Geology and evolution of the Natal Belt, South Africa. Journal of African Earth Sciences, v. 46, p. 71-92.

Nicolaysen, L.O. and Burger, A.J. 1965. Note on an extensive zone of 1000 million-year old metamorphic and igneous rocks in Southern Africa, Sci. Terre, v. 10, p. 487-516.

Robb, L.J., Armstrong, R.A., and Waters, D.J. 1999. The history of granulite facies metamorphism and crustal growth from single zircon U-Pb geochronology: Namaqualand, South Africa. J. Petrol., v. 40, p. 1747-1770.

Scheepers, R. and Armstrong, R. 2002. New U-Pb SHRIMP zircon ages of the Cape Granite Suite: implications for the magmatic evolution of the Saldanian Belt. S.A. Journal of Geol., v. 105, p. 241-256.

Van Niekerk, H.S. 2006. The origin of the Kheis Terrane and its relationship with the Archean Kaapvaal Craton and the Grenville Namaqua Province in southern Africa. PhD Thesis (unpubl.), University of Johannesburg, p. 402.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 9 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Precambrian tectonic domains of São Paulo, Paraná and Santa Catarina States, South Brazil: identification of major neoproterozoic suture zones

Miguel Angelo Stipp Basei 1; Oswaldo Siga Junior1; Claudia Regina Passarelli1 ¹ Instituto de Geociências-USP, Rua do Lago 562, CEP-05508-080, SP, Brasil

Abstract In southeast South America magmatic, metamorphic, structural and geotectonic features record the superposition of Neoproterozoic (Brasiliano)-Eopaleozoic (Rio Doce) orogenies. The present geometry of these geotectonic units reflects collages of distinct terranes, culminating with the amalgamation of the Gondwana supercontinent. The Brasiliano orogeny that took place in the region (850-620Ma) is documented by remnants of magmatic arcs and metavolcano-sedimentary covers well represented by the Ribeira and Dom Feliciano Belts. The volcano- sedimentary basins and alkaline-peralkaline granitoids of the Serra do Mar suite (590 ± 10 Ma) mark the end, in extensional regime, of events related to the Brasiliano Cycle. The foreland-type basins (Itajaí, Camaquã) with sedimentation around 570 ± 20 Ma and deformation around 530 Ma, represent the main volcanic-sedimentary record of the Rio Doce orogeny in southern Brazil. The mosaic of tectonic blocks identified from southern São Paulo to northern Santa Catarina States is composed of five major units: southern Ribeira Belt, Curitiba and Luis Alves Microplates, and northern Dom Feliciano Belt. Lancinha-Itariri, Piên and Major Gercino shear zones represent the main neoproterozoic sutures observed in the region.

Geological Setting Southern Branch of the Ribeira Belt The southernmost part of the Ribeira Belt is characterized by a series of NE-SW trending domains with Meso-Neoproterozoic supracrustal rocks at low metamorphic grade, intruded by Neoproterozoic granitic batholiths and stocks (Campanha & Sadowski 1999; Campos Neto, 2000; Janasi et al., 2001; Prazeres 2005). Most of these domains are separated by shear zones that have marked dextral horizontal displacements. This represents a polycyclic evolution, during which the Meso- and Neoproterozoic supracrustal units became juxtaposed. Neoproterozoic tectonothermal evolution is marked by northwestwards vergence towards the Paranapanema Craton (Mantovani & Brito Neves

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 10 2005). In contrast, Neoproterozoic transport in the southeastern portion of the belt was towards the Curitiba domain. From northwest to southeast, five major supracrustal belts are recognized: Itaiacoca (shallow-water platformal deposits of mostly meta-arkose and dolomites and subordinate felsic metavolcanic rocks intruded by the Cunhaporanga batholith), Água Clara (metacalcarenite, micritic metalimestone, metacalcisiltite, calc-schist and subordinately mica schist and iron formation intruded by the Três Córregos Batholith), Lageado (shallow-water platformal deposits composed of metaconglomerate, metasandstone, metarhythmite and calcitic-marble), Votuverava-Perau (composed of quartzite, rhythmic laminated phyllite, metasiltite, metaconglomerate, dolomitic marble, graphitic quartzite beds and metabasite lenses interfingered with volcaniclastic rocks and iron-manganese formations) and Capiru domain (composed of dolomitic marble, quartzite, and subordinated phyllite that comprises a metasedimentary cover of the Curitiba Microplate, south of the Lancinha-Itariri Suture Zone).

Luis Alves and Curitiba Microplates The terranes that constitute the basement of the northern part of southern Brazil, from Ribeira to Dom Feliciano fold belts, can be grouped in two distinct geotectonic units: Luís Alves and Curitiba microplates (Basei et al., 1998) representing preexistent continental fragments totally involved by the processes associated with ocean closing between Africa and South America during the formation of the Gondwana supercontinent. These domains are separated by the NE-trending Pien Suture Zone, extending along the southern limit of a Neoproterozoic calc-alkaline granitoid belt generated in the southern border of the Curitiba Microplate, as a consequence of the northwestward subduction of the oceanic crust between these geotectonic units. An important landmark in the evolution of the Luis Alves and Curitiba domains is represented by expressive non-deformed granite genesis of alkaline–peralkaline nature (Serra do Mar Suite) as well as by intense volcanism, related to the development of Corupá, Campo Alegre and Guaratubinha extensional basins. The main interval (590 +/-10 Ma) of the formation of these rocks is restricted to the final stages of the Brasiliano and the begin of Rio Doce orogenesis. The northern portion of the Curitiba Microplate is overlain by metasediment sequences (Capiru and Setuva formations), which have paleogeographic affinities with it. The gneissic-migmatitic basement rocks that occur in this microplate (Atuba Complex) is constituted by banded migmatitic amphibole gneisses, of Paleoproterozoic age (2.1Ga), which underwent intense deformation and migmatization during the Neoproterozoic. Whole rock Rb-Sr isochrons and K-Ar cooling ages are always Neoproterozoic. On the other hand, Nd model ages for metamorphic rocks are predominantly Archean. Within these gneisses, nuclei of high metamorphic-grade rocks of predominantly charno-enderbitic composition occur. These rocks are preferentially distributed in the northern portion of these terranes. The

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 11 best exposition occurring in the Peruíbe region (Itatins Massif), southeastern São Paulo State. The Paulo Leminski Quarry, in Curitiba City, is another site of excellent expositions. The Lancinha-Itariri Suture Zone separates the Curitiba Microplate from the Ribeira Belt. The Luís Alves Microplate is characterized as a crustal segment where its basement of regional expression is the only crustal block of the Brazilian southeastern sector that had already been cold since the Paleoproterozoic, not undergoing Neoproterozoic tectonothermal superposition. The high-grade rocks are predominantly depleted orthoderived charno-enderbites (Basei et al., 1998).

24º PB Pr C Cn Pe ru íb e RB C

TC Cp Pr N SZ LI P So u t h Cp Am e ric a C Pa ra n a g u á 200 km

Z Pr 50º P S 26º P LA

Br PB SZ G M Flo ria nó p o lis

28º N F A

O 56º 54º DFB Tertiary and Quaternary covers PB Paraná Basin - Palaeozoic sediments Porto Ale gre Neoproterozoic Magmatic Arc s 30º Cn - Cunhaporanga; TC- Três Córregos PB Pr - Paranaguá; P - Piên Pe F - Florianópolis; Pe - Pelotas; A - Aiguá SG B Neoproterozoic Belts B Po R RB - Sou U AZ T thern Ribeira Belt RU IL Z G S DFB - Dom Feliciano Belt U C AY (Br- Brusque; Po- Porongos; La - Lavalleja) R - Ro c h a Be lt R Pe 32º Neoproterozoic terranes C SGB - São Gabriel Block I PB T PET - Punta del Este Terrane

N Palaeoproterozoic-Archaean terranes

A C - Curitiba Terrane L LA - Luis Alves Terrane NP T SZ Rio de La Plata Craton: PA- Piedra Alta; SC A

NP- Nico Perez; R- Rivera; T- Taquarembó

Z S

34º PA Y PET Neoproterozoic - Cambrian Suture Zones S La LISZ: Lancinha - Itarirí Suture Zone La Pa l o m a A R PSZ : Pien Suture Zone SZ : Major cino Z Major Tectonic Transport MG Ger Suture Zone S PA B S CSZ : Cordilhera Suture Zone Montevideo Thrust Fault SYSZ : Sarandy Del Y Suture Zone SCSZ: San Carlos Suture Zone 56º Transcurrent Shear Zones SBSZ : Sierra Ballena Suture Zone

Figure 1 – Distribution of the main tectonic units of the southern Brazil and Uruguay (modified from Basei et al. 2007).

The existence of two Paleoproterozoic high-grade metamorphic events is suggested. The first (2.35Ga) predominates in the Luís Alves Microplate, generating the Santa Catarina Granulitic Complex, the second (2.10Ga), occurs along the northern portion of the Curitiba Microplate being represented by a discontinuous orthogranulitic belt (mangerites to charnockites).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 12 Northern branch of Dom Feliciano Belt In Santa Catarina State (northernmost segment of the 1,200 km-long Dom Feliciano Belt) three tectonic domains can be found (Fig. 1). These are described from ESE to NNW: (i) Eastern Granitoid Belt, which comprises the calc-alkaline Florianópolis Batholith and represent the roots of a Neoproterozoic magmatic arc; (ii) Supracrustal Schist Belt, composed by the Brusque Group, characterized by multiply folded low-grade metavolcanosedimentary rocks with northwestward tectonic transports, intruded by several generations of late- to post-tectonic granitoids that developed contact metamorphic aureoles; (iii) Foreland basin, represented by a thick pack of sedimentary (mostly turbidites) rocks filling the Itajaí Basin, less affected by deformation and metamorphism than the adjacent Schist Belt.

Tectonic Model Several models for tectonic correlation across the southern Atlantic Ocean were proposed with the later ones (for example: Frimmel & Folling 2004; Valeriano et al. 2004 and Basei et al. 2005, 2007) within the framework of West Gondwana assembly. The available

Nd (Tdm) ages for a roughly W-E transect across the units observed on both sides of the Atlantic Ocean show that there is a conspicuous decrease of model ages eastwards; with the Damara Belt displaying the youngest values. The basement of Curitiba and Luis Alves microplates shows Archean model Nd ages. There is a concentration of model ages around 2.0 Ga for the Schist Belt of the Dom Feliciano Belt, whereas for the Granite Belt the average falls to 1.3-1.6 Ga. For the Damara Belt (mainly its granitoids), the younger values do not exceed 1.1 Ga, but the average also falls in the 1.3-1.6 Ga interval. Detrital zircon ages decreasing from west to east can also be observed. The 0.9-1.2 Ga detrital zircons are lacking in the southern Ribeira Belt, are rare in the Dom Feliciano Belt but dominate in the African belts. This indicates an African source for such young zircons. Figure 2 shows hypothetical NW-SE section from the Ribeira to the Gariep Belt, where figure 2a is a representation at c. 0.61 Ga, when subduction zones were active with magmatic arcs being generated. These arcs are now represented by the Cunhaporanga and Três Córregos batholiths in the Ribeira Belt, the Piên batholith between the Curitiba and Luís Alves terranes, and the Florianópolis-Pelotas-Aiguá batholith between the Dom Feliciano and Gariep/Damara belts at the African side. In the proposed tectonic model the Lancinha –Itariri lineament represents a suture zone related to the subduction of an oceanic crust towards NW preceding the juxtaposition of Curitiba Microplate and Paranapanema craton. Similar process is suggested by the generation of Pien Suture Zone that marks the passage between Curitiba and Luis Alves microplates. The Major Gercino–Sierra Ballena lineament should be viewed as a Neoproterozoic suture (Passarelli, 1996; Basei et al. 2005) separating the Granite Belt

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 13 (terranes of African affinities) from those further west (South American affinities) represented by terranes that are geologically, geochronologically and isotopically distinct. An important gravimetric anomaly along the lineament also supports this interpretation (Hallinan & Mantovani 1993).

A 610 Ma

3C P FP RB Cp DFB GB

LA K P C

LISZ PSZ M G SBSZ 3C P DFB FP 530 Ma B RB GB P C LA K

Magmatic Arcs Metasedimentary cover Cratons and Microplates Su tu re Zo n e s 3C -Trê s Có rre g o s RB - Ribeira Belt P - Paranapanema Craton LISZ - Lanc inha-Ita riri P- Piên Cp - Capiru Formation C - Curitiba Microplate PSZ - Piê n FP-Florianópolis DFB - Dom Feliciano Belt LA - Lu i s Alv e s Mi c ro p l a t e MGSBSZ - Ma jor Gerc ino-

GB - Gariep Belt K - Kalahari Craton Sierra Ballena

Figure 2– Simplified tectonic model emphasizing the main geologic units that were juxtaposed during the collisions associated with the formation of Gondwana. 2A.- situation before collisions; 2B – after collisions. (Basei et al. 2007)

The westward collision between the Florianópolis-Pelotas-Aiguá batholiths and the Dom Feliciano supracrustal belt occurred at c. 0.60 Ga. However, on the African side, only at c. 545 Ma did eastward-directed nappes and regional metamorphism affect the supracrustal units (Frimmel & Frank 1998). This event was expressed on the South American side by reactivation of 0.60 Ga structures, and deformation in the foreland basins (Itajaí, Camaquã and Arroio del Soldado). The deformation of African foreland basins (e.g. Nama) also started at around 0.54 Ga (older units) but continued through the Cambrian. Figure 2b represents the tectonic situation at c. 0.53 Ga, just after the collisions during Gondwana assembly, and displays the location of the major suture zones. The terranes east of the Major Gercino-Sierra Ballena lineament (proposed suture) are interpreted as remnants of African terranes that were juxtaposed during the formation of West Gondwana. After opening of the South Atlantic, only small parts of these belts were preserved in South America.

References Basei, M.A.S., McReath, I., Siga JR. O. 1998. The Santa Catarina Granulite Complex of Southern Brazil: A Review. Gondwana Research, 1(3-4): 383-391.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 14 Basei M A S., Frimmel, H E., Nutman, A P., Preciozzi. F, Jacob, J. 2005. The connection between the Neoproterozoic Dom Feliciano (Brazil/Uruguay) and Gariep (Namibia/South Africa) orogenic belts. Precambrian Research, 139 (3-4): 195-221.

Basei M A S., Frimmel, H E., Nutman, A P., Preciozzi. F. 2007 - West Gondwana amalgamation based on detrital zircon ages from Neoproterozoic Ribeira and Dom Feliciano belts of South America and a comparison with coeval sequences from southwestern Africa. Journal of Geological Society of London (in press).

Campanha, G. A. C., Sadowski, G. R. 1999. Tectonics of the Southern Portion of the Ribeira Belt (Apiaí Domain). Precambrian Research, 98(1): 31 - 51.

Campos Neto, M.C., 2000. Orogenic Systems from Southwestern-Gondwana: an approach to Brasiliano-Pan African Cycle and orogenic collage in southeastern-Brazil. In: Cordani, U. G.; Milani, E. J.; Thomaz-Filho, A.; Campos, D. A. (eds): Tectonic Evolution of South America. 31st International Geological Congress, Rio de Janeiro, Brazil p. 335-365.

Frimmel, H.E & Frank, W. 1998. Neoproterozoic tectono-thermal evolution of the Gariep Belt and its basement, Namibia/South Africa. Precambrian Research, 90, 1-28.

Frimmel, H.E., Fölling, P.G., 2004. Late Vendian closure of the Adamastor Ocean: Timing of tectonic inversion and syn-orogenic sedimentation in the Gariep Basin. Gondwana Research, 7 (3): 685-699.

Halinann, S.E.& Mantovani, S.M.M. 1993. Structural Framework of the Southern Brazilian Shield: the perspective from the Gravity models. III International Geophysical Congress of the Brazilian Geophysical Society, Anais, SBGF, Rio de Janeiro, 2: 1078-1083.

Janasi, V. A., Leite, R. J., Van Schmus, W. R. 2001. UPb chronostratigraphy of the granitic magmatism in the Agudos Grandes Batholith (west of São Paulo, Brazil): implications for the evolution of the Ribeira Belt. Journal of South American Earth Sciences, 14 (4): 363-376.

Mantovani, M. S. M. & Brito Neves, B. B. de 2005. The Paranapanema Listhospheric Block: Its importance for Proterozoic (Rodinia, Gondwana) Supercontinent Theories. Gondwana Research, 8 (3): 303-315.

Passarelli, C.R. 1996. Análise estrutural e caracterização do magmatismo da Zona de Cisalhamento Major Gercino, SC. São Paulo. M.Sc. Thesis. Institute of Geosciences, University of São Paulo, 179 pp.

Prazeres Filho, H.J. 2005. Caracterização Geológica e Petrogenética do Batolito Granítico Três Córregos (PR- SP): geoquímica isotópica (Nd-Sr-Pb), idades (ID-TIMS / SHRIMP) e 18O em Zircão. PhD. Thesis. Institute of Geosciences University of São Paulo, 207p.

Valeriano, C.M., Machado, N., Simonetti, A., Valladares, C.S., Seer, H.J., & Simões, L.S. 2004. U-Pb Geochronology of the Southern Brasilia belt (SE Brasil): Sedimentary Provenance, Neoproterozoic orogeny and assembly of Western Gondwana. Precambrian Research, 130 (1-4): 27-55.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 15 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Some structural and stratigraphic enigmas of the eastern Cape Fold Belt, South Africa

Peter Booth Nelson Mandela Metropolitan University, Port Elizabeth, South Africa, [email protected]

Abstract Recent research on the eastern part of the Cape Fold Belt has yielded new information on thrust fault patterns which have to be factored into interpretations of the stratigraphy of this part of the fold belt. Enigmas such as anomalously thick arenaceous sequences in the Cape Supergroup, and the omission of the Cedarberg Shale Formation in the eastern part of the fold belt are explained by taking the structural characteristics of the fold belt into account. Further enigmas that still require satisfactory explanations include the irregular variation in structural style from south (hinterland) to north (foreland) through the fold belt, as well as the perception that the eastern part of the fold belt is more deformed than the western part.

Introduction Strata of the Cape Supergroup which form part of the Cape Fold Belt crop out in the southern part of South Africa over a distance exceeding 1000 km (Fig.1). The basement to the Cape Supergroup includes 700 to 500 million year old metamorphosed sediments of the Malmesbury, Kango, Kaaimans and Gamtoos Groups. Basement rocks occur in juxtaposition with the cover rocks. This situation is the product of large displacements along normal faults which developed at the break-up of Gondwana, during the late Mesozoic. Unconformably overlying the Cape Supergroup is the Karoo Supergroup whose lower and southernmost sediments were affected by the Cape Orogeny some 300 Ma ago. The approximately 12 000 m thick stratigraphic sequence of cover rocks (Cape Supergroup) is divided into three units viz. the Table Mountain Group (TMG) at the base, followed by the Bokkeveld Group and, at the top of the sequence, the Witteberg Group (Shone & Booth, 2005). All three groups of rocks are composed essentially of quartzites and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 16 shales. These rocks represent arenaceous and argillaceous sediments deposited in a continental shelf setting, between 500 and 330 Ma (Tankard et al., 1982). Deformation and metamorphism of the sediments to greenschist facies took place during the late Palaeozoic (about 300 Ma). Each of the three groups is differentiated from one another by the variable proportions of quartzite to shale which in turn gives rise to characteristic topographic and stratigraphic features where each group of rocks is easily recognisable in the field. The fold belt has two distinct “arms”, one in the west where strata and structures trend approximately north-south, and the other in the south which has an overall east-west trend.

Witteberg Group Bokkeveld Group Table Mountain Group

Steytlerville #

Cape # Port Elizabeth Tow n #

0 200 400 Kilometres N

Figure 1 - Map showing distribution of the Witteberg, Bokkeveld and Table Mountain Groups (Cape Supergroup), southern South Africa.

Stratigraphic enigmas A question that all researchers working on the Cape Fold Belt have yet to answer satisfactorily is, “Are we confident about correlating strata of the Cape Supergroup over their vast area of outcrop from the western part of the fold belt to the eastern part?” The Council for Geoscience have produced excellent maps on a scale of 1:250 000 showing continuity of strata across the entire fold belt, i.e. strata are correlated on a macroscale. However, on a more detailed scale, there are certain stratigraphic enigmas that require explanation. The first of the enigmas is the occurrence of anomalously thick TMG strata in an east-west trending trough near the southern part of the fold belt (Rust, 1973). Another anomaly is the omission of the Cedarberg Shale Formation in the extreme eastern part of the fold belt (this formation is a prominent time marker horizon occurring over the major portion of the fold belt).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 17 Structural enigmas Prominent discordant structural features are evident where the two arms of the fold belt meet in the south west known as the syntaxis area. The latter represents an area of interference folding and convergence of fault structures (de Beer, 1992). The western arm of the fold belt has north-northwest trending open folds, whereas the southern arm displays a structural style of variable character from south to north. Throughout the latter area recumbent folds are common in the south, but open and gentle folds are the norm in the northern part of the fold belt. Thrust faulting is also prominent in the southern arm, with a greater abundance of thrust sheets occurring towards the eastern part of the fold belt, whereas thrust faulting appears to be absent in the western arm of the fold belt. Within the Cape Supergroup rocks the structural style is also variable from the base to the top. In the extreme southern outcrop of the central part of the southern arm basement rocks have been thrusted up and northward so that they occur in juxtaposition with the cover rocks. This location is interpreted by Paton et al., (2006) as the dividing line between thick- skinned structural characteristics in the south of the fold belt and thin-skinned characteristics to the north. In the lowermost unit of the cover rocks (TMG) thrust faults within most thrust sheets are closely spaced (of the order of one to two metres apart). Thrusting generally post-dates the major folding event. Outcrops of thrust sheets occur over a strike distance of some 250 km (between Port Elizabeth and Uniondale to the west). Thrust stacking of one quartzite unit upon another is characteristic of these thrust sheets (Fig.2), and thrusting has been so intense that shale beds are largely eliminated within the thrust sheets (Booth & Shone, 1992). Small remnants of shale (and occasionally talc schist) caught between quartzite beds bear evidence of their former existence as argillaceous beds in the original stratigraphic sequence. In some places such as at Port Elizabeth thrust stacking is interpreted as having developed in a break-back sequence, whereas at Uniondale the sequence of thrusting is more complex (Booth and Shone, 1999). In the latter area pop-up structures are common, although the general deformation pattern remains much the same as that recorded in TMG rocks at Port Elizabeth. In general, in both the Port Elizabeth and Uniondale areas the thrust sequence shows a hinterland propagation,,i.e. towards the south (Fig. 3). Although not much work has been carried out on rocks of the Bokkeveld Group, thrusting is evident in black shales and quartzites where thrust sequences display a break– back pattern (Booth et al., 2004). In rocks of the Witteberg Group there is a much closer relationship between folding and thrust faulting than in other units of the Cape Supergroup (Booth, 1996; 1998). At Steytlerville, north-west of Port Elizabeth, thrusting is closely spaced in shale beds of the

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 18 lowermost units, but higher up in the stratigraphic sequence intricate cut-off relationships show that in general thrusts developed progressively towards the south. Fold vergence towards the north and towards the south in several units of this group has been interpreted as a product of simultaneous movement along fore- and back thrusts. Duplexing of strata, usually on a small scale, is evident in rocks of the Witteberg Group, being sporadically developed over a strike length of some 200 km in the eastern part of the fold belt (Steytlerville to Port Alfred).

Figure 2 - Thrust stacking in quartzites of the Table Mountain Group, Target Kloof, Port Elizabeth.

Figure 3 - Thrust faults in quartzites, showing sequence of thrusting in Table Mountain Group rocks, Uniondale Pass.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 19 Extent of deformation From evidence of widespread thrusting and associated folding in the eastern part of the fold belt it is apparent that the original stratigraphic order has been modified at least to some extent (Booth and Shone, 2002). How much modification has taken place has yet to be quantified. Some evidence of the scale with which we are dealing comes from initial assessments made by Hälbich et al., (1983) where they have demonstrated that as much as 70% crustal shortening has taken place in the central southernmost part the fold belt. Northwards the amount of crustal shortening is variable, but is interpreted as being dictated by heterogeneities in the substrate about which little is known. In rocks of the Witteberg Group in the central part of the fold belt Newton (1995) has shown that a displacement of some 4 km has taken place along the Florikraal thrust. In the eastern part of the fold belt south of Steytlerville the Baviaankloof thrust has displaced TMG strata 15 km northwards over strata of the Bokkeveld Group (Theron, 1969; Booth et al., 2004). Just north of Steytlerville a crustal shortening estimate of some 25 % is indicated (Booth, in prep.).

Choice of models Several workers have invoked tectonic models that could explain the structural features of the fold belt. Cascade folding in the western part of the fold belt provided the basis for Newton (1973) to advance the gravity gliding model, although this model has only local application. Lock (1980) suggested a flat plate subduction model to explain the absence of volcanic rocks in the cover sequences. Most researchers favour a plate tectonic origin for the fold belt (De Wit & Ransome, 1992). The latter authors proposed a model incorporating rotating microplates to account for the two approximately right angle trends in the syntaxis area of the fold belt. Similarly, the proposed location of a microplate in the eastern part of the fold belt during the main Cape Orogeny compressional deformation event might explain why thrust faulting and associated structures are more prevalent in the eastern part of the folds belt than elsewhere (Booth et al., 2004).

Explanation of enigmas The degree to which deformation has taken place is variable across the fold belt with some areas having undergone crustal shortening of up to 70%, although the average figure is probably of the order of 25%. Thus, over most of the outcrop area, it is justifiable to correlate strata of the Cape Supergroup on a regional basis, but some modification has to be applied when interpreting thrust sheets in certain areas where the stratigraphic order is certainly not correct. To date no attempts have been made to solve this problem in South Africa. This, however, remains a topic which needs to be addressed before correlation of strata can be made with other fragments of Gondwana.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 20 The anomalously thick sequence of quartzitic beds of the TMG is interpreted in the light of a thrust stacking model and not the product associated with Rust’s (1973) depocentre. In TMG rocks the absence of shale horizons (at least within thrust sheets) can be explained by sliding of more competent horizons (quartzites) over less competent horizons (shales). During this process deformation along thrust planes continued to such an extent that shale beds were largely eliminated, or remain as shale or talc schist fragments, along thrust planes. This mechanism would explain the very low proportion of shales within thrust sheets, and the omission of the Cedarberg Shale horizon in the eastern part of the fold belt, where deformation appears to have been greater than in the western part of the fold belt. In Witteberg Group rocks where duplexing is a relatively common feature a large scale duplex model is proposed to account for the structures observed in the upper part of the Cape Supergroup (Booth et al., 2004). Thus duplexing, as a crustal thickening mechanism, can partly explain why some parts of the stratigraphic sequence in the Witteberg Group appear to be anomalously thicker that elsewhere in the fold belt. The intimate relationship between folding and thrust faulting in the Witteberg Group confirms that both folding and faulting contribute towards crustal thickening in at least the upper sequences of the Cape Supergroup.

References Booth, P.W.K., 1996, The relationship between folding and thrusting in the Floriskraal Formation (upper Witteberg Group), Steytlerville, Eastern Cape. South African Journal of Geology, 99, 235 – 243.

Booth, P.W.K., 1998, The effect of thrusting on fold style and orientation, Weltevrede Formation (Cape Supergroup), Steytlerville, Eastern Cape. South African Journal of Geology, 101, 27 – 37.

Booth, P.W.K. and Shone, R.W., 1992, Structure of the Table Mountain Group in the Port Elizabeth area. South African Journal of Geology, 95, 29 – 33.

Booth, P.W.K. and Shone, R.W., 1999, Complex thrusting at Uniondale, eastern sector of the Cape Fold Belt, Republic of South Africa: structural evidence for the need to revise the lithostratigraphy. Journal African Earth Sciences 29, 125 – 133.

Booth, P W K and Shone, R W., 2002, A review of thrust faulting in the eastern Cape fold Belt, South Africa, and the implications for current lithostratigraphic interpretations of the Cape Supergroup. Journal African Earth Sciences 34, 189 – 200.

Booth, P.W.K., Brunsdon, G. and Shone, R.W., 2004, A duplex model for the eastern Cape Fold Belt?: Evidence from the Palaeozoic Witteberg and Bokkeveld Groups (Cape Supergroup), near Steytlerville, South Africa, Gondwana Research, 7, 211 – 222.

De Beer, C.H., 1992, Structural evolution of the Cape Fold Belt syntaxis and its influence on syntectonic sedimentation in the SW Karoo Basin. In: de Wit, M.J. and Ransome, I.G.D. (Eds.), Inversion Tectonics of the Cape Fold Belt, Karoo and Cretaceous Basins of Southern Africa. A.A. Balkema, Rotterdam, pp. 197 – 206.

De Wit, M.J. and Ransome, I.G.D., 1992, Regional inversion tectonics along the southern margin of Gondwana. In: de Wit, M.J. and Ransome, I.G.D. (Eds.) Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous Basins of Southern Africa. Balkema, Rotterdam, pp 15 – 21.

Hälbich, I.W., Fitch, F.J. and Miller, J.A., 1983, Dating the Cape orogeny. In: Söhnge, A.P.G. and Hälbich, I.W. (Eds.) Geodynamics of the Cape Fold Belt. Special Publication Geological Society of South Africa, pp 149 – 164. Lock, B.E., 1980, Flat-plate subduction and the Cape Fold Belt of South Africa, Geology, 8, 35 – 39.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 21

Newton, A.R., 1973, A gravity folding model for the Cape Fold Belt. Transactions of the Geological Society of South Africa, 76, 145 – 152.

Newton, A.R., 1995, Thrusting northwest of Laingsburg, Cape Province. South African Journal of Geology,98, 217 – 223.

Paton, D.A., Macdonald, D.I.M., Underhill, J.R., 2006, Applicability of thin or thick skinned structural models in a region of multiple inversion episodes; Southern South Africa. Journal of Structural Geology, 28, 1933 – 1947.

Rust, I.C.1973, The evolution of the Palaeozoic Cape basin, Southern Margin of Africa. In: Nairn, A.E.M., Stehli, F.G. (Eds). The Ocean Basins and Margins. Plenum Publishing Corporation, New York, 247-276. Shone, R.W. and Booth, P.W K., 2005, The Cape Basin, South Africa: A review. Journal of African Earth Sciences, 43, 196 – 210.

Tankard, A.J., Jackson, M.P.A., Eriksson, K.A., Hobday, D.K., Hunter, D.R., and Minter, W.E.L., 1982, 3.5 Billion years of Crustal Evolution in Southern Africa. Springer Verlag, New York, 532pp.

Theron, J.N., 1969, The Baviaanskloof Range - a South African nappe. Transactions Geological Society South Africa, 72, 29 - 30.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 22 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Tectonic evolution of Neoproterozoic to Eopaleozoic belts in the Southern Brazil and Southern Africa

Farid Chemale Jr. Instituto de Geociências, UFRGS, Porto Alegre-RS Brazil, [email protected]

The region comprised by the SE margin Brazil and SW margin Africa, as part of the SW Gondwana, is well defined by successive amalgamation of basement and juvenile terranes from Neoproterozoic to Early Paleozoic, resulting in the formation of several mobile belts such as Dom Feliciano (DFB), Ribeira (RB), Kaoko (KB), Damara (DB), Gariep (GB) and Saldania (SB) (e.g.: Porada, 1989; Fernandes et al., 1992; Chemale Jr., 2000). These belts were part of a continuous mountain orogenic system similar to the modern ones, separated as individual belts by the later orogenic processes and following events of denudation, Paleozoic basin formation and the Mesocenozoic rift and drift of South America and Africa. However these SE Brazilian belts and SW African belts presented some diachronic evolution that it may fit in broadly geotectonic scenarios. Compared to the African belts, the rifting process in SE Brazil occurred somewhat earlier, around 1000 to 900 Ma, which evolved to passive margin or back-arc setting (Fig 1A). The tectono-stratigraphic studies showed that a main period of island arc accretion and subsequently continental magmatic subduction (Andean-type) occurred first from 900 Ma to 700 Ma in the Dom Feliciano and Ribeira belts, which are related to the docking of microplates and Rio de la Plata Paleoplate, in the DFB, or terranes/microplates (as the Encantadas Microcontinet, see Fig. 1A and B) and the São Francisco-Congo and the Rio de La Plata paleoplates, in the RB (Fig. 1B and C). At that time, in the SW Africa region, the Kaoko, Damara and Gariep belts had a clear rifting process around 800-700 Ma with basement raging from Archean, Paleoproterozoic (Eburnean) to Mesoproterozoic (Kibaran), with the formation of larger oceans such as Adamastor and Khomas ones (Fig. 1B and C).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 23 From 700 Ma to 550 Ma, successive amalgamation processes were present in the SW African and SE Brazilian regions with the main subduction westwards (Fig 1D and 2A). In the DFB is recognized a major strike-slip belts with granitic generation from 650 to 590 Ma as result to the continental collision and escape tectonics of Rio de la Plata Plate and Encnatadas microcontinent (Fig. 2B) . In the Gariep Belt and the Kaoko Belt occurred the passive margin subduction with accretion of Archean, Paleoproterozoic, Mesoproterozoic terranes. Many exotic or suspect terranes are regoconized as the 1.7 to 1.8 Ga juvenile Mudorib magmatic arc (Luft Jr., 2005), which were involved in the collisional process during the Pan-African/Brasiliano orogeny. The main magmatic collisional events occurred from 650 to 580 Ma in the Kaoko, Gariep, Damara and Ribeira belts, during the closure of the Adamastor and Khomas oceans (Fig. 2B). All the kinematic indicators in the internal structure of the tectono-stratigraphic terranes related to the 650 to 550 Ma, Brasiliano/Pan-African orogenic process suggest an oblique closure for Adamastor Ocean with involvement of Kahalari, Congo and Rio de la Plata paleoplates. In southernmost Brazil, the magmatism formed form 650 to 550 is characterized as sin-collisional to pos-collisional, suggesting that the collisional process in the region occurred due to collision of the Encantadas Microplate and Rio de la Plata plate (Fig.1). From 590 to 540 Ma, in southern Brazil, formed continental transtensional basins (as Camaquã Basin) associated with a large displacement of the stike- slip tectonic reactivation (Fig. 2B) (Paim et al. 2000). The foreland peripheral basins such as Nama Basin developed in the final stages of continental collision (Fig. 2B and C), well recorded in SW Africa (Grotzinger et al. 1995, Gresse et al. 1996), Uruguay and Tandilla region. Close to the Cambrian-Ordovician collapse orogenic basin formation and associated magmatism with contribution of strongly modified mantle source gave rise. The Paleozoic evolution is mainly recorded by the intracratonic basin units (e.g. Paraná Basin) in most of the studied area. However, the Neopaleozoic orogeny in the Cape Belt and Sierra de la Ventana affected the margin of the SW Gondwana belts in different manners. During the Mesocenozoic rifting/drifting of Africa and South America occurred lithosphere stretching and therefore voluminous magma in the continental margin that lead to strongly modification of the continental margin with up to 4 km denudation of crustal pile (Gallagher & Brown, 1999; Chemale Jr. et al., 2005). Most of exposures of the Pan-African and Brasiliano belts and their present-day shape are strongly influenced by these later denudation processes.

References Chemale Jr., F. 2000. Evolução Geológica do Escudo Sul-Rio-Grandense. In: Holz, M. & De Ros, L.F. 2000. Geologia e Estratigrafia do Rio Grande do Sul, Editora UFRGS, Porto Alegre, 13-52.

Chemale Jr. F. and Luft Jr., J.L. 2005. Modelo tectônico para a margem SE do Brasil SW da África durante o Ciclo Brasiliano-Pan-Africano. I Simpósio sobre anomalias profundas. CENPES-PETROBRAS.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 24 Chemale Jr., F., Hadler Neto, J.C., Borba, A.W, Saenz, C.T., Santos, F.V., Luft, F. F., Ávila, J. N., Vignol, M.L., Iune, P.J., Hackkspacher, P.C., Baitelli, R. and Guedes, S. 2005 Projeto Evolter – Termocronologia por traços de fissão em apatita – Margem SE do Brasil e Uruguai. Internal Report, UFRGS-UNESP-UNICAMP- FINEP-PROFEX-PETROBRAS, Porto Alegre, 132 pp.

Gallagher, K and Brown, R. 1999. The Mesozoic denudation history of the Atlantic margins of southern Africa and southeastr Brazil and the relatiosnship to offshore sedimentation. In: Cameron, N.R., Bate, R.H., Clure, V.S. (eds.), The oilk and gas habitats of the South Atalntic, -pp.41-53.

Grotzinger, J.P., Bowring, S.A., Saylor, B.Z., Kaufman, A.J.,1995. Biostratigraphic and geochronologic constraints on early animal evolution. Science 270, 598_/604.

Luft Jr., J.L. 2005. Evolução Crustal do Vale do Rio Hoanib, Cinturão Kaoko, NW da Namíbia. Doctorade Thesis, UFRGS, Poroto Alegre.

Fernades, J. A. D., Tomasi, A. & Porcher, C. C. (1992) Deformation patterns in the southern Brazilian branch of the 1–11.

Gresse. P. G.; Chemale Jr., F.; Silva, L. C.; Walraven, F. & Hartmann 1996 Late to Post-Orogenic Basins of the Pan-African-Brasiliano Collision Orogen in Southern Africa and Southern Brazil. Basin Research, 8:157-171

Paim, P. S. G., Chemale Jr., F. & Lopes, R. C. 2000. A Bacia do Camaquã. In: Holz, M. & De Ros, L.F. 2000. Geologia e Estratigrafia do Rio Grande do Sul, Editora UFRGS, Porto Alegre, 231-272

Porada, H. (1989) Brasiliano rifting and orogenesis in southern to equatorial Africa and eastern Brazil. Precambr. Res. 44, 103-136.

Figure 1 - Tectonic evolution model for the SE Brazilian and SW African margin during the Neoproterozoic. The evolution of the both sides are not directly connected (modified after Chemale Jr. Luft Jr. 2005)

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 25

Figure 2 - Tectonic evolution model for the SE Brazilian and SW African margin during the late Neoproterozoic to Eopaleozoic (modified after Chemale Jr. Luft Jr. 2005).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 26 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Late Carboniferous to Permian sequence stratigraphy in the main Karoo Basin of South Africa and its application in Southwestern Gondwanaland

Douglas Ingle Cole Council for Geoscience, Bellville – South Africa, [email protected]

Abstract The Late Carboniferous to Permian sedimentary succession in the retro-arc Karoo foreland basin is divided into three stratigraphic sequences. The lowermost sequence consists of glaciogenic strata of the Dwyka Group, which starts at approximately 312 Ma in the southern foredeep and finishes at about 290 Ma when the ice sheets finally collapsed and rapid melting coincident with a Gondwana-wide marine transgression occurred. The second sequence consists of marine shales and a fluvio-deltaic sandstone wedge in the northeastern part of the basin derived from a local cratonic source. These form the lower Ecca Group and the sequence is terminated by a second marine flooding event at approximately 271 Ma. The third sequence comprises the upper Ecca and lower Beaufort Groups and is terminated at the Permian-Triassic boundary (251 Ma). It comprises a major regressive succession starting with marine shales that encompass deep water submarine fan sandstones in the southern foredeep. The fluvio-deltaic sandstone wedge, which forms a retrogradational sequence, was still being deposited in the northeastern part of the basin. Marine shales derived from suspension settling of mud predominated in the basin until about 267 Ma when deltaic sands emanating from southerly and westerly orogenic provenances, prograded northeast across the basin. At approximately 254 Ma, another set of deltaic sands prograded southwest into the northeastern part of the basin. The deltaic sediments were diachronously succeeded by terrestrial mudstone and sandstone of the lower Beaufort Group (Adelaide Subgroup). The lowermost sequence is present in most basins of southwestern Gondwanaland and all three apparently occur in the Paraná and distal part of the Kalahari Basin. Recognition of the second marine flooding event or its proximal-equivalent at the base of the third stratigraphic sequence might be achieved by the systematic dating of juvenile magmatic zircons present in volcanic tuffs within the Permian sediments of many of the basins.

Introduction The Main Karoo Basin evolved as a retro-arc foreland basin during the Late Carboniferous and is one of several such basins that formed in southwestern Gondwanaland inboard of the subducting Panthalassan Plate and which were first defined by Alex du Toit as his “Samfrau Geosynclne” (Du Toit, 1937; Veevers et al., 1994; Pankhurst et al., 2006).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 27 Strata up to 12 km thick accumulated in the main Karoo Basin until the Early Jurassic when sedimentation was terminated by the outpouring of at least 1.4 km of basaltic lavas (Cole, 1992). The sedimentary succession is characterized by initial glaciogenic deposits, followed by marine deposits, which shallow up through a deltaic transition into terrestrial deposits. The latter covered the entire basin by the latest Permian and continued until the Early Jurassic igneous event. Lithostratigraphically, the glaciogenic deposits form the Dwyka Group, the marine and deltaic deposits the Ecca Group and the terrestrial deposits the Beaufort and Stormberg Groups, separated by a mid-Triassic lacuna. The presence of diachronous boundaries and variable source areas promoted sequence-stratigraphic studies with several schemes being devised based upon tectonic control caused by subduction and mountain building along the Panthalassan margin of Gondwanaland (Veevers et al., 1994; Catuneanu et al., 1998). Previous misinterpretation of lithostratigraphic correlations and the recognition of marine flooding events in South Africa and India have allowed a simpler and possibly more extensive sequence-stratigraphic scheme to be delineated within the Late Carboniferous to Permian sedimentary succession (Cole et al., in prep.). This scheme and its applicability to other successions in southwestern Gondwanaland is described here, but is limited to the pre- Triassic period, since younger sediments are missing in some basins, e.g. Ellsworth and Sierra de la Ventana, due to non-deposition and/or subsequent erosion.

Stratigraphy The Late Carboniferous to Permian sedimentary succession in the main Karoo Basin is divided into three stratigraphic sequences, resting on a basal diachronous boundary and finishing at the Permian-Triassic boundary (Figure 1). The lowermost sequence consists of glaciogenic strata (Dwyka Group) up to 700 m thick, with the oldest sediments occurring in the southern part of the basin in the foredeep, starting at approximately 312 Ma. Four deglaciation sequences have been recognized in the foredeep, consisting of thick diamictite representing debris rain and debris flow from a melting marine ice sheet, overlain by subordinate shale, sandstone and diamictite representing suspension sedimentation, proglacial subaqueous debris flow and iceberg-rafted dropstones (Visser et al., 1997). The deglaciation sequences were derived from specific ice-spreading centres located north, east and south of the main Karoo Basin. In the northern part of the basin, only the upper two sequences have been recognized in the deeper distal parts of paleovalleys, with just the fourth uppermost sequence being present in the proximal parts. Here, the strata consist of a heterolithic arrangement of diamictite, shale, rhythmite, sandstone and conglomerate, representing deposition from retreating tidewater glaciers. The upper boundary of the stratigraphic sequence is defined by a Sakmarian marine flooding event, the so-called “Eurydesma transgression” (Veevers et al., 1994). This marks the final melting of the ice sheets at the termination of deglaciation sequence 4, where heterolithic glaciogenic

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 28 lithofacies are overlain by thick shale with iceberg-rafted dropstones in the basal few metres (Figure 1). SHRIMP zircon dates from tuff beds contained within the lowermost shale show ages of 288 and 290 Ma (Bangert et al., 1999). The first tectonic event has been dated at 292 Ma (Hälbich et al., 1983) and this may have triggered shelf subsidence and a final collapse of the ice sheets.

Figure 1 – Distribution in time and space of Late Carboniferous to Permian lithostratigraphic units and environments within a north-south section across the main Karoo Basin, displaying the three major sequences. Modified from Figure 7 of Veevers et al. (1994). The geologic time scale is from Gradstein et al. (2004).

The second stratigraphic sequence is up to 1200 m thick and corresponds with the lower Ecca Group (Figure 1). It consists primarily of shale with a sandstone wedge (Vryheid Formation) in the northeastern part of the basin, but as a result of slow retreat of tidewater glaciers from an ice sheet located north of the basin, the basal part of the sequence in the northern part of the basin contains glaciogenic strata. The shale represents suspension- settling of mud in a deep-water marine environment and the sandstone wedge, which consists of two coarsening-upward cycles, represents prograding fluvio-deltaic successions derived from northerly and northeasterly-located provenances (Veevers et al., 1994). The uppermost part of the sequence consists of a black, carbonaceous shale unit (Whitehill Formation) up to 80 m thick that contains the fossil reptile Mesosaurus and grades northeastwards into the upper part of the second coarsening-upward cycle of the Vryheid Formation (Figure 1). It may represent mud-suspension in a shallow water basin with reducing conditions below algal mats on the basin floor. The upper boundary of the stratigraphic sequence marks the start of a basinwide flooding event above the Whitehill Formation and the second coarsening-upward cycle of the Vryheid Formation (Figure 1).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 29 This flooding event was probably marine, given the presence of glauconite at the top of the second coarsening-upward cycle, and was probably a result of glacio-eustatic sea-level rises (Veevers, 2004) supplemented by regional dynamic subsidence following the second tectonic event dated at 278 Ma (Hälbich et al., 1983). The third stratigraphic sequence attains 7000 m in thickness and corresponds with the upper Ecca and lower Beaufort Groups (Figure 1). The initial strata in the southern and western parts of the basin consist of a succession of shale and sandstone, which represent the episodic incursion of deep water submarine fans from southward- and westward-located orogenic provenances (Veevers et al., 1994). The lowermost shale-dominated unit (Collingham Formation) contains abundant thin beds and lamina of soft grayish-yellow mudstone, which represents fall-out tephra derived from a magmatic arc inboard of the Panthalassan margin of Gondwanaland. U-Pb dating of juvenile magmatic zircons from this mudstone gave an age of 270 Ma, which indicates that the lower sequence boundary is approximately 271 Ma (Figure 1). The initial strata in the central part of the basin comprise shale that represents suspension settling of mud within a marine/paralic environment. In the northeastern part of the basin, a sandstone wedge comprising up to four coarsening-upward cycles (upper Vryheid Formation) is present and this represents a retrogradational sequence of prograding fluvio-deltaic units that were derived from a northeasterly-located provenance (Figure 1). These cycles give way to the shale-dominated Volkrust Formation, which contains a fossil of the wide-ranging marine bivalve, Megadesmus, and corresponds with a marine highstand. The sandy submarine fans in the southern and western parts of the basin also give way to shale (Fort Brown, Kookfontein and Tiergberg Formations) representing suspension settling of mud, prior to the start of a major regressive sequence that has been linked to a period of orogenic unloading following the 278 Ma tectonic event (Catuneanu et al., 2002). This regressive sequence is diachronous and started in the southern and western parts of the basin at about 267 Ma with the progradation of deltaic sands (Waterford Formation) from southward- and westward-located orogenic provenances (Figure 1). These deltas advanced north and northeast across the basin and were succeeded by terrestrial sediments of the lower Beaufort Group (Adelaide Subgroup). The third tectonic event dated at 258 Ma (Hälbich et al., 1983) resulted in renewed orogenic loading and uplift of the forebulge region north of the main Karoo Basin followed by progradation of fluvio-deltaic sediments (Normandien Formation) into the northern part of the basin (Figure 1). Terrestrial sedimentation continued up to and across the Permo-Triassic boundary, which corresponds to the upper boundary of the third stratigraphic sequence (Figure 1). This separates an upper Lystrosaurus from a lower Dicynodon fossil tetrapod biozone (Rubidge, 2005) and was accompanied by rapid climatic warming with initial deposition of floodplain mud followed by fluvial sand from a rising Cape Fold Belt south of the basin, an uplifted backbulge north of the basin and an uplifted block on the eastern side of a major shear system.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 30 Stratigraphic correlation in Southwestern Gondwanaland The same sequence stratigraphic division has been recognized in the Gondwana Sequence of India (Cole et al., in prep.) and some of the sequence boundaries are present in other basins of southwestern Gondwanaland (Figure 2). The lowermost sequence again comprises mostly glaciogenic strata, beginning at about 320 Ma in the basins of western Argentina and terminating in most basins by about 290 Ma, with shale representing suspension settling of marine mud at the onset of the Sakmarian flooding event (Figure 3; Collinson et al., 1994; López-Gamundí et al., 1994; Veevers, 2004). The four deglaciation sequences recognized in the southern part of the main Karoo Basin are present in the distal part of the Kalahari Basin (Bangert et al., 1999) and possibly three sequences are present in the Paraná Basin (Eyles et al., 1993). The identification of deglaciation sequences is important for locating the position of ice-spreading centres in southwestern Gondwanaland and whether or not they retreated simultaneously. Although the sequence boundary that corresponds with the Sakmarian flooding event can be recognized in most basins, the boundary at the top of the second sedimentary sequence, which defines the start of a second basinwide flooding event, can only be identified in the distal part of the Kalahari Basin and in the Paraná Basin, above or within the distinctive black, carbonaceous shale unit (Whitehill/Irati Formations) that contains the fossil reptile Mesosaurus (Zalán et al., 1990). Generally, there is an overall regressive sequence during the Permian, following the Sakmarian flooding event (Figure 4). This culminated in terrestrial sedimentation after approximately 260 Ma and continues up to and across the Permo-Triassic boundary, where present (Figure 5; Veevers, 2004).

Figure 2 – Lithostratigraphic sections from four basins in Southwestern Gondwanaland showing possible correlation with the three major sequences in the main Karoo Basin during the Late Carboniferous - Permian. The geologic time scale is from Gradstein et al. (2004).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 31

Figure 3 – Paleogeographic map of Southwestern Gondwanaland during the start of the Sakmarian flooding event (~290 Ma), which corresponds with the boundary between stratigraphic sequences 1 and 2.

References Bangert, B., Stollhofen, H., Lorenz, V. and Armstrong, R., 1999. The geochronology and significance of ash-fall tuffs in the glaciogenic Carboniferous-Permian Dwyka Group of Namibia and South Africa: Journal of African Earth Sciences, v. 29, p. 33-49.

Catuneanu, O., Hancox, P.J. and Rubidge, B.S., 1998. Reciprocal flexural aviour and contrasting stratigraphies: a new basin development model for the Karoo retroarc foreland system, South Africa: Basin Research, v. 10, p. 417-439.

Catuneanu, O., Hancox, P.J., Cairncross, B. and Rubidge, B.S., 2002. Foredeep submarine fans and forebulge deltas: orogenic off-loading in the underfilled Karoo Basin: Journal of African Earth Sciences, v. 35, p. 489- 502.

Cole, D.I., 1992. Evolution and development of the Karoo Basin. in: De Wit, M.J. and Ransome, I., eds., Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa: Rotterdam, Balkema, p. 87-99.

Cole, D.I., Mukhopadhyay, G., Mukhopadhyay, S.K., Parui, P.K., Neveling, J. and Roychowdhury, M., in preparation. Correlation between the Karoo Supergroup of South Africa and the Gondwana Sequence of India: South Africa/India Bilateral Research Project: Council for Geoscience of South Africa, Memoir.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 32

Figure 4 – Paleogeographic map of Southwestern Gondwanaland during the early part of stratigraphic sequence 3 (~267 Ma).

Collinson, J.W., Isbell, J.L., Elliot, D.H., Miller, M.F. and Miller, J.M.G., 1994. Permian-Triassic Transantarctic Basin. in: Veevers, J.J. and Powell, C.M.A., eds., Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland: Geological Society of America, Boulder, Colorado, Memoir, 184, p. 173-222.

Du Toit, A.L., 1937. Our Wandering Continents: London, Oliver and Boyd, 366 p.

Eyles, C.H., Eyles, N. and França, A.B., 1993, Glaciation and tectonics in an active intracratonic basin: the Late Palaeozoic Itararé Group, Paraná Basin, Brazil: Sedimentology, v. 40, p. 1-25.

Gradstein, F.M., Ogg, J.G., Smith, A.G., Bleeker, W. and Lourens, L.J., 2004. A new geologic time scale, with special reference to Precambrian and Neogene: Episodes, v. 27, p. 83-100.

Hälbich, I.W., Fitch, F.J. and Miller, J.A., 1983. Dating the Cape Orogeny. in: Söhnge, A.P.G. and Hälbich, I.W., eds., Geodynamics of the Cape Fold Belt: Geological Society of South Africa, Special Publication, 12, p. 149- 164.

López-Gamundí, O.R., Espejo, I.S., Conaghan, P.J. and Powell, C.M.A. 1994. Southern South America. in: Veevers, J.J. and Powell, C.M.A., eds., Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland: Geological Society of America, Boulder, Colorado, Memoir, 184, p. 281-329.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 33

Figure 5 – Paleogeographic map Southwestern Gondwanaland during the later part of stratigraphic sequence 3 (~ 255 Ma).

Pankhurst, R.J., Rapela, C.W., Fanning, C.M. and Márquez, M., 2006. Gondwanide continental collision and the origin of Patagonia: Earth-Science Reviews, v. 76, p. 235-257.

Rubidge, B.S., 2005. Re-uniting lost continents – Fossil reptiles from the ancient Karoo and their wanderlust: South African Journal of Geology, v. 108, p. 135-172.

Veevers, J.J., 2004. Gondwanaland from 650 – 500 Ma assembly through 320 Ma merger in Pangea to 185 – 100 Ma breakup: supercontinental tectonics via stratigraphy and radiometric dating: Earth-Science Reviews, v. 68, p. 1-132.

Veevers, J.J., Cole, D.I. and Cowan, E.J., 1994. Southern Africa: Karoo basin and Cape Fold Belt. in: Veevers, J.J. and Powell, C.M.A., eds., Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland: Geological Society of America, Boulder, Colorado, Memoir, 184, p. 223-279.

Visser, J.N.J., Van Niekerk and Van der Merwe, S.W., 1997. Sediment transport of the Late Palaeozoic glacial Dwyka Group in the southwestern Karoo Basin: South African Journal of Geology, v. 100, p. 223-236.

Zalán, P.V., Wolff, S., Astolfi, M.A.M., Vieira, I.S., Conceição, J.C.J., Appi, V.T., Neto, E.V.S., Cerqueira, J.R. and Marques, A., 1990, The Paraná Basin, Brazil. In: Leighton, M.W., Kolata, D.R., Oltz, D.F. and Eidel, J.J., eds., Interior cratonic basins: American Association of Petroleum Geologists, Tulsa, Memoir, 51, p. 681-708.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 34 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Chrono-, chemical-, seismic- , electrical- and tectono-stratigraphy across parts of the Cape Fold Belt – Karoo Basin of South Africa: New foundations for correlations across the South Atlantic

Maarten de Wit1, Richard Armstrong, Sam Bowring, James Alexander, Thomas Branch, John Decker, Joy Ghosh, Shaun Moore, Ansa Lindeque, Jacek Stankiewicz, Nicolas Rakatolosofo ¹AEON- Africa Earth Observatory Network, and department of Geological Sciences, University of Cape Town, South Africa; email: [email protected]

More than 100 years ago, Alex du Toit and Juan Keidel both appreciated the first order geologic similarities on either side of the Atlantic with which to champion continental drift. Subsequent workers, using mainly litho- and bio-stratigraphy, have vindicated a co- evolving system of the Paleozoic to Mesozoic Cape-Karoo and Paraná basins and their flanking mountain ranges – the Cape Fold Belt (CFB) and the Sierra de la Ventana, as first detailed by du Toit (1927). Yet to date, prevailing models fail to provide a much better resolution with which to further detail the shared history of these once conterminous geological provinces. Uncertainties are large, and linked to the possibility that geologic events along the ‘Samfrau’ or ‘Gondwanide’ orogen, and its flanking basins, were diachronous. There is a notable lack of relevant modern data with which to test correlations between these Paleozoic- early Mesozoic provinces. Here, we highlight some of these shortcomings and report on new data from southern Africa, based on various modern stratigraphic tools that were not available in the Du Toit/Kiedel era, and which should be compared to similar data from South America. The Cape-Karoo basin stretches across much of South Africa, and intermittently northwards into Namibia and Zimbabwe, with an original area in excess of 1 x 106 km2. The basin was once more extensive, with local remnants preserved as far as in central Africa and Madagascar. In this regional context the rocks span a history of almost 400 Ma. The main

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 35 depocentres are, however, confined to South Africa. In places their total thickness is reported to reach over 10 km, and their preserved stratigraphic record spans just over 300 Ma. In the southernmost regions, the crust below the Karoo basin is about 38 km thick, increasing to 43 km below the CFB. Below the southern CFB ranges, however, the thickness decreases abruptly to 30 km at the coast, and from there to less than 20 km thick across ~250 km of continental shelf up to the Agulhas Fracture Zone. The Cape-Karoo basin is divided into two supergroups, traditionally treated as having formed in two separate successor basins, in turn subdivided into a number of groups. The lowermost of these is the siliciclastic Cape Supergroup that ranges in age from mid Cambrian (ca 500 Ma) to upper Devonian (ca 360 Ma) and comprises a number of well- defined marine transgression-regression sequences devoid of any volcanic material. Sedimentation in the Cape Basin terminated in the Lower Carboniferous period (Visean) with deposition of tidal flat to lacustrine sandstones of the Tournasian Waaipoort Formation of the Witteberg Group (Evans, 1999). Directly underlying the Cape Supergroup is a sequence of silici- and calci-clastic sediments and volcaniclastic rocks (e.g. the Kango Group) that have been traditionally grouped with the Precambrian basement to the Cape Supergroup within the CFB, but are now dated (U/Pb on detrital zircons) to be at least in part early Cambrian in age. This represents the lowermost rift sequence of the Cape Supergroup (Armstrong et al., 1998; Barnet et al., 1997). An angular unconformity separates these deposits from underlying Neoproterozoic metasediments intruded by Cambrian A-type granites as young as 520-540 Ma. U/Pb dating and geochemistry has shown that the granites and associated rhyolitic extrusives below the famous Cape unconformity, which separates them from the overlying siliciclastics of the Cambrian-Ordovician Table Mountain Group, to overlap in age and chemical composition with similar rock types below the unconformity of the thick clastic sediments (Curamalal Group) in the Sierra de la Ventana (Rapela et al., 2003). This is the most dramatic modern follow-up correlation work todate vindicating the earliest lithostratigraphic correlations between Africa and South America documented almost a century ago by Kiedel (1916). In the south-central Karoo Basin, new lithostratigraphic mapping and seismic imaging show that the Kango Group is absent there and that the Cape Supergroup lies directly on Mesoproterozoic basement (the Natal-Namaqua Kibaran-age). Near the northern tectonic front of the Cape Fold Belt, this basement contains the largest continental magnetic anomaly in the world (i.e. the Beattie Anomaly). A possible extension of this anomaly has recently been recognized just north of the Sierra de la Ventana in Argentina. In South Africa, the origin of this anomaly has been debated for nearly 30 years. The most accepted model relates it to a paleosuture zone of either Pan African or Kibaran age. Recent seismic and magnetotelluric soundings show this interpretation to be unlikely, because the anomaly is

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 36 confined to the upper crust (Weckman et al., 2007; Stankiewicz et. al., 2007; Lindeque et al., 2007). Resolving this feature further in South America and South Africa should test possible correlations across the Atlantic of Grenville-aged structures in the basement to the Karoo- Paraná basins. Such potential correlations would constrain any orogen-parallel displacements within the intervening late Neoproterozoic belt that runs parallel to the Atlantic margins (e.g. the now separated Dom Feliciano and Gariep-Saldanian belts of South America and southern Africa, respectively). The Karoo Supergroup starts with an extensive sequence of glacial sediments with up to seven major ice advance-retreat episodes representing the Carboniferous-early Permian ‘Dwyka’ Gondwana glaciations (Opdyke et al., 2001). The age of the overlying Dwyka deposits has been a matter of long debate. It was previously believed that a hiatus of ~30 Ma may have occurred before deposition of the Karoo Sequence commenced. However, dropstones and soft sediment deformation features believed to be glacial in their origin are present in the uppermost beds of the Witteberg Group (Streel and Theron, 1999). Therefore, the onset of glaciation was in progress during deposition of the uppermost Visean-age sediments of the Cape Supergroup (~350 Ma), and there is in fact minimal hiatus before glaciers advanced across the Witteberg glaciogene surface. Rhyolitic-andesitic volcanic tuffs are present in the Dwyka Group of southern Africa. U/Pb dating on zircons from tuffs about 400 meters above the base of the Dwyka yield a date of 297± 1.8 Ma (Bangert et al., 1999). The age of the top of the Dwyka has been similarly derived using zircons dates from tuffs in the lowermost beds of the overlying Prince Albert Formation (288 ±3 and 289 ±3.8). Thus, the Dwyka Group in southern Africa spans an extraordinary long time period of ~50 million years. The environment of deposition of the Dwyka also remains enigmatic. Most investigators infer a marine setting (e.g.Visser, 1997), but there is no unequivocal evidence for this. It is possible also that the Dwyka sequence was deposited in a terrestrial setting, as originally suggested by du Toit (1926). Marine fossils occur only at the top of the Dwyka along the northern and western margins of the basin. These probably relate to a short marine transgression related to eustatic sealevel rise following rapid Gondwana deglaciation (du Toit, 1954). Thereafter, post-glacial isostatic rebound of the northern and eastern provenances resulted in an influx of fluvial deltaic sands and onset of extensive coal deposition in the northeastern part of the Karoo Basin. Sediments of the overlying carbonaceous mudstones of the Prince Albert and Whitehill Formations also indicate non-marine conditions and a time that the Karoo Basin had become a gigantic highly reduced inland lake up to three times the size of the present Black Sea (du Toit 1926; Faure and Cole, 1999). Geochemistry indicates that most of these deposits are fresh to brackish water lake-deposits (Zawada and Cadle, 1988; Faure and Cole, 1999; Herbert and Compton, 2007) and not marine-deposits as is often inferred (e.g. Visser 1997). These observations support a model for the diamictites as glaciogenic lake-

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 37 sediments deposited beneath, and peripheral to, the major continental ice sheet that covered much of Gondwana at that time, much as we see beneath the Antarctic ice sheet today (e.g. Lake Vostok). How then should the glacial Dwyka deposits be correlated with the Itararé Groups in South America, beyond the general facies concept, to test if there are major differences between them in time, duration and environmental conditions? The distinctive black pyrite-rich mud horizon (the Whitehill Formation) extended across into the Paraná Basin of South America in which the contemporaneous Irati Formation was deposited. Both contain the characteristic fauna of the fresh water Mesosaurus and dragonflies (du Toit, 1927, 1937; Araújo et al., 2001). This distinctive geochemical marker is a true ‘time line’ for correlation; and it has recently been shown that it can be traced accurately in subsurface using relatively fast and simple magnetotelluric resitivity sounding (Branch et al., 2007). Both the Whitehill and Irati shales have a very high total organic carbon content (up to 24% in condensed sections of the Irati), and a sulphur content as high as 8%, and elevated uranium concentrations, to provide a highly conductive layer (~120m thick) ideal for electrical mapping. Overlying turbidites illustrate a regional polarity change in the source of the clastic sediments of the southern Karoo. These turbidites grade northwards into a complex diachronous delta-shoreline sequence that separates a distinct change in depositional style between the lower and upper Karoo Basin deposits that roughly coincided with the Permo- Triassic boundary (Rubidge et al., 2000; Smith, 1995; Ward et al., 2000; 2005; MacLeod et al., 2000). The latter is now reasonably well-located on a meter scale, using bio- chemo-and magneto-stratigraphy. However, this world-class site for terrestrial studies across the major global extinction boundary cannot yet be correlated to the Paraná basin with any certainty, and still lacks precise definition needed for testing feedbacks between biological, climate and ecodynamic changes. The uppermost sedimentary rocks of the Karoo represent a terrestrial sequence formed under increasingly arid conditions (e.g. Smith 1995). Around the Triassic-Jurassic boundary, chemical-stratigraphy, based on carbon isotopes, has revealed at least 7 episodes of relatively wet conditions followed by droughts (Decker and de Wit, 2005) before the start of ~50 myrs of sandy desert conditions, punctuated at ~182 Ma by the outporings of vast volumes of Karoo basalts. Farther north, in Namibia and southern Angola, the sand-dunes of this extensive desert were finally ‘drowned’ by the ca. 134 Ma Etendeka basalts and rhyolites, a small African remnant outlier of the great Paraná continental flood basalt province. Within 10 myrs thereafter, progressive opening of the South Atlantic Ocean followed (Eagles 2007), and warm humid conditions set in. The continental margins south of the Walvis Ridge became the foci of extensive seaward dipping basalt sequences and thick basaltic underplating, as indicated by recent seismic surveys, as well as extensive dyke swarms that cut obliquely across the margin, and which may be useful to test detailed cross-

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 38 Atlantic correlations (e.g.Trumbull et al., 2007). Refraction images shows that crustal thinning from the Gariep-Saldanian belt (~40 km) and across the continental shelf (30-20 km over ~300 km) is surprisingly similar as that seen along the south coast. Between 90-120 Ma, most of the uplift of the Kalahari Plateau took place, forcing rapid extensive erosion of the

Karoo/Parana basalts, accelerating draw-down of atmospheric CO2 to such degree that by the end Cretaceous a cooling trend was well on the way (de Wit, 2007). During the time of turbidite deposition in the Karoo basin, many thin rhyolitic-andesitic airfall tuffs were deposited (e.g. Cole, 1992). U/Pb dating on zircons from these tuffs (Bangert et al 1999; Bowring and de Wit, unpublished) shows them to be especially common around 270-280 Ma (Collingham Formation), but their total age range is dated between 250 and 300 Ma (Bangert et al., 1999). Several workers have linked these Karoo tuffs to volcanic activity along the convergent margin of the Gondwanides in Argentina. In South Africa, these tuffs predate and/or overlap with the first episodes of tectonic activity in the CFB, and record the onset of the deeper water deposition across southern Africa. For the first time, sediments became dominantly sourced from the arc terranes along the Gondwana margin to the south and east of the CFB (e.g. Cole, 1992; Cloetingh et al., 1992; Catuneanu et al., 2002). Similar airfall tuffs in sequences of the Paraná basin have also been dated to range between ca 278 and 299 Ma (Guerra-Sommer et al., 2005; Santos et al., 2006), offering a robust basis for potential detailed chronostratigraphic correlation between the Palaeozoic sequences of the two basins. Integrated with near continuous chemical stratigraphy that is now available throughout the Karoo basin, including the Dwyka glacial deposits, distinct carbon isotope excursions should be used to track climate changes and further cross-Atlantic correlations. The upshot of this work, albeit in early stages, yields a very different view as that of traditional interpretations on which Karoo basin modeling has relied. The southern Karoo Basin is believed to have evolved into a well-defined foreland basin around ~250 Ma, linked to the emerging CFB (Catuneanu et al., 1998, 2002). Yet detailed backstripping analyses show this basin not to have a typical subsidence history of a classic foreland basin-type (e.g. Cloetingh et al., 1992; Milani and de Wit, 2007). New high- resolution seismic stratigraphy (Lindeque et al., 2007) does not support the drastic thickening of the Karoo sequences towards the CFB on which the foreland basin models are based. In addition, the above-mentioned models are rooted in lithostratigraphic analysis complemented by biostratigraphy of limited value due to a scarcity of zone fossils with precise time resolution. This has hampered detailed tectonic modelling and basin analysis, and is the cause of significant controversy about the age and position of lithostratigraphic boundaries (e.g. Zawada and Cadle, 1988; Rubidge et al., 2000) as well as sequence stratigraphy (e.g. Catuneanu et al., 1998; Turner, 1999). There is simply not enough reliable chronostratigraphy to verify detailed correlations; and the little that does exist, shows that some of the basin evolution models are based on erroneous biostratigraphic extrapolations.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 39 For example, Catuneanu et al. (1998) place the lithostratigraphic top of the Dwyka sequence at ca 268 Ma and 265 Ma in the southern and northern parts of the Karoo Basin, 30-35 Ma too young, respectively, when the lithostratigraphy is framed in more recent chronostratigraphic data based on U/Pb zircon analyses. Similarly, their age of the Collingham Formation is at least 12 Ma too young in light of the ~ 270 Ma U/Pb dates derived from zircons in the widespread airfall tuffs from this formation (Bowring and de Wit, unpublished data). The age of the CFB orogeny also remains poorly documented, as it is based essentially on whole rock Ar/Ar dating. Whilst detailed structural history within this belt remains to be resolved, traditional fold models are clearly incorrect (Craddock et al., 2007). Extensive early thrusting and downward facing sequences have been recognized and documented (e.g. Booth 1998), and large-scale duplexes within the Table Mountain Group of the central CFB yield minimum subhorizontal shortening ~75% over distances >2km (de Wit and Moore, unpublished). These new findings will have major impact on our understanding of the original paleogeography, tectonic evolution and correlations to South America. 100 years after du Toit and Kiedel, the best is yet to come.

References Araújo, L.M., Rodrigues, R. & Scherer, C.M.S. 2001. Expressão estratigráfica do carbono orgânico, resíduo insolúvel, enxofre, parâmetros de pirólise e elementos-traço nas seqüências 40 deposicionais Irati, Bacia do Paraná. Curitiba, VII Congresso Brasileiro de Geoquímica,abstract 207.

Armstrong, RA., de Wit, MJ., Reid, D., York, D. & Zartman, R. 1998. Cape Town’s Table Mountain reveals rapid Pan-African uplift of its basement rock. Journal of African Earth Sciences, 27-1A, 10-11.

Bangert, B., Stollhofen, H., Lorenz, V. & Armstrong, R. 1999. The geochronology and significance of ashfall tuffs in the glaciogenic Carboniferous-Permian Dwyka Group of Namibia and South Africa. Journal of African Earth Sciences, 29, 33-50.

Barnett, W., Armstrong, R. & de Wit, M. 1997. Stratigraphy of the upper Neoproterozoic Kango and lower Paleozoic Table Mountain Groups of the Cape Fold Belt, revisited. South African Journal of Geology, 100, 237-250.

Booth, P.W.K.1998. The effect of thrusting on fold style and orientation, Weltevreden Formation, Stellterville, easterb Cape. South African Journal of Geology, 101, 27-37

Branch, T., Weckmann, U. & Ritter, O. 2007. The Whitehill Formation - a highly conductive marker in the Karoo Basin. South African Journal of Geology, in press.

Catuneanu, O., Hancox, P.J., Cairncross, B. & Rubidge, B.S. 2002. Foredeep submarine fans and forebulge deltas: orogenic off-loading in the underfilled Karoo Basin. Journal of South African Earth Sciences, 35, 489- 502.

Catuneanu, O., Hancox, P.J. & Rubidge, B.S. 1998. Reciprocal flexural behaviour and contrasting stratigraphies: a new basin development model for the Karoo retroarc foreland system, South Africa. Basin Research, 10, 417-439.

Cloetingh, S., Lankreijer,A., de Wit, M.J. & Martinez, I. 1992. Subsidence history analyses and forward modelling of the Cape and Karoo Supergroups. In: de Wit, M.J. & Ransome, I.D. (eds). Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa. Balkema, Rotterdam, 239-248.

Cole, D.I., 1992, Evolution and development of the Karoo Basin. In: de Wit, M.J. & Ransome, I.D. (eds). Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa. Balkema, Rotterdam, 87- 100.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 40 Craddock, J.P., McKiernan, A.W., de Wit, M.J. 2007. Calcite twin analysis in syntectonic calcite, Cape Fold Belt, South Africa: Implications for fold and cleavage formation within a shallow thrust front. Journal of Structural Geology 29, 1100-1113

Decker, JE. & de Wit, MJ. 2005. Carbon isotope evidence for Crassulacean Acid Metabolism in the Mesozoic. Terra Nova, 18, 9-17. de Wit, M.J. 2007. The Kalahari Epeirogeny and climate change: differentiating cause and effect from Core to Space. South African Journal of Geology, in press. du Toit, A.L. 1926. The Geology of South Africa. Oliver and Boyd, Edinburgh. du Toit, A.L. 1927. A geological comparison of South America with South Africa. The Carnegie Institution, Washington, publication number 381. du Toit, A.L. 1937. Our wandering continents. Oliver and Boyd, Edinburgh. du Toit, A. L. 1954. The Geology of South Africa. Haughton and Boyd, London.

Evans, F.J. 1999. Paleobiology of Early Carboniferous biota of the Waaipoort Formation (Witteberg Group), South Africa. Palaeontologia Africanus, 35, 1-6.

Eagles, G. 2007. New angles on South Atlantic opening. Geophysical Journal International, 166, 353-361.

Faure, K. & Cole, D. 1999. Geochemical evidence for lacustrine microbial blooms in the vast Permian Main Karoo, Paraná, Falkland Islands and Haub basins of southwestern Gondwana. Palaeogeography Palaeoclimatology Palaeoecology, 152, 189-213.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Formoso, M.L., Menegat, R. & Basei, M. A. S. 2005. New radiometric data from ash fall rocks in Candiota coal-bearing strata and the palynostratigraphic framework in southern Paraná Basin (Brazil). In: Pankhurst, R.J. & Veiga, G. D. (eds), Gondwana 12: Geological and Biological Heritage of Gondwana, Abstracts, Academia Nacional de Ciencias, Córdoba, Argentina, 189.

Herbert, C.T. and Compton, J.S. 2007. Depositional environments of the lower Permain Dwyka diamictite and prince Albert shale inferred form the geochemistry of early diagenetic concretions, southwest Karoo Basin, South Africa. Sedimentary Geology, 194, 263-277.

Keidel, J. 1916. La geología de las sierras de la Província de Buenos Aires y sus relaciones con las montañas de Sud Africa y los Andes. Buenos Aires, Anales del Ministerio de Agricultura de la Nación, Sección Geología, Mineralogía y Minería, 3, 1-78.

Lindeque, A.S., Ryberg, T., Stankiewicz, J., Weber, M.H. & de Wit, M.J. 2007. Deep crustal seismic reflection experiment across the southern Karoo basi, South Africa. South African Journal of Geology, in press.

MacLeod, K.G., Smith, R.M.H., Koch, P.I. & Ward.P. 2000. Timing of mamal-like reptile extinction across the Parmian-Triassic boundary in South Africa. Geology, 28, 227-230.

Milani and de Wit, 2007. Correlations between the classic Paraná and Cape-Karoo sequences of South America and southern Africa and their basin infills flanking the Gondwanides:Du Toit revisited In: Pankhurst, R. et al. (eds), Western Gondwana - the Ties that Bind. The Geological Society, London (in press).

Opdyke, N.D., Mushayandebun, M. & de Wit, M.J. 2001. A new palaeomagnetic pole for the Dwyka System and correlative sediments in sub-Saharan Africa. Journal of African Earth Sciences, 33, 143-154.

Rapela, C.W., Pankhurst, R.J., Fanning, C.M. & Grecco, L.E. 2003. Basement evolution of the Sierra de la Ventana Fold Belt: new evidence for Cambrian continental rifting along the southern Margin of Gondwana. Journal of the Geological Society, London, 160, 613-628.

Rubidge, B.S., Hancox, P.J. & Catuneanu, O. 2000. Sequence analysis of the Ecca-Beaufort contact in the southern Karoo of South Africa. South African Jornal of Geology, 103, 81-96.

Santos, R.V., Souza, P.A., Alvarenga, C.J.S., Dantas, E.L., Pimentel, M.M., Oliveira, C.G. & Araújo, L.M. 2006. Shrimp U-Pb zircon dating and palynology of bentonitic layers from the Permian Irati Formation, Paraná Basin, Brazil. Gondwana Research, 9, 456-463.

Smith, R.M.H. 1995. Fluvial environments across the Permian-Triassic boundary in the Karoo Basin, South Africa and possible causes of the tetrapod extinctions. Palaeogeography Palaeoclimatology Palaeoecology, 117, 84-104.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 41 Stankiewicz, J., Ryberg, T., Schulze, A., Lindeque, A.S., Weber, M.H. & de Wit, M.J. 2007. Initial results from the wide-angle seismic refraction lines in the Southern Cape. South African Journal of Geology, in press.

Streel, M. & Theron, J.N. 1999. The Devonian-Carboniferous boundary in South Africa and the age of the earliest episode of the Dwyka glaciation: New palynological result. Episodes, 22, 41-44.

Trumbull, R.B., Reid, D.L., de Beer, C., van Acken, D., Romer, R.L. 2007 Magmatism and continental breakup at the west marginof southern Africa:A geochemical comparison of dolerite dikes from northwestern Namibia and the Western Cape. South African Journal Geology, in press.

Turner, B.R. 1999. Tectonostratigraphical development of the upper Karoo foreland basin: orogenic unloading versus thermally-induced Gondwana rifting. Journal of African Earth Sciences, 28, 215-238.

Visser, J.N.J. 1997. Deglaciation sequences in the Permo-Carboniferous Karoo and Kalahari basins of southern Africa: a tool in the analysis of cyclic glaciomarine basin fills. Sedimentology, 44: 507-521.

Ward, P.D., Montgomery, D.R. & Smith, R. 2000. Altered river morphology in South Africa related to the Permian- Traissic extinction. Science, 289, 1740-1743.

Ward, P., R. Buick, J. Botha, M. O. de Kock, D. H. Erwin, G. Garrison, J. Kirschvink and R. Smith. 2005. Abrupt and gradual extinction among land vertebrates in South Africa: Evidence against a K/P-type impact extinction at the end of the Permian. Science 307: 709-713.

Weckmann,U., Ritter, O., Jung, A., Branch, T. & de Wit, M.J. 2007. Magnetotelluric measurements across the Beattie magnetic anomaly and the Southern Cape Conductive Belt, South Africa. Journal of Geophysical Research, 112, doi:10.1029/2005JB003975.

Zawada, P.K. & Cadle, A.B. 1988. Position of the Ecca-Beaufort boundary contact in the southwestern Free State: an evaluation of four possible alternatives. South African Journal of Geology, 91, 49-56.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 42 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Tectonic and climatic induced changes in depositional styles of the Mesozoic sedimentary record of Southern Paraná Basin, Brazil

Ubiratan Ferrucio Faccini Programa de Pós-graduação em Geologia, Universidade do Vale do Rio dos Sinos, São Leopoldo – RS – Brazil, [email protected]

Abstract The stratigraphy and depositional architecture of the Mesozoic of southern Paraná Basin are presented, emphasizing the Triassic changes in depositional styles, interpreted as a response to tectonic and climatic changes. The sedimentary record can be grouped into three major chronostratigraphic intervals based on vertebrate ages (Late Permian-Early Triassic; Middle-Late Triassic; Early Cretaceous). Furthermore, at least five third-order sequences can be identified: (1) Sanga do Cabral (Early Triassic; Scythian): alluvial braidplain; (2) Santa Maria/Caturrita (Middle-Late Triassic): fluvial braided, floodplains and sinuous/anastomosing, rich vertebrate fauna; (3) Mata (Rhaetian?): fluvial, low sinuosity, silicified wood; (4) Guará (Late Jurassic?), fluvial/eolian, also composes the Mesozoic record, occurring to the west of the study area focused here. (5) Botucatu (Early Cretaceous): dry eolian system. The basalts extrusions of the Serra Geral Formation close the sedimentation in this portion of Paraná Basin.

Introduction The Mesozoic sedimentary record in southern Paraná basin can be divided, at least, in five depositional sequences, limited by regional unconformities and marked by changes in depositional styles interpreted as a response to tectonic and climatic processes related to the period between the final disconnection to the Panthalassa during the end of Paleozoic and the breakup of Gondwana in Cretaceous times. This sedimentary succession marks the final phases of progressive continentalization of the basin and includes several lithostratigraphic units, ranging in age from Late Permian to Early Cretaceous times (Rio do Rasto, Pirambóia, Sanga do Cabral, Santa Maria (Passo das Tropas and Alemoa Members), Caturrita (Mata Sandstone), Guará and Botucatu formations. This work presents a stratigraphic analysis of the Triassic record exposed in the Central Region of the Rio Grande do Sul State (Figure 1), that corresponds to a second order sequence (Milani, 1997), and can be divided in this area into four third-order allostratigraphic units (Figure 2), based on changes of depositional styles (Figure 3; Faccini et al. 2003) and ages obtained from paleontologic data (Schultz et.al, 2000).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 43

Stratigraphic evolution and accommodation changes The continentalization of the basin during Late Permian is well recorded by the progressive implantation of an extensive aeolian sand sea (Pirambóia Formation) over the lacustrine and deltaic sediments of the Rio do Rasto Formation. This unit is characterized by deposits of dunes, wet interdunes and the occurrence of minor isolated sandy channels (wadis), enclosed in the dominant eolian facies. In this stratigraphic level is common the occurrence of soft deformation structures, interpreted as sismites and related to the first manifestations of the distensive Middle Triassic phase of the Gondwana. The end of this aeolian sedimentation is marked by a regional unconformity surface (supersurface) produced by the development of a braidplain (Sanga do Cabral Formation), suggesting a climatic change during the Early Triassic. These alluvial deposits contain vertebrate fossils of Scythian age and are dominated by tabular, unconfined sandstones, with common flat lamination and small to medium scale bedforms, indicating low accommodation conditions. The onset of the Middle Triassic extensional rifting phase of the Gondwana is represented in this portion of the basin by an extensive erosional surface, coincident with an important change in the alluvial style characterized by the deposition of a conglomeratic to coarse-grained sandy braided fluvial system (Passo das Tropas Member, Santa Maria Formation), dominated by macroforms and downstream accretion deposits (Figure 3). This remarkable fluvial incision and sedimentation, related to a lowering of the stratigraphic baselevel, is interpreted as a sedimentary response to tectonic reorganization of the basin during Middle Triassic times (Figure 2). Following the low accommodation Passo das Tropas fluvial system, the basin experiences an abrupt onset of a high accommodation phase. It occurs as an expansion of the sedimentation with the predominance of fine-grained floodplain deposits, including small lakes, loess and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 44 paleosols (Alemoa Member, Santa Maria Formation), with minor isolated fluvial channels (ribbons) and the development of a rich vertebrate fauna.

Toward the top of this succession (Caturrita Formation, Middle to Late Triassic), the progressive decreasing of the rate of accommodation creation leads to the amalgamation of channels and changing in fluvial style, from isolated (sinuous/anastomosing) to sheet-like deposits of channels filled by lateral accretion macroforms, characteristics of a meandering system. This coarsening upward trend coincides also with the reduction of fine-grained lithofacies, possibly due to the channel migration and progressive reworking of the floodplains. During the Late Triassic times (Rhaetian?) the basin records a new lowering in baselevel and a reduction of accommodation with the incision of a braided fluvial system, mainly composed by downstream sandy macroforms, with minor fines of floodplain, and characterized by an expressive content of silicified trunks (Mata Sandstone). These fluvial deposits are extensive and correspond to a third or fourth order sequence, presumably related to combined effects of tectonic and climatic factors. The upper unconformity recorded in this portion of the basin corresponds to the onset of a very extensive dry aeolian system of the Botucatu Formation. This surface corresponds to an interregional unconformity recognized along the almost entire basin. The record of Botucatu desert consists mainly of aeolian dunes and dry interdunas. This system is covered and dated (Early Cretaceous) by the volcanic rocks of Serra Geral Formation. The depositional hiatus at bottom is interpreted as produced during a long period of uplift and erosion Figure 3). The aeolian deposition, in dry climatic conditions, finishes the sedimentary history of the basin in this region. Final remarks This work focused only the Central Region of Rio Grande do Sul. The East and West regions have different stratigraphic frameworks (see Figure 2), including the Guará Formation, not analyzed here. Detailed geologic mapping, new paleontologic data and, particularly, subsurface information (e.g. Zerfass et al. 2003; Wankler, 2006) are improving and refining greatly the correlations available today.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 45

References Faccini, U. F.; Giardin, A.; Machado, J. L. F., 2003. “Heterogeneidades litofaciológicas e hidroestratigrafia do Sistema Aqüífero Guarani na Região Central do Rio Grande do Sul”. In Paim, P. S. G.; Faccini, U. F. & Neto, R. G.(Ed.): Geometria, Arquitetura e heterogeneidades de corpos sedimentares - estudo de casos. Editora Unisinos,São Leopoldo.

Milani, E.J. 2004. Comentários sobre a origem e evolução tectônica da Bacia do Paraná. In: Mantesso-Neto, V., Bartorelli, A., Carneiro, C.D.R. and Brito-Neves, B.B. (eds) Geologia do continente Sul-Americano: evolução da obra de Fernando Flávio Marques de Almeida. Beca Editora, São Paulo, 265-279.

Schultz, C.L., Scherer, C.M.S., Barberena, M.C. 2000. Biostratigraphy of the southern Brazilian Upper Triassic. Ver. Bras. De Geociências, v30, CD Version-2000:IGC-064.

Wankler, F. L. 2006. Arquitetura Deposicional e Compartimentação Estrutural do Aqüífero Passo das Tropas, na região de Santa Maria, RS: influências no comportamento Hidrogeológico. Tese de Doutorado, PPGEO/UNISINOS, São Leopoldo, 193 p.

Zerfass, H.; Lavina, E.L.; Schultz, C.L.; Garcia, A.J.V.; Faccini, U.F. & Chemale Jr., F. 2003. Sequence stratigraphy of continental Triassic strata of Southernmost Brazil: a contribution to Southwestern Gondwana palaeogeography and palaeoclimate. Sedimentary Geology, 161(1-2):85-105.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 46 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Stratigraphy and Sedimentology of the Late Paleozoic Glacial Record of the Paraná Basin: Brazil

Almério Barros França, Fernando Farias Vesely Petróleo Brasileiro S.A. – Petrobrás, Rio de Janeiro – RJ – Brazil, [email protected], [email protected]

Abstract The Itararé Group comprises the record of the Late Paleozoic glaciation in the Paraná Basin. The unit crops out in the eastern outcrop belt of the Paraná Basin and the most complete sections are in Paraná and São Paulo States. In the western and northern side of the basin, the Itararé Group is composed of sand-rich red bed units and is called the Aquidauana Formation. Diamictites, sandstones, shales, and rhythmites (varve-like) are the most common types of rocks present in the Itararé Group. Sedimentation took place mainly in marine settings, influenced by temperate glaciers flowing northwestward. The fossil content is meager, composed mostly of pollens and spores, except for the middle to uppermost interval where fish remains, foraminifers and even crinoids are present; most likely reflecting an improving in the life condition at the end of the long period of glaciation in Gondwana.

Introduction The Late Paleozoic glaciation in the Paraná Basin lasted for about 36 million years, from Late Carboniferous (Westphalian) to Early Permian (Sakmarian/Artinskian). The rock record of such a long period of time is the Itararé Group, a thick siliciclastic unit covering over one million square kilometers. The Itararé Group lies unconformably over Devonian rocks of the Ponta Grossa and Furnas formation, and locally over Precambrian basement. It is overlain by Permian post glacial sandstones and shales of the Rio Bonito and Dourados formations (Figure 1). The maximum thickness of the Itararé Group is about 1300m thick, in the State of São Paulo. A good average thickness throughout the basin is around 600 meters, Figure 2. The main purpose of this work is to present a sedimentological model and stratigraphic analysis for the Itararé Group. The data set used is based on the description of more than 600 meters of cores, outcrop studies, and well log analysis from more than 100 wells throughout the basin.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 47

Figure 1 - Partial Stratigraphic Chart for the Paraná Basin showing the Itararé Group. The western side of the basin is on the left hand side of the figure. Note deep channel cutting the basal section of the Itararé Group. The Roncador Layer (Westphalian) is an important key layer, both in outcrop and subsurface.

Figure 2 - Isopach Map of the Itararé Group (in meters), based on well data. The outcrop areas are shown in blue color. Circular symbols inside the basin are oil wells.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 48 Stratigraphy The Itararé Group is subdivided into three main formations: Lagoa Azul, Campo Mourão, and Taciba (França & Potter, 1991). The oldest, Lagoa Azul Formation is composed of a basal highly silicified sandstone unit and a widespread pebbly shale known as the Roncador Layer (Westphalian), which is an important key bed in the Itararé Group. There are good outcrops of the Roncador Layer, as well as cores in the central part of the basin, Figure 3. The Roncador Layer is characterized by black to greenish shales and rhythmites containing ice-rafted debris from granules to pebble sizes. Brownish color is also typical in cores and outcrops suggesting subaerial exposure during Late Carboniferous times. Gray to black slumped and massive diamictites are present locally. Another important aspect of the Roncador Layer is its typical anomalous high gamma ray log, very distinct for the entire Itararé Group. The Roncador Layer was deposited during an important maximum flood time that has covered basement areas such as in São Bento do Sul, northeastern Santa Catarina State, where the Roncador Layer lies directly on Precambrian granitic rocks. The fossil content according to Daemon & Marques-Toigo (1991) is predominantly composed of pollens and spores of the Interval G1 (Westphalian). Most common are: Potonieisporites cf. simplex; Krauselisporites; Potonieisporites neglectus; Potonieisporites novicus (P500); and tasmanites. The Figure 4 shows a correlation for the Roncador Layer from the central part of basin (Well Roncador) to the outcrop area in Paraná and Santa Catarina states.

A B

Figure 3 - A) Core from the Roncador Well (3594 m deep) illustrating the Roncador Layer; B) Outcrop of the Roncador Layer near Lapa in Paraná State. The brownish color is typical of the unit, both in outcrop and cores, suggesting exposures.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 49

Figure 4 - Geological cross-section showing the correlation for the Roncador Layer. Note the high gamma ray and the position of Core 5 (Figure 3A). Vila Velha and Lapa are outcrops in Paraná State and São Bento do Sul in Santa Catarina State. TS: Transgressive Surface; MFS: Maximum Flooding Surface; SB: Sequence Boundary. Modified from Franca et al. (1996).

The Roncador Layer was exposed due to a major sea-level drop, oxidizing shales and carving valleys (subaerial and/or subglacial) - running from South to North. These valleys are mapped by seismic and can be 4km large, and as deep as 200m, mostly filled with sandstones and conglomerates of the Campo Mourão Formation. The middle unit of the Itararé Group is the Campo Mourão Formation, a sand-rich unit interbedded with siltstones, diamictites and rhythmites, deposited from the Stephanian to middle Sakmarian (Figure 1). The most distinct features are the basal channelized sandstones filling paleovalleys, some of them drilled and cored in many wells for oil exploration. The Lapa Channel, which crops out near the city of Lapa in Paraná State, is the best outcrop example of these valleys and its geological column is shown in Figure 4. The Lapa Channel is about 65 km long, about 2 km wide in average, and 100m thick. The lithology inside the channel is composed of massive and stratified sandstones, with conglomerates in the lower sections (Figure 6). The Figure 5 is a “Google view” of part of the Lapa Channel by Lapa Town. Stratigraphic correlation is more difficult within the Campo Mourão Formation because of its poor fossil content and presence of laterally discontinuous channelized sandstone bodies. However, marine horizons constitute radioactive markers that can be traced hundreds of kilometers across the basin. Some examples are the Lontras-Guarauna and Ortigueira shales. The youngest unit of the Itararé Group is the Taciba Formation (Middle Sakmarian to Artinskian), composed of slumped and massive diamictites throughout the Paraná Basin. In southern areas the diamictites interfinger with black shales and rhitmites of the Rio Sul

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 50 Member. A sandy and porous unit underlies the diamictites for most of the basin. The Taciba Formation was deposited during a maximum flood as result of the major ice melting at the end of the Gondwana Glaciation. The gamma ray log shows the maximum flood surface near the top of the diamictites. The Taciba Formation and the Roncador Layer are excellent key units in the Itararé Group. The fossil content is basically pollens and spores: Krauselisporites sp (P-824); Cristatisporites sp.; Horriditriletes ramosus; Granulatisporites angularis; Complexisporites polymorphus. However, because of temperature improving at the end of the glacial time, it is also present brachiopods, gastropods, fish scales and arenaceous foraminifers (Weinschütz & Castro, 2004). The Figure 7 shows two examples of the main lithologies in the Taciba Formation. In a sequence stratigraphic perspective, the Itararé Group comprises several unconformity-bounded units, which are mostly composed of deglaciation facies successions deposited during major phases of glacial retreat (Vesely & Assine, 2006). Each sequence shows fining and thinning upward stacking patterns resulting from the retrogradation of proglacial glaciomarine depositional systems.

Figure 5 - Google image of part of the Lapa Channel, composed of Sandstones and conglomerates. In this image the “channel” is 1 km wide and 100 meters thick. Lapa town is on the left hand side.

Sedimentology. The sedimentological aspects for the Itararé Group are summarized in Figure 8, considering deposition related to a grounded marine glacier. Most of the glacial-related sedimentary rock of the Itararé Group is associated with gravity flows, ranging from slumps to turbidity currents. Distal facies, such as rhythmites and shales, result of suspension alone, or suspension plus traction.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 51 A B

Figure 6 - Campo Mourão Fm. A), Conglomerates and Coarse sandstones in the lowermost part of the Lapa Channel; B): Amalgamted massive, fine to medium sandstones in Lapa Channel.

A B

Figure 7 - Taciba Formation: A), Slumped diamictites with large boulders; B), Black shales of the Rio do Sul Member.

A typical facies is slumped diamictites, deposited originally by rain-out processes and sindepositional downslope movements. Interbedded and laterally associated with these diamictites occur conglomerates and cross bedded sandstones deposited in subaqueous outwash fans. Some of the mass movements may develop into sediment gravitational flows, segregating sand out of the main flow, depositing sandy debris flows and, ocassionaly bipartite flows may develop depositing turbidite-like deposits and thin bedded turbidites. The sandstones deposited herein are generally argillaceous with high gamma ray, low porosity, and are not considered reservoir rocks. These rocks were deposited during either, transgressive or high stand systems tracts. During low sea level times, incised valleys paved the way to depositing clean sandstones such as the Lapa Channel and the fairly good reservoirs in the channels of the Barra Bonita gas field. The channels were filled with sediment gravity flows issued from glaciers, some of them likely as jökulhlaups, as well as by periglacial rivers. França et al

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 52 (1996) and D’Ávila (1999) are good sources about the depositional environment of the Lapa Sandstone. By this context, two other trigger mechanisms for turbidity current initiation are also suggested: 1. Hyperpycnal flows derived from flood-dominated proglacial rivers and 2. hyperconcentrated jet flows emanating from a grounded marine glacier. These dynamics are suggested by the association of turbidites with prodeltaic facies and subaqueous outwash respectively. The distal facies of the channelized sandstones are massive amalgamated sandstones as thick as 100m, cropping out as distinct lobes (30 meters thick each) separated by thin clay beds (Figure 9). The lobes are good reservoirs and distinguish from the channelized sandstones for its much better sorting and lacking of rock fragments. Lodgement tills and subaerial outwash deposits are present but quite rare in the Paraná Basin. Striated and grooved surfaces are common, occurring on basement rocks (Precambrian to Devonian) as well as on intraformational soft-sediment substrata. These features indicate that ice flowed from southeast to northwest during Itararé Group deposition (Figure 10).

Figure 8 - Schematic depositional settings for the Itararé Group, considering processes and facies related to a grounded marine glacier.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 53

Figure 9 - Sandstone lobes occurring distal to channelized sandstones in Vila Velha State Park, Paraná State. The lobes are separated by thin layers of shales.

A B

Figure 10 – A) Directions of Late Paleozoic ice flow in the Paraná Basin inferred from striated surfaces. Black arrows: soft-sediment surfaces; white arrows: hard rock surfaces (after Gesicki et al, 2002 and Vesely, 2006). B) Example of a soft-sediment striated surface developed on proglacial sandstone (flow towards the right side of the photo).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 54 References Daemon, R.F. & Marques-Toigo, M., 1991, An integrated biostratigraphic column for the Paraná Basin, Brazil. In International Congress of Carboniferous-Permian Stratigraphy and Geology, 12, Buenos Aires, September 1991. Resumos, p. 25.

D'Ávila, R.S.F. 1999. Análise de Fácies e Estratigrafia Física do Arenito Lapa, Grupo Itararé, Bacia do Paraná, Brasil. Porto Alegre. 349 p. (Dissertação de Mestrado, Instituto de Geociências, Universidade Federal do Rio Grande do Sul).

França, A.B.,Potter ,P.E., 1991, Stratigraphy And Reservoir Potential Of Glacial Deposits Of The Itarare Group (Carboniferous-Permian), Parana Basin, Brazil.. American Association of Petroleum Geologists Bulletin. Tulsa, Estados Unidos: , v.75, p.62 - 85, 1991.

Franca,A.B., Winter,W.R., Assine,M.L., 1996, Arenitos Lapa-Vila Velha: Um modelo de trato de sistemas subaquosos canal-lobos sob influência glacial, Grupo Itararé (C-P), Bacia do Paraná.. Revista Brasileira de Geociências. Brasília, DF:, v.26, n.1, p.43 – 56.

Gesicki, A. L. D., Riccomini, C., Boggiani, P. C. 2002. Ice flow direction during Late Paleozoic glaciation in western Paraná Basin, Brazil. Journal of South American Earth Sciences 14, 933-939.

Vesely, F.F. 2006. Dinâmica Sedimentar e Arquitetura Estratigráfica do Grupo Itararé (Carbonífero – Permiano) no Centro-leste da Bacia do Paraná. Curitiba. 226 p. (Tese de Doutorado, Departamento de Geologia, Universidade Federal do Paraná).

Vesely, F.F.; Assine, M.L. 2006. Deglaciation sequences in the Permo-Carboniferous Itararé Group, Paraná Basin, southern Brazil. Journal of South American Earth Sciences, v. 22, p. 156-168.

Weinschütz, L.C., Castro, J.C., 2004, Arcabouço cronoestratigráfico da Formação Mafra (intervalo médio) na região de Rio Negro/PR-Mafra/SC, borda leste da bacia do Paraná. Geociências, Rev. Esc. Minas de Ouro Preto, 57 (3), p. 151-156, junho 2004.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 55 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Correlation of Neoproterozoic terranes between SE-Brazil and Africa: comparative tectonic evolution and open questions

Monica Heilbron1,3; Claudio M. Valeriano1,3; Colombo C. G. Tassinari2,3; Júlio Almeida1,3; Miguel Tupinambá¹; Oswaldo Siga Jr2,3 ¹ TEKTOS Research Group, Faculdade de Geologia, Universidade do Estado do Rio de Janeiro (UERJ), heilbron @uerj.br , Rua São Francisco Xavier 524/4006- A, Maracanã, 20559-900, Rio de Janeiro, Brazil. ² CPGeo/IG-USP, Centro de Pesquisas Geocronológicas, Instituto de Geociências, Universidade de São Paulo (USP). Rua do Lago, Cidade Universitária, São Paulo, SP, Brazil. ³ Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPQ

Abstract A comparative tectonic model for Southeastern Brazil and Western Africa in the scenario of Western Gondwana is presented, based on geological and geochronology data, whereby ca. 790 and 610 Ma major episode of generation of magmatic arc, both intra-oceanic and cordilleran, along both sides of the Adamastor Ocean. Diachronic collisions of the arc terranes and small plates followed at ca. 600, 580, 560 and 530 Ma. The São Francisco-Congo and Angola palaeocontinents were probably not monolithic, rather may have accommodated considerable convergence during the Brasiliano-Panafrican episodes. The final docking of Cabo Frio and Kalahari in the Cambrian was coeval with the arrival of Amazonia on the opposite side, resulting in lateral reactivation and displacement between the previously amalgamated pieces. The transition between the Cambrian and the Ordovician is marked by the extensional collapse of the metamorphic core zones of the orogens, with regional cooling and scattered bimodal magmatism probably related to slab detachment and underplating process.

Introduction The Ribeira Belt is one the Neoproterozoic belts of West Gondwana, extending for 1400 km along the Atlantic coast of SE Brazil. The African counterpart, in Angola and Namibia, is represented from north to south by the West Congo Belt, the Angola craton and the Kaoko Belt.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 56 Tectonostratigraphic terranes of southeast Brazil and southwest African counterparts do not provide a clear match when detailed palaeogeographical reconstructions of West Gondwana are considered. Comparison of both sides of the Atlantic yields reasonable continuity between terranes of the northern segment of the Araçuaí Belt and of the West Congo Belt. South of this area, however, there is mismatch between the Ribeira Belt (SE Brazil), including the Rio Negro magmatic arc (790–620 Ma) and the Cabo Frio terrane, and the southern segment of the intra-continental West Congo Belt and the basement of the Angola craton, south of Luanda. A detailed tectonic model for the Neoproterozoic evolution of the belt related to its African counterparts will be presented, besides the discussion of some open questions. All the data summarized is discussed in detail in a review by Heilbron et al. (2007, in press).

Pre-Gondwana palaeo-continents: Archaean to Mesoproterozoic history The major Archaean to Mesoproterozoic palaeo-continental blocks that collided in the Neoproterozoic to form the southern sector of the West Gondwana are the São Francisco/Congo, Angola, Rio de la Plata, Paranapanema (presently covered by the Phanerozoic Paraná basin), and Kalahari cratons. Smaller fragments, possibly micro- continents, were involved: the Curitiba/ Luis Alves, Paraíba do –Embú, Apiaí and Cabo Frio terranes. The basement of the São Francisco palaeo-continent exposed in the reworked southern passive margins is typically composed of Palaeoproterozoic plutonic or supracrustal metamorphic rocks (2.2 to 1.9 Ga), but containing well characterized Archaean nuclei (2.8– 2.6 Ga). Platform conditions in the São Francisco–Congo palaeo-continent were only achieved after a very important and widespread orogeny around c. 2.05 Ga (“Transamazonian” event). A very similar situation is reported for the basement of the Congo area, where rocks formed predominantly between c. 2.9 Ga and 2.1 Ga, followed by metamorphism at 2.05 Ga. Relicts of Palaeoproterozoic supracrustal successions with sedimentary context suggesting a passive margin setting is also present in the classic area of the “Quadrilátero Ferrífero” (Iron Quadrangle), Minas Gerais State, Brazil. An additional characteristic of the tectonic history of the São Francisco– Congo palaeo-continent is a very important Statherian (c. 1.7 Ga) episode of continental rifting recorded by the diamond-bearing clastic successions and subordinated felsic volcanic rocks of the Espinhaço Supergroup. The basement associations in the Paraíba do Sul–Embu and Curitiba terranes are also characterized by c. 2.2 to 2.1 Ga orthogneisses and granulites with Nd isotopic signatures indicating derivation from Archaean crust. In the Cabo Frio terrane the basement association (Região dos Lagos complex) is comparatively younger, with the majority of isotopic ages of crystallization between 2.0 and 1.9 Ga.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 57 The basement of the Apiaí terrane contrasts with the other blocks of eastern Brazil with the occurrence of expressive Mesoproterozoic volcanosedimentary sequences, besides Palaeoproterozoic orthogneisses (c. 2.2–1.7 Ga) and granitic rocks. These Mesoproterozoic sequences are thought to represent both extensional settings, rift to ocean floor, and compressional settings, ocean floor to volcanic arc. The Angola and Kalahari cratons in SW Africa display contrasting tectonic associations with predominant Archaean rocks affected by several episodes of magmatism during the late Palaeoproterozoic and Mesoproterozoic.

Neoproterozoic passive margins The following geological features mark the onset of continental rifting in the São Francisco–Congo and Angola Kasai palaeo-continents during the Mesoproterozoic– Neoproterozoic transition: a) the intrusion of c. 900 Ma mafic dyke swarm in the São Francisco craton; b) Tonian alkaline plutons in the Araçuaí Belt; c) c. 999 and 912 Ma expressive bimodal volcanism in the West Congo Belt; and d) dolerites and gabbros of 1.2– 1.1 Ga in southern Angola. This important rifting event was successful enough to evolve into spreading of oceanic lithosphere and consequently to the isolation of several palaeo-continents, such as São Francisco, and smaller fragments that later amalgamated to form Gondwana Although no complete ophiolitic relicts have been recognized in the central sector of the Ribeira Belt, MORB-type metabasic rocks dated at 848 ± 11 Ma (U–Pb TIMS) are interpreted as representing the oldest magmatic event compatible with the generation of oceanic lithosphere between the São Francisco palaeocontinent and the Oriental terrane. This matches the similar 816 ± 72 Ma Sm–Nd isochron age interpreted to date the formation of ocean floor in the Araçuaí Belt, the along-strike continuation of the Ribeira Belt to the north. The passive margin successions around the São Francisco palaeocontinent are represented by the siliciclastic Andrelândia Megasequence in the Occidental terrane. In Africa the Neoproterozoic successions were deposited in the West Congo rift to sag basin. The reworked Palaeoproterozoic basement rocks crop out along the northern coast of Angola. The Neoproterozoic passive margin of the western Angola-Kasai palaeo- continent is represented by the sedimentary successions (Namibian Series) of the central and external domains of the Kaoko Belt. The Angola craton as the source area is clearly indicated by U–Pb ages of detrital zircons and by palaeocurrent distribution in turbidites of the southern part of the Kaoko Belt. The maximum age of sedimentation of the West Congolian Group is constrained by U–Pb data to around 650 Ma. Orogenic inversion during the Neoproterozoic (c. 566 Ma) developed through thrusting of the rift succession over the sag-basin deposits of the West

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 58 Congolian Group. The Inkisi Group is a post-orogenic succession that completes the depositional history of the region.

Cryogenian to Ediacaran subduction Eastward subduction of the oceanic lithosphere belonging to the São Francisco palaeo-plate caused the development of a long-lived magmatic arc that developed from c. 790 to 610 Ma, presently incorporated in the Ribeira Belt as the Rio Negro complex. The Rio Negro arc was emplaced into passive margin metasedimentary successions of the Oriental terrane, with tonalites, granodiorites, granites and gabbros with juvenile isotopic signatures. Similar granitoid rocks of c. 630–610 Ma in the Apiaí terrane were also described as formed in magmatic arc environment. Generation of granite took place in the Coastal terrane of the Kaoko Belt simultaneously to the generation of the Rio Negro arc, at c. 655–645 Ma. This was followed by the generation of c. 635 Ma granites and by an episode of highgrade metamorphism. The country rocks of these arc intrusive complexes in the Coastal terrane in the Kaoko Belt are cordierite-garnet-biotite gneisses, very similar to those of the Oriental terrane in the central Ribeira Belt. The summary above indicates long-lived subduction of oceanic lithosphere, with generation of intra-oceanic to cordilleran magmatic arcs within and around the Adamastor ocean from c. 790 Ma to 600 Ma. This scenario is not compatible with a palaeogeography of a narrow ocean between the São Francisco–Congo and Angola palaeo-continents during the Neoproterozoic. Subduction was associated with the development of fore-, intra- and backarc basins that were inverted during several subsequent tectonic episodes. This is exemplified by the volcano-sedimentary record of the Neoproterozoic basins of the Apiaí and Cabo Frio terranes.

Late Ediacaran (c. 605–550 Ma) major continental amalgamation Between 605 and 550 Ma most pieces of West Gondwana were already attached together. Using the São Francisco–Congo craton as reference, this time interval is marked by diachronous collisions of Palaeoproterozoic microplates and Neoproterozoic magmatic arcs. The available geochronological database indicates that docking of the Socorro terrane took place between c. 640 and 610 Ma related to the evolution of the southernmost Brasília Belt. The docking of the Apiaí and Curitiba–Paraíba do Sul–Embú terranes followed at c. 605–580 Ma. Finally, docking of large magmatic arcs of the Oriental and Coastal terranes took place at c. 580–550 Ma, both in the Ribeira and Kaoko belts.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 59 These tectonic episodes were dominated by the oblique convergence of colliding terranes and resulted in pervasive deformation observed in the Ribeira and Kaoko belts. Deformation began with frontal thrusting with isoclinal to tight folds and a low dipping foliation defined by metamorphic minerals related to the main metamorphic event. The deformation evolved to a late phase with a major right-lateral component in the Ribeira Belt, and left- lateral in the Kaoko Belt. Coeval metamorphic parageneses in the collisional zones indicate intermediateto high-pressure regimes in the lower plates (reworked passive margins) and of lower pressure regime in the upper plates (magmatic arc terranes). Around 560 Ma, closure of the rift and sag successions of the West Congo Belt was in course. Latest stage of West Gondwana amalgamation in the Cambrian (c. 530–510 Ma) By c. 550 Ma the majority of blocks of the central portion of West Gondwana were attached together. However, an important Cambrian tectonicmetamorphic event is recorded around c. 530–510 Ma along the border of proto- West Gondwana. It is recorded in the Ribeira Belt and in the Kaoko and Damara belts, but is also detected in the Dom Feliciano, Paraguai-Araguaia and Pampean belts. The tectonic setting that emerges from this tectonic correlation is that of the closure of another large ocean around proto-West Gondwana with the arrival of large blocks such as the Amazonian and Kalahari palaeo-continents, and of minor blocks such as the Cabo Frio terrane in the Ribeira Belt, and the Pampean terrane in Argentina. Cambrian–Ordovician (c. 510–480 Ma) magmatism: a record of the collapse of Gondwana orogens? Post-collisional deformation recorded in the Oriental and Cabo Frio terranes in central Ribeira Belt marks the transition to extensional tectonic regimes, represented by two groups of structures: a) N–S to NE–SW brittleductile subhorizontal shear zones striking parallel to the orogen with down-dip movement and associated down-dip verging folds; and b) dextral transtensional subvertical NW–SE shear zones that are transversal to the orogen. In the Oriental terrane, this deformational episode is associated with widespread post- collision calc-alkaline granitic intrusions in the form of circular stocks, sills or dykes, with U– Pb ages between 510 and 480 Ma. The shear zones behaved as channels for the ascending magmas as shown by frequently observed magmatic flow structures. An important characteristic of these intrusive bodies is the common association with mafic enclaves and magma mingling along the border zones. In the Cabo Frio terrane this tectonic episode is represented by pegmatites striking parallel to NW-SE subvertical shear zones. The extensional collapse following thrust-stacking was possibly triggered by slab detachment, causing exposition of the high temperature core of the orogen and associated bimodal magmatism.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 60 Open Questions for discussions Several important open points arise from the comparison of the geological evolution of the Ribeira Belt and Southwest African pre-Atlantic counterparts: a) how do the pieces of the Neoproterozoic collage fit? b) How to accommodate juvenile arcs and a multiple collision history with the frequently postulated narrow ocean between the cratonic blocks of South America and Africa during the Neoproterozoic? c) Did the São Francisco–Congo and Angola cratons compose a monolithic block since the Palaeoproterozoic? d) When was the assembly of West Gondwana finally completed?

References Basei, M. A. S., Frimmel, H.E., Nutman, A. P., Preciozzi, F. & Jacob, J. 2005. A connection between the Neoproterozoic Dom Feliciano (Brasil/Uruguay) and Gariep (Namibia/south Africa) orogenic belts - Evidence from a reconnaissance provenance study. Precambrian Research, 139, 195-221.

Bossi, J. & Gaucher, C. 2004. The Cuchilla Dionisio terrane, Uruguai: An Allochthonous Block Accreted in the Cambrin to SW-Gondwana. Gondwana Research, 7, 661-674.

Campos Neto, M.C. 2000. Orogenic Systems from Southwestern Gondwana, an approach to Brasiliano-Pan African Cycle and Orogenic Collage in South-eastern Brazil. In: Cordani, U.G., Milani, E.J., Thomaz Filho, A., & Campos, D. A. (eds). Tectonic Evolution of South America, 335-365.

Carvalho, H. & Tassinari, C.C.G. 1992. Idades do magmatismo da região do Caraculo-Bibaia (SW deAngola) e sua implicações na correlação geológica com o Cinturão Ribeira no sudeste do Brasil. Revista Brasileira de Geociências 22, 73-81.

Cordani, U.G., Dágrella-Filho, M.S., Brito Neves, B.B. & Trindade, R.I.F. 2003. Tearing up Rodinia: the Neoproterozoic paleogeography of South American cratonic fragments. Terra Nova 15(5), 350-359

Frimmel, H. E., Tack, L., Basei, M. S., Nutman, A. P. & Boven, A. 2006. Provenance and chemostratigraphy of the Neoproterozoic West Congolian Group in the Democratic Republic of Congo. Journal of African Earth Sciences, 46, 221-239.

Goscombe, B., Gray, D., Armstrong, R., Foster, D., Hand, M., Mawby, J. & Vogl, J. 2005. Event geochronology of the Pan-African Kaoko belt, Namibia. Precambrian Research,

Goscombe,B., Hand, M., Gray, D., & Mawby, J. 2003. The metamorphic Arquitecture of a Tranpressional Orogen: The Kaoko Belt, Namibia. Journal of Petrology, 44, 679-711 Heilbron M. & Machado N. 2003. Timing of terrane accretion in the Neoproterozoic-Eopaleozoic Ribeira orogen (SE Brazil). Precambrian Research, 125, 87-112.

Heilbron, M., Machado, N., Simonetti, T. & Duarte B. 2003. A Palaeoproterozoic orogen reworked within the Neoproterozoic Ribeira Belt, SE Brazil. In: IV South American Symposioum on Isotope Geology. Salvador, Brazil. August 2003. Short Papers, 1: 186-189.

Heilbron, M., Pedrosa-Soares, A.C., Campos Neto, M., Silva, L.C., Trouw, R. A J. & Janasi, V.C. 2004b. Brasiliano Belts in SE Brazil. Journal of Virtual Explorer, 17, www.virtualexplorer.au

Heilbron, M.; Valeriano, C.M., Tassinari, C.G., Almeida, J.C.A., Tupinambá;M.,Siga Jr, O. 2007 (in press). Correlation of Neoproterozoic Terranes between SE-Brazil and Africa: Comparative Tectonic Evolution and Open Questions. In: Pankhurst, R.J., Brito-Neves, B.B., de Witt, M.J., Trouw, R.R.(Eds). Western Gondwana Pre-Atlantic Reconstructions: Geological Society of London Special Paper, in press

Janasi, V.A., Leite, R.J. & Van Schmus, W.R. 2001. U–Pb chronostratigraphy of the granitic magmatism in the Agudos Grandes Batholith (west of São Paulo, Brazil) – implications for the evolution of the Ribeira Belt. Journal of South American Earth Sciences, 14, 363-376.

Juliani, C., Hackspacher, P., Dantas, E. & Fetter, A.H. 2000. The Mesoproterozoic volcanosedimentary Serra do Itaberaba Group of the central Ribeira Belt, São Paulo State, Brazil: implications for the age of the overlying São Roque Group. Revista Brasileira de Geociências, 30, 82-86.

Kröner, a & Cordani, u. 2003 African, southern India and South American cratons were not part of Rodinia supercontinent: evidence from filed relatioships and geochronology. Tectonophysics, 375,325-352

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 61 Passchier, C. W., Trouw, R.A.J., Ribeiro, A. & Paciullo, F.V.P. 2002. Tectonic evolution of the Southern Kaoko belt, Namibia. Journal of African Earth Sciences, 35, 61-75.

Pedrosa-Soares, A.C., Noce, C. M., Wiedemann, C. & Pinto, C.P. 2001. The Araçuaí-West- Congo Orogen in Brazil: An overview of a confined orogen formed during Gondwanaland assembly. Precambrian Research, 110 (1-4), 307-323.

Schmitt, R.S., Trouw, R.A.J., Van Schmus, W.R. & Pimentel, M.M. 2004. Late amalgamation in the central part of Western Gondwana: new geochronological data and the characterization of a Cambrian collision orogeny in the Ribeira Belt (SE Brazil). Precambrian Research, 133, 29-61.

Seth, B., Kronner, A., Mezger, K., Nemchin, A.A., Pidgeon, R.T., & Okrusch, M. 1998. Archaean to Neoproterozoic magmatic events in the Kaoko belt of Nw Namibia and their geodynamic significance. Precambrian Research, 92, 341-363

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 62 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Examples of climatic, tectonic and eustatic controls on the stratigraphic signatures of the Early Permian succession in the Paraná Basin

Michael Holza,1; Juliano Küchlea,2; Paula Dariva dos Reisa, Junia Casagrandea,2 aInstituto de Geociências, Universidade Federal do Rio Grande do Sul, Brazil 1 CNPq researcher, 2 PPGGeo/UFRGS and Agência Nacional do Petróleo -ANP-PRH-12 (corresponding author: [email protected])

Abstract Debate on control of sedimentation is almost as old as sedimentary geology and the recognition and distinction of those facts in the sedimentary record is not always clear or easy. The present paper aims on some examples regarding climatic and tectonic control on specific stratigraphic signatures of the Early Permian succession of the Paraná Basin in southernmost Brazil and to trigger a debate about these factors and the manner to recognize their relative role in the sedimentation history of a basin. This is archived by showing regional examples of tectonic and climatic control on the sedimentary record of the Sakmarian/Artinskian, coal-bearing Rio Bonito Formation.

Introduction Discussion about climate, sediment supply, eustacy and tectonics as the main controls on sedimentation and its cyclicity is almost as old as sedimentary geology. For the Paraná Basin, the traditional viewpoint has been that of an intracratonic, “stabilized” basin with “no tectonic influence” if compared to rift or foreland basins. Hence, eustasy has traditionally been seen as the main if not only important control on sedimentation in that basin. However, since the papers by Cloething in the late eighties (e.g., Cloething, 1988) and the studies by De Wit and Ransome (1992) Milani & Ramos (1998), base upon the structural evolution of the Cape Fold Belt (South Africa) and the La Ventana trust belt (Argentina), the intraplate stress induced by subduction and accretion on the occidental margin of Gondwana (i.e., the Gondwaindes belt) are responsible for reactivation and uplifts of large terrains,

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 63 hence controlling accommodation space and overprinting a tectonic signal to the supposed eustatic control of the sedimentary infill of basins such as the Paraná Basin (e.g., see the second order sequence stratigraphy of the Paraná Basin by Milani, 2000). The present paper intents to show the reflex of this tectonism upon the third-order depositional sequences. The climatic control is also aimed on, we want to discuss the role of paleowind directions as main control on specific stratigraphic signatures and to show how specific confluences of climatic conditions and paleogeography can influence sedimentation and induce variations within the basin.

Tectonic versus eustatic control The Paraná Basin has been emplaced over different geotectonic domains, comprising Archean and Paleoproterozoic cratonic terrains and Neoproterozoic mobile belts related to the Pan-African and Brasiliano Cycles, responsible for the assemblage of Gondwana. Hence, the geotectonic framework of the basement is characterized by several cratonic blocks and intervening mobile belts, forming a complex framework of lineaments and crustal discontinuities. This complexity is also present in a regional scale: the southern margin of the Paraná Basin is nowadays forming the shield areas of the Santa Catarina and the Rio Grande do Sul states in southernmost Brazil, and both shield areas are tectonically dominated by a NE-SW transcurrent shear zone and by NW-SE associated fault and fracture systems. This tectonic patchwork defined several geotectonic blocks as shown in figure 1.

Figure 1 – The shield areas of the southern Brazil is actually a complex tectonic patchwork, indicating that several lithologically different blocks form the basement of the Paraná Basin in a regional scale (cf. Holz et al., 2006). Note the dotted line which indicates the correlation section shown in figure 3. The limits of the tectonic blocks were projected basinwards in order to depicture the kind of basement at each stratigraphic profile (e.g. the basement of profile HV-60 is block C, while the adjacent profile DP-01 has block A as its basement).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 64 The Early Permian succession has an overall transgressive sedimentation regime, recorded by a clear retrogradation of the facies belt (Figure 2). The sequence stratigraphic framework established by Holz et al. (2000) has shown that the sequences have variable thicknesses and alongstrike variation of depositional systems, suggesting differential subsidence, hence, a tectonic overprint on the eustatic control of the stratigraphic signature (Figure 3).

Figure 2 – Stratigraphic overview, showing sequence stratigraphic framework and lithostratigraphic equivalents of

the studied interval. Lithostratigraphy is based on Schneider et al (1974), third order sequence stratigraphy is from

Holz et al. (2000) Abbreviations: SB - sequence boundary, TS -transgressive surface, MFS - maximum flooding

surface, LST - lowstand systems tract, TST - transgressive systems tract, HST - highstand systems tract.

Figure 3 - Regional correlation section, the location is indicated in figure 1. The section shows the distribution of

the Early Permian sequences, enhancing the coal-bearing, fluvial to paralic sequence 2. as established by Holz et

al. (2000). Datum is the maximum flooding surface of this sequence. Note the presence of depressions preserving

sequence 1 and thickness variation of sequence 2 close to the limits of the tectonic blocks (cf. Holz et al. 2006).

Hence, it seems clear that the different reology of the tectonic blocks which form the basement have a decisive influence on the stratigraphic signature.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 65 Correlation of sequences Seq-2 and Seq-3 (see Figure 2) in a regional scale (i.e., from Rio Grande do Sul to Santa Catarina) reveals that there is a clear contrast between the Santa Catarina and Rio Grande do Sul areas, because the marine interval (Paraguaçú Member) in the north can be correlated to the coal-bearing coastal systems in the south (Figure 4). Therefore, we interpret that in Rio Grande do Sul, the coal seams are linked to the transgressive systems tract of sequence 2, while the Santa Catarina area, by this time, was mostly a marine embayment setting, with a stratigraphic record expressed by tens of meters of lenticular and massive mudstone and fine sandstones with wavy and hummocky cross bedding. Hence, the retrogradational coal-bearing barrier system mapped as nondivided Rio Bonito Formation in the Rio Grande do Sul state is time-equivalent to the marine rocks labeled the Paraguaçu Member of the Rio Bonito Formation in Santa Catarina State. Sequence stratigraphic analysis suggests that both lithostratigraphic units are parts of the same transgressive systems tract (labeled TST2 - see correlation in figure 4).

Figure 4 - Correlation between the Early Permian succession of Rio Grande do Sul and Santa Catarina states, based upon sequence stratigraphic analyses. Datum is the maximum flooding surface of sequence 2. Note that the marine interval called the Paraguaçú Member in Santa Catarina is equivalent to the coal-bearing shoreface- barrier system in Rio Grande do Sul. After formation of sequence boundary SB 3 the main source area is Santa Catarina, with sandstones and coals assigned to the Sideropolis Member. This member is time-equivalent to offshore facies in Rio Grande do Sul, labeled the Palermo Formation. (cf. Holz et al. 2006).

The stratigraphic relations of sequences 2 and 3 within the study area, regarding the contrast between marine and coastal sedimentation, seems to indicate tectonic movement. The fact is that the northern part of the study area (i.e., Santa Catarina State) was a marine embayment by the time the southern region had fluvial and coastal sedimentation, and than, after formation of sequence boundary SB 3, the opposite occurs: most of the southern area recorded shallow marine facies, while the northern area shows recorded fluvio-deltaic and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 66 coal-bearing coastal sediments (Figure 4). Hence, from the epoch of the maximum flooding surface of sequence 2 to the epoch of lowstand to trangressive systems tract deposition of sequence 3 the shoreline configuration has dramatically changed. The sin-SB 3 relief inversion and the uplift of areas which were marine during most of the deposition of the previous sequence strongly suggests tectonic control, i.e., the formation of sequence boundary SB 3 is not of eustatic origin (or not exclusively), but has a strong tectonic component in its origin.

Figure 5 - Schematic map comparing the shoreline configuration before and after sequence boundary SB 3 has

been formed. Note that after formation of sequence boundary SB3 most of the southern area recorded shallow

marine facies due to a up to 60 km coastal encroachment, while the northern area recorded fluvio-deltaic and

coal-bearing coastal sediments due to regression (cf. Holz et al. 2006).

Our conclusive opinion about this issue is that basin analysis should include this decisive methodogical step: it is useful to project the basements heterogeneities (as depictured by surface maps, such as the one shown in figure 1) into the basin, hence, producing a kind of geological map of the basin’s basements, in order to take the heterogeneity and the possible differential subsidence into account when analyzing the stratigraphy of the basin, and be aware of along-strike variations, because depending on local subsidence rate and its alongstrike variation and on the presence of point-sourced feeding systems with frequent shift of input loci, a sequence may be developing lowstand, transgressive and highstand sedimentation at the same time, (e.g., Wehr, 1993; Helland- Hansen & Martinsen, 1996), as demonstrated by the above example.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 67 Climatic control on stratigraphic signatures Concerning climatic control, the occurrence of coal seams, a climate sensitive facies, can be used as an example. In southernmost Brazil, there are areas with extensive and economic important coal occurrences, while in other areas almost no coal occurs. What factor controls this uneven and patchy distribution of the coals of southernmost Paraná Basin? A lesson can be learned analyzing the Candiota area (Figure 6 A, 6 B), where the most important coal seams in southernmost Brazil occur. The coal seams are genetically linked to paralic sandstones formed in a barrier/lagoon system (e.g., Holz, 2003). Lithostratigraphically it corresponds to the Rio Bonito Formation and its equivalent in Uruguay, the Tres Islas Formation. During Early Permian, the study area was located approximately 430 South. The open sea was located westwards, and the eastern region of Uruguay and the southwestern region of Rio Grande do Sul state were characterized by an irregular palocoastline, with at least one huge embayment (Candiota paleobay, Figure 6 B).

Figure 6 – ( A) location map of the Paraná Basin and the study area; (B) paleogeographic reconstruction of the study area, enhancing shoreline configuration and coal occurrences. Note that significant coal occurrences are linked to specific loci such as the Candiota paleobay, while in the northern shoreline no coal is present.

This paleophysiographic reconstruction was based upon a database integrated by more than a hundred well logs and complementary outcrop analysis. Coal and sandstone facies distribution is clearly not uniform along the paleoshoreline (Figure 6 B). The most extensive coal formation occurs within the Candiota mining area, with cumulative coal

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 68 thickness of up to 18 meters. Adjacent areas show minor coal occurrences. On the eastern border of the basin, almost no coal is registered in the Rio Bonito/Tres Islas succession. This peculiar concentration of coal seams in particular areas rises the question about the depositional control. Why does coal accumulate only in certain areas? What is the role of the shoreline configuration on the coal and sand facies distribution? What kinds of dynamics were controlling deposition on the Uruguayan-Brazilian Early Permian paleocoastline? The answer can be found by a paleoclimatic approach. Data provided by Patzkowsky et al. (1991) for the Early Permian shows a unveiling retrodiction: a low pressure cell over southwestern Gondwana during summer and prevailing wind directions from south and southwest (Figure 7 A), while during winter season, a high pressure system formed in that region of Gondwanaland due to the cooling of the extensive land mass, hence the prevailing winds are retrodicted as moving from east/northeast to west/southwest (Figure 7 B). Based on these retrodiction, the climate within the study area was humid almost all around the year, and very cold: the Early Permian is the time of maximum Paleozoic icehouse effect with very low winter temperatures (e.g, Crowley et al, 1989; Visser, 1983). The wind regime was alternating from southern winds in summer to western wind during winter. These winter winds must have been intense due to the fact that the study area is located close to a zone of high barometric gradient (note dense spacing of the isobars 1025 to 1010 at figure 7 B). Overlying the retrodicted wind regime and the regional paleogeographic reconstruction (Figures 7 C and D), one can note that during summers the northern paleocoastline and the area of the here called Candiota paleobay received almost frontal wind, while during winter the winds came from the highlands in the east.

Figure 7 - Global atmospheric circulation during the Early Permian during (A) summer and (B) winter in the southern hemisphere (after Patzkowsky l, 1991). The gray area is the Pangea, the black rectangle indicates the paleoposition of the study area. Figures (C) and (D) show the prevailing winds during summer and winter in the study area, as retrodicted from the global model.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 69 For epicontinental seas, relatively isolated from the open oceanic wave conditions, the most important control on wave propagation is the wind direction and its seasonal variation (e.g., Carter, 1988). Insofar, the peculiar wind regime in the study area certainly had a strong control on the wave action and controlled the accumulation of coaly and sandy facies. The main depositional process on the eastern paleoshoreline must have been controlled by river mouth bypassing, delivering sediments to the platform. During the summers, there was a strong longshore drift towards north, reworking and transporting almost all of the sediment input from the eastern shoreline, precluding the development of broad beaches and strandplains. The bulk of sediment input on the eastern shoreline was transported towards north, and therefore the most important sand accumulation and the consequent development of widespread strandplains, occurred on the northern part of the paleoshoreline, where the important coal seams were formed. During winter, the wave direction changed doe to the east-to-west directed winds, and the paleoshoreline must have experienced an increasing fluvial input, because the heavy raining during the winter season must have increased the rate of sedimentary influx. During the subsequent summer, most of these sediments were re-mobilized, crossing the marine platform towards north. The fact that the northern shoreline had two main headlands (delimiting the Candiota paleobay) must have induced wave refraction. As studies on modern shorelines conclusively show (e.g., Carter, 1988), an unrefracted wave, as it comes closer to an irregular shoreline, undergoes differential dispersion of its energy. The wave segment heading towards the headlands encounters the sea floor sooner than the wave segment heading to the bay, and therefore decreases its velocity. The difference in speed of propagation causes the wave front to bend and the energy contained in the segment close to the headlands is concentrated, while the energy of the segment in front of the embayment is dispersed. Therefore, sediments are entrapped in the area between the headlands, forming beaches and strandplains. That is what occurred in the area of Candiota Paleobay: the areas around the headland of the embayment have a relatively narrow strandplain and few coal, while the strandplain within the embayment is broad and developed extensive and thick coal seams Two facts, (1) during summers the wave energy was frontal to the strandplains on the northern part of the shoreline and (2) the retrodicted paleoclimatic condition indicating strong winter storms, may be the explanation for the minoritary presence of tidal features in the rock succession preserved within the Candiota paleobay. Although the paleogeographic configuration should favor tides to be enhanced, there are not only tidal features in the rock succession, but many wave-induced features also, characterizing a mixed - wave and tide- influenced depositional system, as depictured by Holz & Kalkreuth (2002) and Holz (2003). Under the bottonline, one can state that the coal occurrence on the shoreline in southeastern border of the Paraná Basin had a climatic control allied to peculiar

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 70 paleogeographic settings. The coal accumulates only in those areas far away enough from the active shoreline to develop coastal mires under a peculiar climatic regime which controlled the wind and sediment transport regime at the paleoshoreline. More northwards (in the region of Rosario do Sul city) this conditions are not given, and no coal occurs in that region. Insofar, we think that basin analysis should take the clima retrodiction into account, not only in a basinwide scale like it’s normally done, but even in a local scale, as demonstrated by the example above.

Conclusions Tectonic influence in a third order scale can be detected by alongstrike variation of the thicknesses and depositional systems of the depositional sequences and its close association with the tectonic blocks, delimited by the main lineaments and associated faults of the basins basement. In the specific example shown herein, the Artinskian/Kungurian tectonic event, generating the sequence boundary SB3, has caused a regional inversion of the paleorelief in the study area. Coal of the Paraná Basin, on the other hand, seems to have had a strong climatic control, because specific paleoclimatic conditions induced a peculiar wind regime, which in some loci leaded to the development of broad strandplains with protected mires, while in the other parts of the paleo-shoreline the wave energy formed narrower standplains, where few or no coal was formed due to the proximity of the active shoreline. This teaches us that climatic retrodiction should be considered even in a local scale, not only in basinwide analysis.

Acknowledgments M. Holz acknowledge the Brazilian National Research Agency (CNPq) for research support (grant 302666/04-4). P. Dariva dos Reis, J. Küchle and J. Casagrande acknowledge Agência Nacional do Petróleo (ANP) and CNPq for scholarships. The authors acknowlege the organizers of the meeting for the invitation to contribute to this publication and specially Prof. Dr. R. Iannuzzi for his great editorial effort. The Brazilian Geological Survey - Companhia de Pesquisas de Recursos Minerais, Porto Alegre office, is thanked for providing access to sample material and well cores.

References Carter, R.W.G. 1988. Coastal environments. Academic Press, London. 617p.

Crowley, T.J.; Hyde,W.; Short, D. 1989. Seasonal cycle variations on the supercontienent of Pangea. Geology 17:457-460.

Cloething, S. 1988. Intraplate stress: a tectonic cause for third-order cycles in apparent sea level? In: Wilgus, C.K.; Hastings, B.S.; Kendall, C.G. St. C.; Posamentier, H.W.; Ross, C.A.; Van Wagoner, J.C. (Eds.) 88. Sea- level changes: an integrated approach. Society of Economic Paleonologists and Mineralogists Special Publication, 42:19-29.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 71

De Wit, M.J. & Ransome, I.G.D. 1992. Regional inversion tectonics along the southern margin of Gondwana. In: DE WIT, M.J. & RANSONE, I.G.D. Inversion Tectonics of the Cape Fold Belt, in Karoo and Cretaceous Basins of Southern Africa. Balkema, Rotterdam.

Helland-Hansen, W. & Martinsen, O.J. 1996. Shoreline trajectories and sequences: description of variable depositional-dip scenarios. Journal of Sedimentary Research 66(4):670-688.

Holz, M. 2003. Sequence stratigraphy of a lagoonal estuarine system - an example from the lower Permian Rio Bonito Formation, Paraná Basin, Brazil. Sedimentary Geology, 162:301-327.

Holz, M. & Kalkreuth, W. 2002. The Early Permian paleo-shoreline of the Paraná Basin in southernmost Brazil and northern Uruguay: paleoceanographic approach to coal and coastal sandstone facies distribution. Gondwana 11 contributions (see proceedings of this meeting).

Holz, M., Vieira, P. E., Kalkreuth, W. 2000. The Early Permian coal-bearing succession of the Paraná Basin in southernmost Brazil: depositional model and sequence stratigraphy. Revista Brasileira de Geociências, Rio de Janeiro, v.30, n.3, p.420-422.

Milani, E.. J. 2000. Geodinâmica Fanerozóica do Gondwana sul-ocidental e a Evolução Geológica da Bacia do Paraná. In: Geologia do Rio Grande do Sul, Holz, M & De Ros, L.F. eds., CIGO-UFRGS, Porto Alegre, p.275-302.

Milani, E. J. & Ramos, V. 1998. Orogenias Paleozóicas no domínio sul-ocidental do Gondwana e os ciclos de subsidência da Bacia do Paraná. Revista Brasileira de Geociências, 28(4): 527-544.

Patzkowsky,M.E.; Smith,L.H.; Markwick,P.J. Engberts, C.J.; Gyllenhaal, E.D. 1991. Application of the Fujita- Ziegler paleoclimate model: Ealy Permian and Late Cretaceous examples. Palaeogeography, Palaeoclimatology, Palaeoecology, 86(1991):67-85.

Schneider, R.L.; Mühlmann, H.; Tommasi, E.; Medeiros, R.A.; Daemon, R.F. & Nogueira, A.A. 1974. Revisão estratigráfica da Bacia do Paraná. XXVIII Congresso Brasileiro de Geologia, Porto Alegre. Anais, v.1, p. 41- 65.

Visser, J.N.J. 1983. The problem of recognizing anchient subaqueous debris flow deposits in glacial sequences. Transactions of the Geological Society of South Africa. 86:127-135.

Wehr, F.L. 1993. Effects of variation in subsidence and sediment supply on paraseqeunce stacking patterns. In: WEIMER. P. & POSAMENTIER, H. 1993. Siliciclastic seqeunce stratigraphy. AAPG Memoir 58. p.369-379.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 72 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Biostratigraphic versus Geochronologic frameworks in the Early Permian from Paraná Basin: looking forward a possible consensus.

Roberto Iannuzzi Universidade Federal do Rio Grande do Sul, Porto Alegre – RS – Brazil, [email protected]

Abstract Recent radiometric zircon ages reported from the Rio Bonito Formation, in southern Paraná Basin, show significant discrepancies when compared one another. The data have been obtained from tonstein layers interbedded with coal seams that belong approximate to equivalent stratigraphic interval or sequence within this unit. The absolute ages estimated varies from the Carboniferous/Permian boundary to the early Middle Permian (Rodian), according to one of the well-known geological time-scales used. Consequently, different time intervals can be attributed to the same biozone or fossil assemblage depending of the absolute age chosen. Then, each age support a specific and distinct biostratigraphic correlation with the fossiliferous deposits of neighboring basins. Taking into account these controversial ages, a biostratigraphic point of view is emphasized herein and compared with the geochronologic framework published for the basin until this moment. Finally, the absolute ages that corroborate the biostratigraphic framework are compared with the geochronologic framework known from southern Africa.

Introduction The main biostratigraphic problem of the most of Gondwanan Carboniferous-Permian deposits, including the Paraná Basin ones, is the absence of significant chrono-correlating marine faunal elements (such as foraminifers or amonoids) preventing correlations with the Late Carboniferous and Early Permian international stages. On the other hand, radiometric data are still scarce and conflictive in the Paleozoic sequences of Gondwana, militating against accurate age calibration of the available biostratigraphic schemes found throughout the super-continent. Thus, correlation with international stratigraphic stages can be considered as difficult and speculative in the most part of the cases. For all these reasons, a stronger geochronologic framework is needed throughout Gondwana in order to enable to get more accurate calibration of regional biostratigraphic schemes with the global time-scale. However, the calibration of fossil-bearing sections around world, including Gondwana ones, have demonstrate the coherence of the biostratigraphic correlations and relative datings previously obtained from the fossil contents (see Roberts et al., 1995, as a Gondwana example). Thus, it is expected that the radiometrical datings corroborate regional and intercontinental correlations established on the basis of fossils.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 73 Biostratigraphic framework of Paraná Basin The most useful biostratigraphic tool for the Upper Carboniferous - Middle Permian interval of the Paraná Basin is the palynologic content, due the abundance, diversity and widespread distribution of spore-pollen assemblages. Thus, four interval palynozones have been proposed in the last years for this interval, two corresponding to Pennsylvanian times and two attributed to Permian, which are namely in, ascending order, as Ahrensisporites cristatus (AcZ), Crucisaccites monoletus (CmZ), Vittatina costabilis (VcZ), and Lueckisporites virkkiae (LvZ) Zones (Souza and Marques-Toigo, 2003, 2005; Souza, 2006). The AcZ and CmZ occur in the basal and middle parts of the Itararé Subgroup, being considered as Late Carboniferosu in age (Moscovian-Gzhelian interval; Souza, 2006). VcZ has been recognized from the top of Itararé Subgroup to the uppermost Rio Bonito Formation/lowermost Palermo Formation. LvZ extends from the uppermost Rio Bonito Formation/lowermost Palermo Formation into Irati Formation, being found some elements of this zone in the basal part of Rio do Rastro Formation. Both VcZ and LvZ are considered as Permian in age, related respectively to the Asselian-Artinskian (VcZ) and the Artinskian-Wuachiapingian (LvZ) intervals (Souza et al., this volume). In contrast to palynozones, there is no applicable formal plant macrofossil based zonation for entire basin in the stratigraphic interval under consideration. Recently, Iannuzzi and Souza (2005) improved the classical scheme of Rösler (1978), proposing three successive informal floras, representing the Upper Carboniferous - Lower Permian strata of Paraná Basin, in ascending order: Pre-Glossopteris Flora (Pre-G), Phyllotheca- Gangamopteris (P-G) and Glossopteris–Brasilodendron (G-B). This scheme was adopted in order to enable better plant correlations with other Gondwana deposits. Pre-G extends from the basal and middle parts of the Itararé Subgroup, being considered as Late Carboniferosu in age (Iannuzzi & Souza, 2005). P-G has been recognized in the top of Itararé Subgroup (Iannuzzi et al., in press) and dated as earliest Permian (Asselian-Sakmarian) while GB is restricted to Rio Bonito Formation, being considered as Artinskian in age (Iannuzzi & Souza, 2005). Marine invertebrate assemblages (mollusks and brachiopods) are associated with the major transgressive phases in the interval under consideration, being restricted to specific horizons. As plant record, no formal zonation has been erected based on the invertebrate megafaunal succession. Concerning to biostratigraphic correlations, there are few important assemblages with Gondwanan affinities: Rio da Areia, Baitaca, Passinho and Taió. The former three show some elements associated with Australian “Eurydesma Fauna” and are reported from the top of Itararé Subgroup. The Taió Assemblage is related to Australian Permian faunas, being recorded in the middle part of Rio Bonito Formation (Rocha-Campos & Rösler, 1978). “Eurydesma Fauna” is considered throughout Gondwana as Early

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 74 Sakmarian in age (Rocha-Campos & Rösler, 1978; Iannuzzi & Souza, 2005). Taió Assemblage is dated as Artinskian. Vertebrate assemblages are very useful for biostratigraphic proposes but mostly restricted to the upper units of the basin, such as Rio do Rastro Formation. The basalmost vertebrate faunal record corresponds to the named “Mesosaurus Fauna” that occurs in the Irati Formation and is considered as Artinskian in age. Having in mind that it is the only vertebrate fauna important to the present contribution, the others are not been showed or discussed herein. The ages suggested to the palynozones and floras of Paraná Basin are based on correlation with palynological and plant assemblages, respectively, and on available radiometric data across Gondwana (see Iannuzzi & Souza, 2005; Santos et al., 2006; Césari, 2007). Marine invertebrate faunas have been dated based on correlation with Gondwana, mainly Australian, invertebrate assemblages (Rocha-Campos & Rösler, 1978; Iannuzzi & Souza, 2005). The age of “Mesosaurus Fauna” is on the basis of correlation with identical fauna reported to southern African basins (Iannuzzi et al., in press).

Radiometric Ages from Rio Bonito and Irati Formations Matos et al. (2001) obtained a date of 267.1 ± 3.4 Ma (U/Pb) from a tonstein interbedded in the upper coal seam of Candiota coalfield, situated in middle Rio Bonito Formation from southern Rio Grande do Sul, which corresponds to the Protohaploxypinus goraiensis Subzone of Vittatina costabilis Zone (VcZ) (according to Souza and Marques- Toigo, 2005). In the context of the time-scale of Gradstein et al. (2004), this age is assigned to the earliest Wordian stage. However, this age was recently re-dated by Guerra-Sommer et al. (2005) giving the new ages that varies from 299 + 2.6 to 296 + 1.4 Ma for the tonstein from Candiota. These ages correspond to Carboniferous/Permian boundary and Asselian stage according to above-mentioned time-scale. Recently, Guerra-Sommer et al. (2007) furnished a new radiometric zircon age of 285.4 + 8.6 Ma geochronologic data from a tonstein interbedded with coal seams of Faxinal coalfield, situated in middle Rio Bonito Formation, from Rio Grande do Sul, which corresponds to the Hamiapollenites karroensis Subzone of Vittatina costabilis Zone (VcZ). According to the time-scale of Gradstein et al. (2004), this age correspond to latest Sakmarian stage. On the other hand, a recent published SHRIMP zircon dating of 278.4 + 2.2 Ma (Santos et al., 2006) obtained from tuff beds in the overlying Irati Formation indicated a late Artinskian age for this unit, according to Gradstein et al.’s time-scale.

Stratigraphic position of coal seams The upper Itararé Subgroup/Rio Bonito Formation interval comprises three third-order

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 75 depositional sequences based on the stratigraphic scheme proposed by Holz (1999). The first sequence corresponds to upper Itararé Subgroup. The Sequence 2 extends from the base to the middle part of Rio Bonito Formation. The third sequence is recognized from the upper part of Rio Bonito Formation into to Palermo Formation. According to Holz (1999) and Holz et al. (2000), the main coal seams of Rio Bonito Formation in the Rio Grande Sul State lie below Sequence Boundary 3, extending through the Sequence 2. In this context, both coal seams of Candiota and Faxinal coalfields are situated in the middle portion of Rio Bonito Formation, as showed by Iannuzzi et al. (2007, in press).

Other Relevant Radiometric Ages from Gondwana Given the inconsistencies of age obtained from Rio Bonito Formation, the geochronologic framework adopted in the correlative sections from southern Africa can be useful in discussion on the biostratigraphic correlations. From the upper Dwyka Group in southern Namibia, in the Ganigobis Shale Member, Bangert et al. (1999) obtained SHRIMP U–Pb zircon ages of 302±3 Ma and 299.2±3.2 Ma from tuffaceous intervals. More specifically, they gave an age of 297±1.8 Ma for the top of deglaciation sequence III which is covered by the overlying “Eurydesma Fauna”. Besides, the bentonitic tuffs found in the lowermost Prince Albert Formation in South Africa has been dated as 288±3 Ma and 289.6±3.8 Ma also by Bangert et al. (1999). This interval overlies glacial deposits corresponding to the uppermost part of the Dwyka Group, which contains the “Eurydesma Fauna”. Because this, deposits associated with the “Eurydesma Fauna” have been considered as Early Sakmarian in age based on overlying tuffs dated by radiometric methods. This age has already been proposed previously on the basis of biostratigraphic correlations with Eastern Gondwana sequences. Other age control is given by U/Pb age of 270+1 Ma (= Kungurian/Roadian boundary) determinate from tuff beds of the Collingham Formation in the Main Karoo Basin of South Africa (Turner in Stollhofen et al., 2000). This unit overlies the Whitehill Formation, which is clearly correlated to the Irati Formation based on the “Mesosaurus Fauna”. This date corroborate the Artinskian age proposed by Santos et al. (2006) for Irati Formation.

Biostratigraphic correlations and radiometric ages The Analysis of the radiometric ages obtained from Rio Bonito Formation versus the biostratigraphic framework adopted in the Lower Permian interval of the Paraná Basin can be useful in the future discussion on the geochronologic mean of each dating. In order to adopt the age proposed by Matos et al. (2001; 267.1 ± 3.4 Ma), one should assume that the “Mesosaurus Fauna” is younger than Wordian and the Taió Assemblage is slightly older, corresponding approximately to Rhodian in age. Besides, a relatively long gap should be admitted between Rio Bonito and Itararé units, having in mind

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 76 the Sakmarian age of the “Eurydesma Fauna” registered in the top of Itararé Subgroup. In this context, the correlations of “Mesosaurus Fauna” and Taió Assemblage with African and Australian equivalent assemblages became totally impossible. Other controversy is related to the age obtained from Irati Formation. That one is older than age proposed by Matos et al. (2001). Thus, one of them is totally wrong once the Rio Bonito Formation is stratigraphically lower than the Irati Formation. Adopting the more recent dating proposed by Guerra-Sommer et al. (2005; 299 + 2.6 to 296 + 1.4 Ma) to the same horizon worked by Matos et al. (2001), one goes in the opposite direction. In this case, the “Eurydesma Fauna” should be considered as Late Carboniferous in age. This fact is totally against the well-dated sequences reported not only from African deposits but also from the other Gondwana basins. The presence of this fauna has been used as important stratigraphic marker throughout Gondwanan sequences, in which is associated with a widespread transgressive event coined as “Eurydesma transgression”. Therefore, the results showed by Guerra-Sommer et al. (2005) seem inconsistent with the independent ages proposed for “Eurydesma Fauna” in the other Gondwana areas. Besides, the widespread index palynomophs assigned throughout Gondwanan sequences to the earliest Early Permian (Asselian/Tastubian; Archbold et al., 2004; Césari, 2007) should be also considered as Carboniferous in age based on the data furnished by those authors to the middle part of Rio Bonito. Consequently, no palynological correlation is possible in order to establish the Carboniferous/Permian boundary in the Paraná Basin deposits. Finally, the last age analyzed herein is that one furnished recently by Guerra-Sommer et al. (2007; 285.4 + 8.6 Ma). This age is slightly better than the others. However, the same problems persist. For instance, the Taió Assemblage, situated in laterally equivalent horizons, should be considered as late Sakmarian in age in this proposal. However, the worse picture emerges from the correlative chart published in that paper (Figure 8), where the authors show the deposits of Itararé Subgroup quite inserted in the Carboniferous times. This interpretation implies again in a “Eurydesma Fauna” being recorded as much older in Paraná Basin when compared with equivalent assemblages dispersed through the other Gondwana sequences.

Conclusions Despite the fact of radiometric ages are welcome and desirable, the recent ages published to the Paraná Basin in southermost Brazil deserve more detailed analysis and discussion in order to check what could be done to improve those results. For this moment, the biostratigraphic framework is still more consistent and robust than geochronologic one and permits a more accurate inter-basinal or intercontinental correlation. We are looking forward to possible consensus in the near future.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 77 References Archbold, N.W., Cisterna, G.A. and Simanauskas, T. 2004. The Gondwana Carboniferous-Permian boundary revisited: new data from Australia and Argentina: Gondwana Research, v. 7, n. 1, p. 125-133.

Bangert, B., Stollhofen, H., Lorenz, V. and Armstrong, R., 1999. The geochronology and significance or ash-fall tuffs in the glaciogenic Carboniferous–Permian Dwyka Group of Namibia and South Africa: Journal of African Earth Sciences, v. 29, p. 33–49.

Césari, S.N., 2007. Palynological biozones and radiometric data at the Carboniferous-Permian boundary in western Gondwna: Gondwana Research, v. 11, n. 4, p. 529-536.

Gradstein, F.M. et al. (plus 38 authors), 2004, A geologic time scale 2004: Geological Survey of Canada, Miscellaneous Report 86, 1 chart.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Formozo, M. L., Menegat, R., and Basei, M.A.S., 2005, New radiometric data from ash fall rocks in Caniota coal-bearing strata and the palynostratigraphic framework in southern Paraná Basin (Brazil): Gondwana Symposium, 12, Mendoza, 2005. Abstracts, p. 189.

Guerra-Sommer, M., Cazzulo-Klepzig, M., Menegat, R., Formozo, M. L., Basei, M.A.S., Barboza, E.G., and Simas, M.W., 2007, Geochronological data from the Faxinal coal succession, southern Paraná Basin, Brazil: a preliminary approach combining radiometric U-Pb dating and palynostratigraphy. Journal of South America Earth Sciences, doi: 10.1016/j/jsames.2007.06.007 (in press).

Holz, M., 1999, Early Permian sequence stratigraphy and the palaeophysiographic evolution of the Paraná Basin in southernmost Brazil: Journal of African Earth Sciences, v. 29, no. 1, p. 51-61.

Holz, M., Vieira, P. E., and Kalkreuth, W., 2000, The Early Permian coal-bearing succession of the Paraná Basin in southernmost Brazil: depositional model and sequence stratigraphy: Revista Brasileira de Geociências, v. 30, no. 3, p. 420-422.

Iannuzzi, R., and Souza, P.A., 2005, Floral succeccion in the Lower Permian deposits of the Brazilian Paraná Basin: an up-to-date overview, in Lucas, S.G. and Zigler, K.E., eds., The Nonmarine Permian: New Mexico, New Mexico Museum of Natural History and Science Bulletin, no. 30, p. 144-149.

Iannuzzi, R., Souza, P.A., and Holz, M., 2007. Lower Permian post-glacial succession in the southernmost Brazilian Paraná Basin: stratigraphy and floral (macro and micro) record: in Extended Abstracts, European Meeting on the Paleontology and Stratigraphy of Latin America, 4a, Madrid: Madrid, Instituto Geológico y Minero de España, p. 207-212.

Iannuzzi, R., Souza, P.A., and Holz, M., in press. Stratigraphic and paleofloristic record of the Lower Permian postglacial succession in the Southern Brazilian Paraná Basin: Geological Society of America Special Papers.

Matos, S.L.F., Yamamoto, J.K., Riccomini, C., Hachiro, J. And Tassinari, C.C.G., 2001, Absolute dating of Permian ash- fall in the Rio Bonito Formation, Paraná Basin, Brazil: Gondwana Research, v. 4, p. 421-426.

Roberts, J., Claoue-Long, J., Jones, P.J., and Foster, C.B., 1995, SHRIMP zircon age control of Gondwanan sequences in Late Carboniferous and Early Permian Australia: in Dunay, R.E., and Hailwood, E.A., eds., Non- biostratigraphical Methods of Dating and Correlation: Geological Society London, Special Publication 89, p. 145- 175.

Rocha-Campos, A.C., and Rösler, O., 1978, Late Paleozoic faunal and floral successions in the Paraná Basin, southeastern Brazil: Boletim IG-USP, v. 9, p. 1-16.

Rösler, O., 1978, The Brazilian eogondwanic floral succession: Boletim IG-USP, v. 9, p. 85-90.

Santos, R.V., Souza, P.A., Alvarenga, C.J.S., Dantas, E.L., Pimentel, M.M., Oliveira, C.G. and Araújo, L.M., 2006, Shrimp U-Pb zircon dating and palynology of bentonitic layers from the Permian Irati Formation, Paraná Basin, Brazil: Gondwana Research, v. 9, p. 456-463.

Stollhofen, H., Stanistreet, I.G., Bangert, B., and Grill, H., 2000, Tuffs, tectonism and glacially related sea-level changes, Carboniferous-Permian, southern Namibia: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 161, p. 127- 150.

Souza, P.A., 2006, Late Carboniferous palynostratigraphy of the Itararé Subgroup, northeastern Paraná Basin, Brazil: Review of Palaeobotany and Palynology, v. 138, p. 9-29.

Souza, P.A., and Marques-Toigo, M., 2003, An overview on the palynostratigraphy of the Upper Paleozoic strata of the Brazilian Paraná Basin: Revista del Museo Argentino de Ciencias Naturales, nueva serie, v. 5, p. 205-214.

Souza, P.A., and Marques-Toigo, M., 2005, Progress on the palynostratigraphy of the Permian strata in Rio Grande do Sul State, Paraná Basin, Brazil: Anais da Academia Brasileira de Ciências, v. 77, p. 353-365.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 78 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The Late Paleozoic Gondwanide Orogen: the Sierra de la Ventana Transpressional Foldbelt

María Silvia Japas CONICET – Universidad de Buenos Aires, Buenos Aires – Argentina, [email protected]

Abstract During the Permian and after the south-western margin of Gondwana was built up by the accretion of different terranes, a stage of Andean type convergence began. As a consequence of this geodynamical scenario, a wide intraplate-pericratonic deformational belt developed, linking southern South America, South Africa, Antarctica and Australia (the Samfrau Orogenic Zone from du Toit, also known as the Gondwanide Orogen). As part of this Late Paleozoic orogenic belt, the Sierra de la Ventana Foldbelt shows a sector of pre-Permian strongly deformed rocks associated cratonwards with less deformed foreland basin deposits. Kinematic and strain fabric reveal transpressional deformation conditioned by oblique NNE convergence (present-day coordinates). The NNW-striking Sierra de la Ventana Foldbelt shows a sigmoidal design with an antitaxial bend to the north and a syntaxial arc southwards. Transected folds, thrusts, reverse and wrench faults seem to be the most striking structures affecting both Early and Late Paleozoic rocks. Minor shear zones, cleavage and strain also contribute to shorten the crust and allow to refine the kinematic and structural evolution of the chain.

Introduction The Sierras Australes de Buenos Aires (SABA) name the geological province which embraces two tectonic units: the Sierra de la Ventana Foldbelt (SVF) and the Sauce Grande Basin (SGB) (Andreis y Japas, 1991; López Gamundi and Rossello, 1998). Until the Triassic, as the south-american part of the du Toit´ Samfrau Geosyncline, the SABA shares a common geological history with the southafrican Cape Fold Belt - Karoo Foreland Basin counterpart (Keidel, 1916, du Toit, 1927, Cobbold et al., 1987, López Gamundi and Rossello, 1998, among others). Early Paleozoic metamorphosed conglomerates, quartzites and sandstones (Curamalal and Ventana Groups - Figure 1 -; Cape Supergroup) overlying the Precambrian-Cambrian basement were covered by a Late Paleozoic sequence (Pillahuincó Group - Figure 1 -; Karoo Supergroup).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 79 In Argentina, the Late Paleozoic deposits which comprise the Gondwanan glaciation (Sauce Grande Fm diamictites) and related postglaciation sequences (Piedra Azul and Bonete Fms) were referred to a sag phase of basin evolution by López Gamundi and Rossello (1998). The upper part of the Pillahuincó Gr, the Tunas Fm, displays a) paleoslope reversion (López Gamundi et al., 1995), b) abundant lithic-volcanic grains derived from a southwestern source area (Andreis and Cladera, 1992, López Gamundi et al., 1995), and c) indicators of syn-tectonic deposition (Japas, 1986, Cobbold et al., 1991, Tomezzoli and Vilas, 1999) and would represent a foreland basin sequence related to the Gondwanide shortening event. From the Tunas Fm exposures to the east, and buried by a relatively thin Late Mesozoic (?) and Cenozoic cover, the Pillahuincó Gr sedimentites fill the Claromecó Basin (Ramos, 1984). Late Paleozoic oblique convergence along the paleo-Pacific margin of Gondwana (Rapalini and Vilas, 1991, Japas and Kleiman, 2004, among others) built up a NE-vergence fold and thrust belt in the SABA, shortening the crust during four phases of contractional deformation (~278 Ma, ~ 262 Ma, 249 ± 8 Ma and ~ 230 Ma, López Gamundi et al., 1994).

Buenos Aires

Figure 1 – Geological map from the Sierras Australes de Buenos Aires.

Different structural models (Ramos, 1984, Sellés Martínez, 1989, Cobbold et al., 1991, von Gosen et al. 1990, among others) and related cross-sections (including Ploszkiewicz, 1999, Tomezzoli and Cristallini, 2004) have been proposed to explain the

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 80 deformation history of the SABA. Although strain data and minor structures are available all over the region, some of these structural profiles only consider geometrical reconstruction. Thus, the aim of this contribution will be to summarise the structural knowledge of the SVF- SGB paying more attention on how strain data and related ductile deformation could support previous interpretations or not.

Regional setting The Sierras Australes and their neighbouring areas are depicted in Figure 2. East of the Sierras Australes, the NW-SE trending Interserrana basin (Claromecó foredeep of Ramos, 1984) comprises a thick paleozoic depocentre of about 7 km of sedimentites. Small scattered exposures of nearly undeformed rocks were assigned to the Pillahuincó Gr. Further east, the Tandil High which is carved in Precambrian to Ordovician rocks. Southwards, lies the Cretaceous Colorado aulacogenic basin with its pre-rift sequences associated with the Sauce Grande basin deposits, and further to the southwest is the North-Patagonian Massif composed of Precambrian and Early Paleozoic rocks deformed and also intruded during the Early Permian. To the west, buried precambrian rocks with a probable Sierras Pampeanas filiation link the Sauce Grande basin to the Permian Carapacha Basin (Chadileuvú Block). Two Cretaceous basins (Macachín, Laboulaye or Levalle) border on the Sierras Austrtales NW margin probably interconnecting the Sierras Australes with the Chacoparanense basin.

Figure 2 – The Sierra de la Ventana Foldbelt in a regional context (modified from Andreis and Japas, 1991).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 81 Ductile deformation: Incidence for a kinematic model At regional and local scales strain reveals heterogenous deformation. Mylonitic (proto- to ultramylonites) shear zones affecting the basement are examples of this inhomogeneous behaviour of deformed rocks. The western area shows NNW folds of similar type with axial planes dipping southwest, associated thrusts and reverse faults (Figure 3). Previous sub-horizontal thrusts are folded. Eastwards folding become similar (to concentric) with SW tilted axial planes. Although fold axes show hinge culminations and depressions, a general tendency of plunging 10°-12° southeastwards was depicted (Harrington, 1947, among others). At a regional scale, folds show refolded inverted and normal limbs. At the easternmost region sedimentites belonging to the Sauce Grande Basin were folded following a parallel-type geometry and show variable amplitude and wavelength, with vertical to steeply-dipping axial planes (Figure 3). NNW- (northern Sierra de las Tunas) to NW- (Southern Sierra de Pillahuincó) trending fold axes plunge 5° - 10° to the SE. Basement rocks crop out as isolated exposures in the westernmost area. In the northern arc region, columnar jointing affecting rhyolitic rocks exhibit heterogeneous deformation with consequent variable strain ratio (so far measurable up to 0,28) as well as visible strain partitioning with top-to-the-NNE and NW dextral motions (Japas and Sellés Martínez, 1999). Shear zones recognized in the southern arc area indicate lower degree of strain partitioning (NW bands with thrust and dextral wrenching components) and high partitioning of motions (NNE dextral vertical and SSW moderately dipping sinistral shear zones) (Cobbold et al., 1991, Delpino and Dimieri, 1992). Strain ratio was estimated at 0.4 (Delpino and Dimieri, 1992). The Pan de Azúcar - Del Corral anticline shows La Lola Fm conglomerates resting tectonically on the basement rocks but both display the same structural grain (Cobbold et al., 1991, Delpino and Dimieri, 1992). Scattered exposures of La Lola Fm conglomerates, which only crop out at the western margin of the NW arc, bear well developed shear zones. Depending on their position in the arc these minor structures show different trends, motions and strain (Japas 1991): while WNW top-to-the-N motions and brittle deformation with low strain ratio (0.87 to 0.56) prevail at the northern end, combined NNE dextral and top-to-the NE movements and ductile strain with higher strain (0.87 to 0.35) dominate the southern segment. Stretching lineation in Del Corral area is oblique to both folds and faults (Cucchi, 1966) and a similar strain ratio to the one obtained from basement rocks was calculated by von Gosen et al. (1990). In the same domain, ultramylonitic shear zones affecting the Trocadero Fm metaquartzites reveal a folded low angle thrust (Ducós, 1994). In the central region of the arc, schists of the Hinojo Fm show strong evidences of thrusting towards the NE. Although major thrusts are not evident in those exposures belonging to the Ventana Group rocks, higher strain (with dynamic recristallization of quartz grains) localize on syncline

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 82 hinges (strain ratio up to 0.28) and inverted limbs belonging to the inverted limb of the major structure (Japas, 1988). Horizontal NE-verging shear zones were detected at different scales. Normal limbs are often stretched during top-to-the-NE shearing leading to the boudinage of beds. Extensional fractures developed perpendicular to folds. Despite the fact that jointing was commonly adscribed to reflect orthorrombic symmetry, Japas (1989) refers to it as monoclinic, further evidence of regional transpression. In the central segment of the arc, flattening of subhorizontal axial planes would reveal subsurficial decollément (Cobbold et al., 1991). Lolén Fm metasedimentites are also tightly folded. As in the quartzites case, anticline hinges and normal limbs (strain ratio: 0.5) are less strained than synclines and inverted ones (0.3), and show maximum elongation direction in coincidence with fold axes. While the former group hold a flattening strain fabric the latter displays a plane one, with a stretching lineation parallel to the dip direction of cleavage, a usual situation observed at the western SVF region. Folded s/c structures with local vergence reversion are common structures at Lolén Fm base and are considered a key structure in SVF kinematic interpretation. Late contractional kink bands developed as a consequence of progressive deformation. Strongly cleaved and highly strained Sauce Grande Fm rocks display flattening strain fabric and gentle folds indicating a main initial stage of layer parallel shortening of the diamictites. Localized shear zones related to the westernmost anticline and syncline hinges induce axial interchange and the development of plane strain fabric, pointing to the presence of subsurficial blind thrusts. Vertical stretched crystal fibres grown on nearly horizontal fractures in clasts and unvariable SW-dipping cleavage, discard the presence of emergent thrusts in the region. Geometrical and strain markers would support a zone of contact strain in association with the underlying competent Lolén Fm rocks (Japas, 1987). Minor structures in clasts (i.e. bookshelf sliding, enveloping cleavage) confirm a dextral component of simple shear associated with regional NNE convergence. Maximum strain ratio was estimated in 0.2 whereas 0.32 could be considered an average value for the more strained areas. As in the Lolén Fm metasedimentites, the strongly cleaved diamictites are kinked. Based on measurements on distorted fossils and concretions, sedimentites of Piedra Azul, Bonete and Tunas Fms indicate relatively low strain (20-30 % of relative shortening) and a flattening strain fabric. Parallel folds with more strained limbs than hinges and well developed pencil cleavage are characteristic. Both, transected folds (Japas, 1986, 1988, 1989, Cobbold et al., 1991) and a direction of maximum elongation nearly parallel to folds axes evidence a dextral NNW component of non-coaxial deformation. In accordance with the evidence supported from regional fold trend rotation, higher strain was observed in the Sierra de las Tunas than in the Sierra de Pillahuincó area.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 83 Relationships between folding lineation, fault trend, strain ratio and strain fabric enable the definition of a transpressive regime in the Sierras Australes during the Gondwanide Orogeny.

Figure 3 – Main structural features and finite strain at the SVF-SGB.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 84 References Andreis, R.R. and Cladera, G., 1992. Las epiclastitas pérmicas de la cuenca Sauce Grande (Sierras Australes, Buenos Aires, Argentina). Parte I: Composición y procedencia de los detritos. Actas 4° Reunión Argentina de Sedimentología, 1 : 127-134. La Plata.

Andreis, R.R. and Japas, M.S., 1991. Cuenca de Sauce Grande y Colorado. In: Archangelsky, S. (ed) El Sistema Pérmico en la República Argentina y en la República Oriental del Uruguay. 12° International Congress on Carboniferous and Permian Stratigraphy and Geology. Academia Nacional de Ciencias, 1 : 45-64. Córdoba.

Cobbold, P.R., Massabie, A.C. and Rossello, E.A., 1987. Hercynian wrenching and thrusting in the Sierras Australes Foldbelt, Argentina. Hercynica, 2 (2) : 135-148.

Cobbold P.R., Gapais, D. and Rossello, E.A., 1991. Partitioning of transpressive motions within a sigmoidal foldbelt: The variscan Sierras Australes, Argentina. Journal of Structural Geology, 13 (7) : 743-758.

Cucchi, R., 1966. Petrofábrica del conglomerado de la Formación La Lola, Sierras Australes de la provincia de Buenos Aire. Revista de la Asociación Geológica Argentina, 21 (2) : 71-106. Buenos Aires.

Delpino, S.H. and Dimieri, L.V., 1992. Análisis de la deformación sobre las rocas del basamento aflorantes en el faldeo occidental del Cerro del Corral, Sierras Australes Bonaerenses. Acta 7° Reunión sobre Microtectónica : 53-61. Bahía Blanca. du Toit, A., 1927. A geological comparison of South America with South Africa. With a paleontological contribution by F. Cowper Reed. Carnegie Institut of Washington, Publication 381, 158 p. Washington.

Ducós, E.I., 1994. Análisis meso y microestructural en el área del Cerro de los Terneros. Sierras Australes de Buenos Aires. Trabajo Final de Licenciatura, Universidad de Buenos Aires, 100 p., unpublished.

Harrington, H.J., 1947. Explicación de las Hojas Geológicas 33m (Sierra de Curamalal) y 34m (Sierra de la Ventana). Provincia de Buenos Aires. Boletín Dirección de Geología y Minería, 61, 43 p. Buenos Aires.

Japas, M.S., 1986. Caracterización geométrico-estructural del Grupo Pillahuincó. I. Perfil del Arroyo Atravesado, Sierra de las Tunas, Sierras Australes de Buenos Aires. Anales Academia Nacional de Ciencias Exactas, Físicas y Naturales, 38 : 145-154. Buenos Aires.

Japas, M.S., 1987. Caracterización geométrico-estructural del Grupo Pillahuincó. II. Formación Sauce Grande. Perfil del Cordón Mambacher y Sierra de las Tunas occidental, Sierras Australes de Buenos Aires. Anales Academia Nacional de Ciencias Exactas, Físicas y Naturales, 39 : 125-144. Buenos Aires.

Japas, M.S., 1988. Análisis cuantitativo de la deformación en el sector oriental de las Sierras Australes de Buenos Aires y su implicancia geodinámica. PhD (Universidad de Buenos Aires), 359 p., unpublished.

Japas, M.S., 1989. Análisis de la deformación en las Sierras Australes de Buenos Aires. Anales Academia Nacional de Ciencias Exactas, Físicas y Naturales, 41 :193-215. Buenos Aires.

Japas, M.S. 1991. Análisis microtectónico de la fábrica deformada del conglomerado de la Formación La Lola. Sierras Australes de Buenos Aires. VII Reunión sobre Microtectónica, Acta : 85-91. Bahía Blanca.

Japas, M.S. and Sellés Martínez, J., 1999. Análisis de la microfábrica deformacional de los "pórfidos riolíticos" del basamento en el área de Pigüé. Revista de la Asociación Geológica Argentina, 53 (3) : 317-324.

Japas, M.S. and Kleiman, L.E., 2004. El Ciclo Choiyoi en el Bloque de San Rafael: de la orogénesis tardía a la relajación mecánica. Asociación Geológica Argentina, Serie D: Publicación Especial N° 7: 89-100.

Keidel, J., 1916. La geología de las sierras de la Provincia de Buenos Aires y sus relaciones con las montañas del Cabo y los Andes. Ministerio de Agricultura de la Nación. Anales Dirección General de Geología y Minería, 9 (3) : 5-77. Buenos Aires.

López Gamundi,O.R and Rossello, E.A., 1998. Basin fill evolution and paleotectonic pattern along the Samfrau geosyncline. Geologische Rundschau, 86 : 818-834.

López Gamundi, O.R., Conaghan, P.J., Rossello, E.A. and Cobbold, P.R., 1995. The Tunas Formation in the Sierras Australes Fold Belt. Journal of South American Earth Sciences, 8 : 129-142.

López-Gamundí, O.R., Espejo, I.S., Conaghan, P.J. and Powell, C.M.A. 1994. Southern South America. In: Veevers, J.J. & Powell, C.M.A. (eds) Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland. Geological Society of America, Boulder, Memoir, 184, 281-329.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 85 Ploszkiewicz, J.V., 1999. ¿ Será Buenos Aires una nueva provincia petrolera ? Antecedentes, hipótesis y certezas. Boletín de Informaciones Petroleras, 58 : 45-56.

Ramos, V.A., 1984. Patagonia: Un continente paleozoico a la deriva ? Actas 9° Congreso Geológico Argentino, 2: 311-325. Bariloche.

Rapalini, A.E. and Vilas, J.F., 1991. Tectonic rotations in the late Paleozoic continental margin of southern South America determined and dated by paleomagnetism. Geophysical Journal International, 107: 333-351.

Sellés Martínez, J., 1989. The structure of Sierras Australes (Buenos Aires – Argentina). An example of folding in a transpressive envirinment. Journal of South American Earth Sciences, 2 (4) : 317-329.

Tomezzoli, R.N. and Cristallini, E.O., 2004. Secciones estructurales de las Sierras Australes de la provincia de Buenos Aires: Repetición de la secuencia estratigráfica a partir de fallas inversas ? Revista de la Asociación Geológica Argentina, 59 (2) : 330-340.

Tomezzoli, R.N., and Vilas, J. F., 1999. Palaeomagnetic constraints on age of deformation of the Sierras Australes thrust and fold belt, Argentina. Geophysical Journal International, 138: 857-870. von Gosen, W., Buggisch, W. and Dimieri, L.V., 1990. Structural and metamorphic evolution of the Sierras Australes (Buenos Aires province/Argentina). Geologische Rundschau, 79 (3) : 797-821.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 86 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Paleoenvironmental evolution and biostratigraphy of Late Paleozoic Andean Gondwana basins

Carlos Oscar Limarino¹, Silvia Nélida Césari² ¹ Departamento de Geologia, Universidad de Buenos Aires, Buenos Aires-Argentina, [email protected] ² Museo Argentino de Ciencias Naturales “B. Rivadavia”, Buenos Aires-Argentina, [email protected]

Abstract Late Paleozoic Andean basins show a complete record of the paleoenvironmental and biostratigraphic evolution of the western margin of Gondwana. In this area two major types of basins are recognized: (1) the western arc- related basins and (2) the eastern retroarc basins. The stratigraphic record of these depositional areas is here divided into five megasequences bounded by regional-scale unconformities. Megasequences span the interval from the Mississippian through earliest Triassic and they are constrained in age on the basis of palynological biozones and radiometric ages.

Introduction The Late Paleozoic Gondwana Andean basins provide an excellent record of the stratigraphic, paleoclimatic and tectonic evolution of western margin of Gondwana. Evidence of a complex paleoclimatic history from icehouse to greenhouse conditions have been registered in these basins (Limarino and Spalletti, 1985; López Gamundí, 1987; López Gamundí et al., 1992) as well as a complete biostratigraphic record (see Césari and Gutierrez 2001 for a revision) and a complex tectonic and magmatic activity linked to the formation and break-up of Gondwana (Ramos et al., 1984; Ramos, 2001). From a regional point of view, three major types of basins are recognized during the Late Paleozoic in southern South America (Limarino and Spalletti, 2006): (1) arc-related basins, (2) retroarc basins and (3) intraplate basins (Figure 1). The here analyzed Andean basins include arc-related and retroarc basins, which encompass the major part of Chile, western Argentina and southwest Bolivia. Andean basins form the western active margin of

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 87 Gondwana, which was subject to persistent tectonic and magmatic activity from the Late Devonian to beyond the end of the late Paleozoic.

Figure 1 - Paleogeography of Late Paleozoic basins and location of the studied area.

Stratigraphic tools for regional correlation Among the tools available for establishing regional scale correlations we have used palynological biozones, radiometric ages and different stratigraphic scales of bounding surfaces that delineate discontinuities in the stratigraphic record. Palynological biozones were preferred because they have been described in several Andean basins (Césari and Gutierrez, 2001). Palynomorphs occur in continental, transitional and marine environments improving the spectrum of depositional settings were biostratigraphic information can be obtained. Figure 2 shows the palynological scheme used in this contribution which is basically that presented by Césari and Gutierrez (2001).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 88 Radiometric data are relatively scarce in the Andean basins, but some interesting ages have been obtained in the San Rafael, Paganzo and Calingasta-Uspallata Basins (Figure 1). In the first area, at least 20 radiometric data belonging to Permian rocks have been obtained while in Paganzo and Calingasta-Uspallata basins some radiometric ages were supplied mainly by Carboniferous rocks. Finally, different hierarchical orders of stratigraphic bounding surfaces were used for regional-scale correlations. Following the model presented by Limarino et al. (2006) the first- order surfaces occur as regional unconformities recognized throughout the region. Second- order surfaces correspond to basinal scale unconformities or significant erosive surfaces mainly resulting from sea level falls. Finally, sedimentary truncations in sedimentary facies define third-order stratigraphic bounding surfaces, which primarily record minor sea level oscillations and/or were controlled by changes in paleoclimatic conditions.

Palynology Paleobotany

. Lueckisporites-

m r

e Weylandites P

Biozone

. m

r Pakhapites fusus- e Gangamopteris

P Vittatina subsaccata

l Biozone Biozone

a E

Subzone C Interval Biozone n

a i

e a

n Subzone B

v l

o y

s NBG Biozone

i n

B n

e Subzone A

p i

s Biozone CV AF Biozone

s i

Figure 2 - Biostratigraphic schemes for Late Paleozoic basins (modified from Césari end Gutierrez, 2001).

Paleogeographic evolution of the Andean Basins Five megasequences bounded by regional-scale unconformities have been identified in Late Paleozoic South American basins (Figure 3). Megasequence 1, Mississippian in age, characterizes the syntectonic sedimentation that took place along the western margin of Gondwana after the collision and amalgamation of the allochthonous terrain of Chilenia (Ramos et al., 1984). The base of this megasequence is bounded by the so-called Chanic

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 89 unconformity that separates Mississipian (and also probably latest Devonian) rocks above from Middle Devonian and older units below. The rocks included in megasequence 2 overlie an important unconformity created by the Río Blanco tectonic movements. This interval is characterized by dominantly postectonic sedimentation; evidence of magmatic activity is rare. Megasequence 3 was deposited during the earliest Permian (Early Cisuralian) after the deformational events associated with the Atacama movements. Interestingly, these movements produced a marked unconformity in the arc-related basins, thought in the retroarc area to pass into a correlative conformity and frequently is only detected by changes in the stacking of facies patterns. During the Late Cisuralian, megasequence 4 was deposited after San Rafael tectonic movements ceased. This megasequence marks an abrupt contrast of facies between the arc-related basins and the retroarc ones. The former were dominated by extensive volcanism and sporadic volcaniclastic sedimentation during intereruptive periods, passing into eolian, lacustrine (mainly ephemeral lake deposits) and low-energy fluvial sediments in retroarc positions. Finally, megasequence 5 (Guadalupian – earliest Triassic?) corresponds to the climax of volcanic activity in the arc-related basins. During this time, the Amaná movements produced dramatic changes in the paleogeography of the basins located in the retroarc area.

Figure 3 - Megasequences, tectonic phases and regional sedimentary patterns in arc-related and retroarc basins. From biozone references see the text.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 90 Megasequence 1 (Mississippian synorogenic sedimentation) Mississippian sedimentation is well represented in the Calingasta-Uspallata, Río Blanco and Tepuel Genoa basins (Figure 1). In the Río Blanco area Early Carboniferous rocks are included in the Angualasto Group, a thick (up to 2300 m) siliciclatic sequence dominated by shallow marine and continental sedimentation. Megasequence 1 is divided into two depositional sequences; the lower one corresponds to the Malimán Formation (and equivalents) beginning with shallow marine and local continental deposits abruptly covered by fine grained sediments belonging to the Tournaisian–Early Visean transgression. The upper sequence consists of the Cortaderas Formation which can be divided into three sections: (1) a basal conglomeratic sequence deposited in fan deltas and alluvial fans, (2) a middle section dominated by sandstones sedimented in alternating shallow marine and fluvial environments and (3) an upper glacial section. Probably, one of the more remarkable aspects of megasequence 1 is the existence of Early Carboniferous glacial diamictites that represent the oldest glacial deposits recognized in the western basins of Argentina and Chile. Megasequence I is interpreted as having been deposited under synorogenic conditions during the final phases of the Protoprecordillera uplift. Palynological associations have been described in the Rio Blanco Basin, where the Angualasto Group provided rich assemblages that contain characteristic species of Mississippian age (Césari and Limarino, 1992, 1995; Amenábar et al., 2006; Perez Loinaze 2005). The composition of these spore associations, preliminary included in the Cordylosporites-Verrucosisporites (CV) biozone by Césari and Gutiérrez (2001), are being revised, but the proposed Tournaisian-Viséan age seems to be confirmed. Macrofloristic remains correspond to the AF monotonous flora characterized by abundant small lycophytes, dominated by Frenguellia, and delicate pteridosperms.

Megasequence 2 (Pennsylvanian postorogenic sedimentation) Along to the western basins, a well exposed unconformity (Río Blanco movements) marks the contact between highly deformed rocks of the megasequence 1 and the less disturbed rocks of megasequence II. This last megasequence consists of the lower section of the Paganzo Group in the retroarc area and the Agua de Jagüel Formation (and equivalents) along the arc-related basins. Despite differences in their paleogeographic positions, a similar pattern of sedimentation can be established throughout the western basins. The lowermost deposits correspond to glacial diamictites belonging to the so-called Namurian glacial event, which is recorded by true tillites and resedimented glacial-related diamictites (López Gamundí, 1987; Pazos, 2002; Marenssi et al., 2005). These coarse grained rocks are succeeded by shales with dropstones overlain in turn by progradational sequences of sandstones and mudstones belonging to the Namurian postglacial transgression (Limarino et al, 2002). This marine interval occurs in all of the western basins from Tarija Basin in Bolivia

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 91 to the Tepuel-Genoa Basin in Patagonia. Toward the end of deposition of megasequence 2, renewed marine transgression entirely flooded the arc-related basins (Agua Negra Formation and equivalents) and the westernmost sector of the retroarc area. The biozone Raistrickia densa-Convolutispora muriornata (DM) that characterizes these sequences is distinguished by the presence of abundant monosaccate pollen and spore species not recognized in the underlying assemblages. A distinctive floristic change is assumed at the Mississippian/Pennsylvanian boundary when CV and DM biozones are compared. The associated Late Carboniferous NBG (Nothorhacopteris-Botrychiopsis- Ginkgophyllum) flora is characterized by the incoming of cordaitales, sub-arborescent lycophytes and pteridosperms with robust fronds. The macrofloristic appearance of conifers and ferns characterizes the Interval biozone at the uppermost levels.

Megasequence 3 (interfingering of shallow marine and continental deposits). In the arc-related basins, an important unconformity (Atacama movements) separates rocks of megasequences 2 and 3, however, easterward, the unconformity surface is missing and the effect of the Atacama movements must be deduced from changes in sedimentation patterns or local unconformities. In the arc-related basins sedimentation was dominated by high energy fluvial systems covered by shallow marine deposits including several levels of limestones. In addition, this interval contains volcanic rocks interspersed throughout the sedimentary sequence (San Ignacio Fm. in Argentina and Quipisca Fm. in Chile). In the western retroarc basins, shallow marine deposits (up to 100 m. thick) form the base of megasequence 3, passing upward into low-energy fluvial sequences. Marine transgressions did not reach the eastern retroarc basins where sedimentation was almost entirely fluvial. The incoming of abundant taeniate pollen grains like Vittatina and Pakhapites and the diagnostic spore species Converrucosisporites confluens characterizes the FS (Pakhapites fusus-Vittatina subsaccata) biozone, associated with the first records of glossopterid leaves. A maximum age of 298-301Ma for these paleofloristic remains was suggested by Césari (2007).

Megasequence 4 (onset of Permo-Triassic volcanism and formation of eolian sand seas) Thick intervals of volcanic flows, breccias, ignimbrites and different types of pyroclastic flows were deposited throughout the arc-related basins over the San Rafael unconformity. This volcanism forms the lower section of the Choiyoi Group in Argentina (Pastos Blancos Group in Chile) and it likely records the re-establishment of subduction along the western margin of Gondwana.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 92 Development of a volcanic chain in the west coupled with arid or semiarid climatic conditions favored the formation of extensive eolian sequences closely associated with playa-lake and ephemeral fluvial deposits. Palynological assemblages of this interval are distinguished by the predominance of bisaccate taeniate pollen and scarce spores. The Argentinian Lueckisporites/Weylandites biozone is associated with a radiometric age of 266.3±0.8 Ma in the San Rafael Basin.

Megasequence 5 (acme of the Permo-Triassic volcanism) Megasequence 5 records the acme of volcanic activity in the arc-related basins in which more than 3000 m. of volcanic rocks accumulated. In the retroarc basins, the Amanaica movement produced substantial paleogeographic modifications as shown not only by the formation of new depocentres but also important changes in the composition of detrital modes of sandstones. While new depocentres are believed to have formed to the east of the Valle Fertil lineament, rocks of megasequence 5 are missing in the western portion of the retroarc basin. This could suggest that the volcanic arc and at least part of the western retroarc area were uplifted during the latest Permian or earliest Triassic producing the inversion of the regional gradient of the basins. This is partially supported by changes in the detrital mode of sandstones which exhibit abundant volcanic fragments that werevery probably derived from the western volcanic arc.

References Amenábar, C., di Pasquo, M., Carrizo, H. and Azcuy, C. 2006. Palynology of the Chigua and Maliman Formations in the Sierra del Volcán, San Juan Province, Argentina. Part I. Palaeomicroplankton and acavate smooth and ornamented spores. Ameghiniana v. 43, p. 339-375.

Césari, S. N., 2007. Palynological biozones and radiometric data at the Carboniferous-Permian boundary in Western Gondwana. Gondwana Research v. 11, p. 529-536

Césari, S.N., Gutiérrez, P.R., 2001. Palynostratigraphy of Upper Paleozoic sequences in Central-Western Argentina. Palynology v. 24, p. 113–146.

Césari, S. and Limarino, C. 1992. Palinomorfos eocarboníferos en la Formación Cortaderas, provincia de San Juan, Argentina. 8° Simposio Argentino Paleobotánica y Palinología (Corrientes 1991), Actas, p. 45-48.

Césari, S. and Limarino, C. 1995. Primer registro palinológico de la Formación Malimán (Carbonífero inferior), Cuenca Río Blanco, Argentina. 6° Congreso Argentino de Paleontología y Bioestratigrafía (Trelew, 2004), Actas, p. 77-83.

Limarino, C.O., Spalletti, L.A., 1985. Eolian Permian deposits in West and Northwest Argentina. Sedimentary Geology v. 49, p. 129–137.

Limarino, C.O. and Spalletti, L.A., 2006. Paleogeography of the upper Paleozoic basins of southern South America: An overview. Journal of South American Earth Sciences, 22: 134-155.

Limarino, C.O., Césari, S., Net, L., Marenssi, S., Gutiérrez, R., Tripaldi, A., 2002. The Upper Carboniferous postglacial transgression in the Paganzo and Río Blanco Basins (Northwestern Argentina): facies and stratigraphic significance. Journal of South American Earth Sciences v. 15, p. 445–460.

Limarino, C.O.; Tripaldi, A.; Marenssi, S. and Fauqué, L., 2006. Tectonic, sea-level, and climatic controls on Late Paleozoic sedimentation in the western basins of Argentina. Journal of South American Earth Sciences v. 22, p. 205-226.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 93 López Gamundí, O.R., 1987. Depositional models for the glaciomarine sequences of Andean Late Paleozoic basins of Argentina. Sedimentary Geology v. 52, p. 109–126.

López Gamundí, O.R. and Breitkreuz, 1997. Carboniferous to Triassic evolution of the Panthalassan margin in southern South America. In: Dickins, J.M. Zuniy, Hongfu, Y., Lucas S.G. and Acharyya, S. (Eds.), Late Paleozoic and Early Mesozoic Circum-Pacific events and their global correlation, World and Regional Series, Cambridge, 10 University Press, p. 8–19.

López Gamundí, O.R., Limarino, C., Césari, S., 1992. Late Paleozoic paleoclimatology of central west Argentina. Palaeogeography, Palaeoclimatology and Palaeoecology v. 91, p. 305–329.

Marenssi, S.A., Tripaldi, A., Limarino, C.O., Caselli, A.T., 2005. Facies and architecture of a Carboniferous grounding-line system from the Guandacol Formation, Paganzo Basin, northwestern Argentina. Gondwana Research v. 8, p. 187–202.

Pazos, J., 2002. The Late Carboniferous Glacial to Postglacial Transition: Facies and Sequence Stratigraphy, Western Paganzo Basin, Argentina. Gondwana Research 5(2) 467-487.

Perez Loinaze, V. 2005. Some trilete spores from Lower Carboniferous strata of the Río Blanco Basin, western Argentina Ameghiniana v. 42, p. 481-488.

Ramos, V.A., 2001, The Southern Central Andes. In: Cordani, U., Milani, E.J., Thomaz Filho, A.,Campos, D.A. (Eds.), Tectonic Evolution of South America, p. 561–604

Ramos, V.A., Jordan, T.E., Allmendinger, R.W., Kay, S.M., Cortés, J.M. and Palma, M.A., 1984. Chilenia: un terreno alóctono en la evolución paleozoica de los Andes Centrales. 9º Congreso Geológico Argentino, Actas 2, p. 84–106.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 94 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Late Triassic-Early Jurassic western Gondwana tetrapod correlations

Claudia Marsicano Dto. de Ciencias Geológicas, FCEyN, Universidad de Buenos Aires – Argentina, [email protected]

Abstract Late Triassic-Early Jurassic western Gondwana intercontinental correlations based on tetrapods are analyzed. Particularly, former evidence that supported the placement of the TJ boundary in the Karoo Basin is demostrated to be weakly constrained when the putative Lower Jurassic tetrapod faunas of the Upper Elliot are compared with those from the Upper Triassic Los Colorados and Laguna Colorada formations of Argentina. These comparisons are based on recent phylogenetic analysis on basal dinosaurs that support closer relationships between the South African and South American taxa than to other early Mesozoic dinosaurs from Pangea. The position of the TJ boundary in the Karoo Basin, one of the major continental sequences of Western Gondwana, is of particular interest as it would have important implications on our understanding of the patterns of diversification of the Early Mesozoic tetrapod faunas in this part of Gondwana and, consequently, on the alleged worldwide tetrapod continental extinction at the TJ boundary.

Introduction Late Triassic-Early Jurassic tetrapod biostratigraphy has been the subject of several discussions, primary based on the intercontinental correlations from the European marine and continental type sections to other fully continental successions from both Laurasia and Gondwana and the placement of the Triassic-Jurassic (TJ) boundary (e.g. Olsen and Galton, 1977, 1984; Olsen and Sues, 1986; Olsen et al., 1987; Benton, 1994; Lucas, 1998; Olsen et al., 2002a, 2002b; Langer et al., 2007). Olsen and Galton in an influential contribution (Olsen and Galton, 1977) correlated the European horizons with the thick continental succession of eastern USA and Canada (Newark Supergroup) and by extension relocated the TJ boundary in several sequences from elsewhere on the basis of their tetrapod content, both body-fossils and footprints (Olsen and Galton, 1977, 1984; Olsen and Sues, 1986).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 95 Of particular interest for the biostratigraphy of Western Gondwana was the reassignment of the upper part of the classic “Stormberg Group” (Karoo Basin) of South Africa to the early Jurassic, well known for its rich tetrapod fauna (e.g. Kitching and Raath, 1984; Olsen and Galton, 1984; Lucas and Hancox, 2001; Knoll, 2004, Rubidge, 2005). Besides, the diverse tetrapod fauna from Los Colorados Formation in central-western Argentina (Bermejo Basin) was also discussed and considered to represent a unique transitional Triassic-Jurassic assemblage therefore of uncertain latest Triassic age (Olsen and Galton, 1977, 1984; Olsen and Sues, 1986; Benton, 1994; Lucas, 1998).

Late Triassic-Early Jurassic successions from Argentina and South Africa The early Mesozoic infilling of the South African Karoo Basin is recorded by the rocks included in the Molteno, Elliot and Clarens formations, which together with the overlying Drakensberg volcanics comprise the informally recognised “Stormberg Group” (Late Triassic- Early Jurassic). The basalts, with radiometric dates between 193-179 Ma (Duncan et al., 1997), indicate the beginning of the break-up of Gondwana and constrained the upper age of the “Stormberg” sediments to the lowermost Jurassic (Hettangian sunsu Gradstein et al., 2005). However, the lack of radiometrically datable material within the Mesozoic Karoo beds has depended on tetrapod biostratigraphic age control. In particular, the stratigraphic position of the TJ boundary in the Karoo succession has always been poorly constrained. Thus, primary based on the equivalent footprint assemblages shared by the upper part of the “Stormberg” group and the Newark Supergroup, Olsen and Galton (1977, 1984) considered it to lie within the middle part of the Elliot Formation (Olsen and Galton, 1977, 1984; Olsen et al., 2002; Knoll, 2004, Rubidge, 2005). The body-fossil tetrapod content of the Upper Elliot Formation includes non-crocodylian crocodylomorphs (sphenosuchians), crocodyliforms (protosuchids), dinosaurs (sauropodomorphs, theropods, and ornithischians), primitive chelonians, non-mammalian therapsids (trithelodontids, tritylodontids), and basal mammals (e.g. Kitching and Raath, 1984; Olsen and Galton, 1984; Lucas and Hancox, 2001, Rubidge, 2005). In Argentina, rich tetrapod Late Triassic deposits are mainly known from central- western Argentina, although some interesting new dinosaur finds are also known from the Late Triassic of southern Patagonia. In the Bermejo Basin (San Juan and La Rioja provinces), the continental red beds of Los Colorados Formation constitutes the final infilling of the basin and conformably overlies the fluvial Carnian levels of the Ischigualasto Formation (Rogers et al., 1993). The Los Colorados rocks have yielded an exceptional tetrapod record represented by both body fossils and tracks, particularly concentrated the upper third of the succession (Caselli et al., 2000; Arcucci et al., 2004). The tetrapod association includes several basal archosaurs (aetosaurs, “rauisuchians”, ornithosuchids), non-crocodylian crocodylomorphs (sphenosuchian), crocodyliforms (protosuchids), dinosaurs

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 96 (sauropodomorphs, theropods), primitive chelonians, non-mammalian therapsids (dicynodonts, trithelodontids), and tetrapod footprints (“chirotheroid”-type) related to crurotarsal archosaurs (e.g. Rougier et al., 1995; Bonaparte 1997; Arcucci and Coria, 1998; Caselli et al., 2000; Arcucci et al., 2004). In southern Patagonia, the rocks of the El Tranquilo Basin (Santa Cruz Province), southern Patagonia, are wholly continental. The upper part of the succession is included in the Laguna Colorada Formation which preserves dinosaur skeletons, both ornithischians and saurischians (sauropodomorphs), and plant remains of the Dicroidium-type flora and thus, considered Norian in age (Jalfin and Herbst, 1995; Báez and Marsicano, 2001; Pol and Powell, 2007).

Faunal comparisons and correlations Tetrapod biostratigraphy essentially relies on systematic resolution ‒the correct identification of taxa to the lowest possible taxonomic level‒ but intercontinental correlations also need a deep consideration of the phylogenetic affinities of the taxa involved therefore, a hypothesis of kin relationships. Particularly concerning to the dinosaur faunas of Western Gondwana, several recent revisions of early Mesozoic dinosaurs include new cladistic hypothesis regarding the South American and South African taxa. Thus, the recent described theropod dinosaur Dracovenator from the Upper Elliot Formation was considered the sister taxon of Zupaysaurus from the Los Colorados Formation (Yates, 2005). The sauropodomorph Massospondylus from the Upper Elliot levels appears to be closely related to the Coloradisaurus from Los Colorados Formation and Eucnemesaurus from younger levels (Lower Elliot Formation) to Riojasaurus, also from Los Colorados beds (Upchurch et al., 2007, Yates, 2007). Besides, the ornithischian Heterodontosaurus is recorded in both the Upper Elliot and the Laguna Colorada formations (Báez and Marsicano, 2001) and heterodontosaurids, in general, are considered now as basal ornithischians close to the Carnian Pisanosaurus from the Ischigualasto Formation (Butler et al., 2007). Likewise, the australochelid turtle Australochelis from the Southafrican upper “Stormberg” beds appears as the sister-taxon of Paleochersis from Los Colorados Formation (Rougier et al, 1995). Recently, a re-evaluation of the stratigraphic position and ichnological content of the classical footprint levels of the Upper Elliot Formation (Moyeni track-site of Ellenberger, 1974) unambiguously evidences the presence of “chirotheroid”-type footprints (Marsicano et al., 2005). It is important to remark that this footprint assemblage was the originally analyzed by Olsen and Galton to position the TJ boundary in the middle part of the Elliot Formation (e.g. Olsen and Galton, 1977, 1984; Olsen and Sues, 1986). The presence of “chirotheroid”- type of footprints in levels of the Upper Elliot Formation is remarkable and biostratigraphically important because of their widespread use as Triassic “markers” and their absence above the TJ boundary, until now (e.g. Olsen and Galton, 1977, 1984; Olsen et al., 2002a). This

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 97 type of footprints is also recorded in levels of the Los Colorados Formation (Arcucci et al., 2004). According to the present discussion, the position of the TJ boundary is still poorly constrained in the Karoo succession and former arguments to place it in the middle of the Upper Elliot Formation are not longer sustained. Moreover, the close phylogenetic relationships of the dinosaurs from the Upper Elliot with undoubtedly Late Triassic taxa from Southern South America (Los Colorados and Laguna Colorada formations) also suggest that the Early Jurassic age of the Upper Elliot fauna is at least very dubious. Tetrapod fossils alone are not probably sufficient to allow fine-scale biostratigraphy and correlations. Besides the systematic/phylogenetic resolution, reliability also depends on the quality of the fossil record and the comparability of the faunas, usually biased by differences in depositional environments and taphonomic processes. Therefore, independent methods of establishing chronology are of primary importance for testing tetrapod faunal hypothesis. Particularly, the position of the TJ boundary in the Karoo Basin, one of the major continental sequences of Western Gondwana, would have important implications on our understanding of the patterns of diversification of the Early Mesozoic tetrapod faunas in this part of Gondwana and, consequently, on the putative worldwide tetrapod continental extinction at the TJ boundary (e.g. Benton, 1994; Olsen et al., 2002a, 2002b).

References Arcucci, A.B., Coria, R.A., 2003. A new theropod from the Triassic of Argentina. Ameghiniana 40: 217–228.

Arcucci, A. B., Marsicano, C. A., Caselli, A. T., 2004. Tetrapod association and paleoenvironment of Los Colorados Formation (Argentina): a significant sample from western Gondwana at the end of the Triassic. Geobios 37: 557-568.

Báez, A. M. and Marsicano, C. A., 2001. A heterodontosaurid ornithischian dinosaur from the Upper Triassic of Patagonia. Ameghiniana, 38: 271-279.

Benton, M.J., 1994. Late Triassic to Middle Jurassic extinctions among continental tetrapods: testing the pattern. In: Fraser, N.C., Sues, H.-D. (Eds.), In the Shadow of the Dinosaurs. Early Mesozoic Tetrapods. Cambridge University Press, NewYork, pp. 367–397.

Bonaparte, J. F., 1997. El Triásico de San Juan-La Rioja, Argentina y sus dinosaurios. Museo Argentino de Ciencias Naturales Bernardino Rivadavia, Buenos Aires, 190 pp.

Butler R.J., Smith, R., and Norman, D.B., 2007. A primitive ornithischian dinosaur from the Late Triassic of South Africa, and the early evolution and diversification of Ornithischia. Proceedings of the Royal Society of London B 274: 2041-2046.

Caselli, A. T., Marsicano, C. A. and Arcucci, A. B., 2001. Sedimentología y paleontología de la Formación Los Colorados, Triásico Superior (provincias de La Rioja y San Juan, Argentina). Revista de la Asociación Geológica Argentina 56: 173-188.

Duncan, R.A., Hooper, P.R., Rehacek, J., Marsh, J.S., and Duncan, A.R., 1997. The timing and duration of the Karoo igneous event, southern Gondwana: Journal of Geophysical Research, 102, 18127-18138.

Ellenberger, P., 1974, Contribution à la classification des Pistes de Vertébrés du Trias: Les types du Stormberg d’Afrique du Sud. Deuxième Partie: Le Stormberg Supérieur: Palaeovertebrata, Memoire Extraordinaire, 141 pp.

Gradstein, F. M, Ogg, J.G., and Smith, A.G., 2005. A geologic timescale 2004. Cambridge University Press, 589 pp.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 98 Jalfin, J. A., and Herbst, R., 1995. La flora triásica del Grupo El Tranquilo, Provincia de Santa Cruz (Patagonia). Estratigrafía. Ameghiniana, 32: 211-229.

Kitching, J.W., and Raath, M.A., 1984. Fossils from the Elliot and Clarens Formations (Karoo Sequence) of the Northeastern Cape, Orange Free State and Lesotho, and a suggested biozonation based on tetrapods: Palaeontologia Africana, 25: 111-125.

Knoll, F., 2004. Review of the tetrapod fauna of the "Lower Stormberg Group" of the main Karoo Basin (southern Africa): implication for the age of the lower Elliot Formation: Bulletin de la Société Géologique de France, v. 175, p. 73-83.

Langer, M., Ribeiro, A.M., Schultz, C.L. and Ferigolo, F., 2007. The continental tetrapod-bearing triassic of south Brazil. In: Lucas, S.G. and Spielmann, J.A., eds., The Global Triassic. New Mexico Museum of Natural History and Science Bulletin 41, 201-218.

Lucas, S., 1998. Global Triassic tetrapod biostratigraphy and biochronology. Palaeogeography, Paleoclimatology, Palaeoecology 143: 347-384.

Lucas, S., Hancox, J., 2001. Tetrapod-based correlation of the nonmarine Upper Triassic of Southern Africa. Albertiana 25: 5-9.

Marsicano, C.A., Smith, R. and Sidor, C., 2005. Tracking the Triassic-Jurassic boundary in the roof of Africa. In: R. Pankhust and G. Veiga (eds.), Gondwana 12: Geological and Biological Heritage of Gondwana, Abstracts, Academia Nacional de Ciencias, Córdoba, 240.

Olsen, P.E., Galton, P.M., 1977. Triassic-Jurassic extinctions: are they real? Science 197: 983-985.

Olsen, P.E., Galton, P.M., 1984. A review of the reptile and amphibian assemblages from the Stormberg Group of southern Africa with special emphasis on the footprints and the age of the Stormberg. Palaeontologia Africana 25: 87-110.

Olsen, P.E. and Sues, H.-D., 1986. Correlations of the continental Late Triassic and Early Jurassic sediments, and patterns of the Triassic-Jurassic tetrapod transition. In: Padian, K. (Ed.), The beginning of the age of dinosaurs, Cambridge University Press, Cambridge, pp. 321-351.

Olsen, P.E., Kent, D.V., Sues, H.-D., Koeberl, C., Huber, H., Montanari, A., Rainforth, E.C., Fowell, S.J., Szajna, M.J. and Hartline, B.W. 2002a. Ascent of dinosaurs linked to an iridium anomaly at the Triassic-Jurassic boundary. Science 296: 1305-1307.

Olsen, P. E, Koeberl, C., Huber, H., Montanari, A., Fowell, S.J., Et-Touhami, M. and Kent, D.V. 2002b. Continental Triassic-Jurassic boundary in central Pangea: recent progress and discussion of an Ir anomaly. Geological Society of America, Special Paper 56: 502-522.

Padian, K., 1986. Summary and prospectus. In: Padian, K. (Ed.), The beginning of the age of dinosaurs, Cambridge University Press, Cambridge, pp. 363-369.

Parrish, J.M., 1989. Phylogenetic patterns in the manus and pes of early Mesozoic archosauromotph reptiles. In: Gillette, D. D., Lockley, M. G. (Eds.). Dinosaur tracks and traces, Cambridge University Press, Cambridge, pp. 249-258.

Pol, D. and Powell, J., 2007. New information on Lessemsaurus sauropoides (Dinosauria: Sauropodomorpha) from the Upper Triassic of Argentina. Special Papers in Palaeontology 77: 223–243

Rogers, R., C. Swischer III, P. Sereno, A. Monetta, C. Forster, and R. Martinez., 1993. The Ischigualasto tetrapod Assemblage (Late Triassic, Argentina) and K/40 Ar/39 dating of dinosaur origins. Science, 260: 794-797.

Rubidge, B., 2005. Du Toit Memorial Lecture: Re-uniting lost continents - Fossil reptiles from the ancient Karoo and their wanderlust . South African Journal of Geology 108: 135-172.

Rougier, G., De La Fuente, M., Arcucci, A. B., 1995. Late triassic turtles from South America. Science 268, 855- 858.

Upchurch, P., Barrett, P. and Galton, P. 2007. A phylogenetic analysis of basal sauropodomorph relationships: Implications for the origin of sauropod dinosaurs. Special Papers in Palaeontology 77: 57-90.

Yates, A. M. 2005. A new theropod from the Early Jurassic of South Africa and its implications for the early evolution of theropods. Palaeontologia Africana 41:105-122.

Yates, A., 2007. Solving a dinosaurian puzzle: the identity of Aliwalia rex Galton. Historical Biology, 19: 93-123.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 99 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The Paraná Basin: a multi-cycle sedimentary and magmatic intracratonic province of W Gondwana

Edison José Milani Petróleo Brasileiro S.A. – Petrobrás, Rio de Janeiro – RJ – Brazil, [email protected]

Abstract The geological development of the intracratonic Paraná Basin was influenced by the geodynamics of southwestern Gondwana. This domain was continuously affected during almost all the Phanerozoic eon by compressional stresses derived from a persistently active convergent motion between the continent and the oceanic lithosphere of Panthalassa. The Paraná Basin, in spite of being supported by a cratonic basement since its inception, had in its neighbourhood evolving collisional belts fringed by foreland basins. Subsidence and sediment accumulation in the Paraná Basin started during Middle to Late Ordovician times when the Precordillera terrane collided against Gondwana and produced the different contractional phases of the Ocloyic Orogeny. The intraplate response to the compressional stresses related to this orogenic cycle was transtensional reactivation of weakness zones, providing the initial subsidence for the Paraná Basin. Repeatedly during the geologic history of the Paraná Basin orogenic cycles left their signature as periods of accelerated subsidence. Subsidence plots revealed that during Early Devonian and Late Permian times were two other periods when intracratonic subsidence rates increased remarkably. An integrated analysis of the sedimentary record of the Paraná Basin, considering eustatic variations of the sea level and subsidence cycles of southwestern Gondwana led to the conclusion that the stratigraphic cyclicity observed in the Paraná Basin was ultimately influenced by its subsidence history as an intraplate response to geodynamic processes affecting southwestern Gondwana margin.

Introduction The Paraná Basin, a vast sedimentation area located in central-eastern South America, developed during Paleozoic and Mesozoic times and holds a stratigraphic record ranging in age from Late Ordovician to Late Cretaceous times, documenting almost 400 million years of the Phanerozoic geological history for this region of the continent (Figure 1). Six major, second order allostratigraphic units or supersequences (Milani, 1997) are recognized: Rio Ivaí (Caradoc-Llandovery), Paraná (Lockovian-Frasnian), Gondwana I (Westphalian-

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 100 Scythian), Gondwana II (Anisian-Norian), Gondwana III (Upper Jurassic-Berriasian) and Bauru (Aptian-Maastrichtian). Three of them correspond to Paleozoic transgressive- regressive cycles, and the others are Mesozoic continental sedimentary packages with associated igneous rocks. These supersequences are the remnant record of successive phases of sediment accumulation that alternated with times of erosion. In this work, five schematic paleogeologic scenarios were drawn representing specific moments of the evolution of the Paraná Basin and of the adjacent, Gondwanides foredeep domain (Figures 2 to 6). Subsidence analysis (Milani and Ramos, 1998; Milani and de Wit, in press) revealed the existence of notably synchronous episodes of accelerated subsidence in the foreland and in the intracratonic domains, suggesting that such areas might have had a common evolutionary history sharing not only regional sedimentary environments but also mechanisms of subsidence.

Figure 1 – Stratigraphic chart of the Paraná Basin (Milani, 1997) displaying its six major sequences and changing basin characteristics through time. Numbers refer to time positioning of maps of figures 2 to 6.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 101 Regional tectonics The southern margin of Gondwana, particularly the sector that now corresponds to the Andean border of South America, behaved during almost all Phanerozoic time as an active domain of convergence between the paleocontinent and the oceanic lithosphere of Panthalassa (Bahlburg and Breitkreuz, 1991; Gohrbandt, 1993); that was the SAMFRAU Geosyncline of du Toit (1937). A series of terranes reached that region and their accretion to the margin of Gondwana produced important orogenic episodes (Ramos, 1988; 1990). This compressional geotectonic setting influenced the evolution of foldbelts and adjoining foreland basins that developed along southwestern Gondwana, but also affected the cratonic interior (Zalán et al., 1990; Milani, 1992; 1997). The succession of orogenies that marked the Phanerozoic history of southwestern Gondwana, as summarized by Ramos (1988), comprised a series of episodes included in two major tectono-sedimentary-magmatic cycles: the Famatinian (Ordovician to Devonian) and the Gondwanic (Carboniferous to Triassic) cycles. The Famatinian cycle encompasses two pulses of compressional deformation and associated phenomena, the Ocloyic and the Precordilleran orogenies, while the Gondwanic cycle includes the Chanic and the Sanrafaelic orogenies. In the same way that the formation of the Andean belt causes the flexural subsidence observed in western South America (the Bolivian/Argentinian Chaco basin), Paleozoic orogenies influenced the development of adjacent foreland basins. This was deduced by regional subsidence analysis (Milani and Ramos, 1998), that revealed a succession of cycles representing the subsidence history for the Paleozoic foreland domain as a whole. The changing subsidence rates expressed in the backstripping diagrams can be interpreted as induced by the varying intensity of tectonic activity along the orogenic belt, and this correlates very well with important stratigraphic events in the Paraná Basin.

Stratigraphic evolution and basin geometries in the multi-cycle “Paraná Basin” The Rio Ivaí Supersequence, comprising the oldest sedimentary rocks of the Paraná Basin, is particularly significant to the understanding of the inception of the basin because it represents the first cycle of sedimentation in this area that was settled over a consolidated floor, witnessing the establishment of cratonic conditions after the Late Proterozoic/Early Paleozoic Brasiliano Orogeny. Thus the intrinsic characteristics of the Ordovician-Silurian package in terms of occurrence, distribution and geometry of its depocenters, together with sedimentological aspects of this section and its association with magmatic rocks (Milani, 2004), allow direct considerations on the nature and evolution of the initial subsidence of the Paraná Basin. Rio Ivaí rocks occur widespreadly across the Paraná Basin (Figure 2). Their thickness, however, shows a non-uniform distribution with some elongated depocenters striking SW-NE. There is also a general trend of thickening to the west, with the package

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 102 reaching about 1,000 m in the Paraguayan portion of the basin. A regional, fluvial paleocurrent pattern from NE to SW has also been identified (Milani et al., 1996) in the lowermost section of Rio Ivaí package. The complete section includes basal conglomerates and sandstones (Alto Garças Formation), diamictites (Iapó Formation) and fossiliferous shales and siltstones (Vila Maria Formation), spanning the range Caradoc-Llandovery. The shales record environmental conditions related to the maximum flooding of the Ordovician- Silurian cycle. The association of Rio Ivaí sediments with magmatic rocks (Três Lagoas basalt) suggests synsedimentary faulting and some kind of rifting related to the inception of the Paraná Basin.The Ordovician-Silurian package as a whole is poorly contrained in age and can not be adequately evaluated in a quantitative perspective through subsidence analysis.

Figure 2 – Paleogeology of SW Gondwana during the Late Ordovician. Thick dashed line is the present-day outline of the Paraná Basin, included as a reference.

The top of the Rio Ivaí Supersequence is defined by an unconformity surface that had deeply eroded Ordovician-Silurian strata (Figure 1). The blanket-like Devonian Paraná Supersequence rests on the unconformity peneplain, lying over the previous sedimentary package or directly over rocky domains of the basement. The Paraná Supersequence is a

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 103 complete transgressive-regressive cycle of sedimentation, starting with continental to transitional Early Devonian sandy rocks (Furnas Formation) covered by marine shaly sediments (Ponta Grossa Formation) aged from Praguian to Frasnian. The Emsian shales have sedimentologic and stratigraphic characteristics compatible with the maximum flooding of the Devonian cycle (Figure 3), corresponding to the rapid drowning of the shallow Furnas platform. Another basin-scale unconformity surface marks the upper limit of the Devonian package. In fact, the Devonian-Carboniferous boundary is a benchmark of the geology of the Gondwana (López-Gamundí & Rossello, 1993), represented in the Paraná Basin by a lacuna that encompasses almost 55 Ma of the Phanerozoic time.

Figure 3 – Paleogeology of SW Gondwana during the Early Devonian.

Remarkable climatic factors, however, also left their contribution to the appearance of the 55 Ma lacuna. The presence and movements of ice caps related to the great Gondwana glaciation, which peaked in the Mississipian, provided important mechanisms of erosion and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 104 a great obstacle to the accummulation of sediments. After that long period of Mississipian erosion, subsidence resumed in the Paraná Basin during the Westphalian and progressed up to the Early Triassic. The Gondwana I Supersequence followed the climax of glacial conditions. Sedimentation was resumed with deglaciation (Eyles et al., 1993), and an intense sedimentary influx coming from those areas laid open by ice melting allowed depositional processes in which mass flows and resedimentation were important, completely reworking the substratum and defining a singular depositional style for the Westphalian to Sakmarian interval of the Paraná Basin. The 1,500 meter-thick deglaciation-related section (Itararé Group in the south and Aquidauana Formation in the north) is composed dominantly by diamictites intercalated with sandstone packages, having both glacioterrestrial and glaciomarine representatives. The glacial package onlaps the sub-Pennsylvanian unconformity from north to south and extends over progressively wider areas. In the Early Permian, onlapping sedimentation reached the southern portion of the basin. An important phase of structural rearrangement of the basin geometry followed. The regional sense of onlap of sedimentary beds was suddenly inverted (Milani and de Wit, in press), as shown by the northward-wedging retrograding sequence of the Guatá Group. Maximum paleobathymetric conditions for the Gondwana I Supersequence are documented in the Palermo Formation of lowermost Late Permian age. The overlying package, accommodated by a renewed cycle of tectonic subsidence of the basement, is an up to 1,400 meter-thick regressive section (Passa Dois Group) that culminates in Early Triassic eolian sandstones (Sanga do Cabral and Pirambóia formations). Accompanying the Middle to Late Permian deformation along southern Gondwana (Cobbold et al., 1992), a progressive and irreversible continentalization of the depositional systems of the Paraná Basin occurred and can be seen from the sedimentary record of the upper portion of the Gondwana I Supersequence onwards. The development of a huge magmatic arc along SW Gondwana margin – the Choiyoi – allowed high freeboard conditions and the definitive closure of remaining connections with the Panthalassa (Figure 4). By Middle to Late Triassic times, extensional rift basins developed all along the suture zones left by the Paleozoic orogens (López-Gamundí et al., 1994). Some of these lacustrine basins developed upon the Permian package of Southern Paraná Basin in Brazil (Figure 5), and are not known in other domains of the basin. Sandy deserts covered definitively the entire basin and extrapolated over a wide adjacent region during the Late Jurassic (Figure 6; Botucatu Formation), followed by the Early Cretaceous basaltic lavas and intrusives of the Serra Geral Formation, heralding SW Gondwana breakup and the opening of South Atlantic ocean. Subsidence and sediment accumulation in the Paraná Basin had their last stage during the Late Cretaceous with the Bauru Supersequence, a thin package of continental sandy to conglomeratic rocks that filled up an interior sag basin developed over the lava pile.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 105

Figure 4 – Paleogeology of SW Gondwana during the Middle Permian.

References Bahlburg, H. and Breitkreuz, C., 1991, Paleozoic evolution of active margin basins in the Southern Central Andes (Northwestern Argentina and Northern Chile): Journal of South American Earth Sciences, v. 4, n. 3, p. 171- 188.

Cobbold, P.R., Gapais, D., Rossello, E.A., Milani, E.J., and Szatmari, P., 1992, Permo-Triassic intracontinental deformation in SW Gondwana, in De Wit, M.J. and Ransome, I.D., eds., Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa: Rotterdam, Balkema, p. 23-26.

Du Toit, A.L., 1937, Our wandering continents. London, Oliver and Boyd, 366 p.

Eyles, C.H., Eyles, N., and França, A.B., 1993, Glaciation and tectonics in an active intracratonic basin: the Late Palaeozoic Itararé Group, Paraná Basin, Brazil: Sedimentology, v. 40, p. 1-25.

Gohrbandt, K.H.A., 1993, Paleozoic paleogeographic and depositional developments on the central proto-Pacific margin of Gondwana: their importance to hydrocarbon accumulation: Journal of South American Earth Sciences, v. 6, n. 4, p. 267-287.

López-Gamundí, O.R., Espejo, I.S., Conaghan, P.J. & Powell, C.M.A. 1994. Southern South America. In: Veevers, J.J. & Powell, C.M.A. (eds) Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland. Geological Society of America, Boulder, Memoir, 184, 281-329.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 106

Figure 5 – Paleogeology of SW Gondwana during the Late Triassic.

López-Gamundí, O. and Rossello, E.A., 1993, Devonian-Carboniferous unconformity in Argentina and its relation to Eo-Hercynian orogeny in southern South America: Geologische Rundschau, v. 82, p. 136-147.

Milani, E.J., 1997, Evolução tectono-estratigráfica da Bacia do Paraná e seu relacionamento com a geodinâmica fanerozóica do Gondwana sul-ocidental [Dr.Sc. dissert.]: Porto Alegre, Rio Grande do Sul, Brasil, Universidade Federal do Rio Grande do Sul, 255 p., il.

Milani, E.J., 1992, Intraplate tectonics and the evolution of the Paraná Basin, S Brazil, in De Wit, M.J. and Ransome, I.D., eds., Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa: Rotterdam, Balkema, p. 101-108.

Milani, E.J., Assine, M.L., Soares, P.C., and Daemon, R.F., 1996, A Seqüência Ordovício-Siluriana da Bacia do Paraná: Boletim de Geociências da PETROBRAS, v.9, n. 2/4, p. 301-320.

Milani, E. J. and de Wit, M. J., in press, Correlations between the classic Paraná and Cape-Karoo sequences of South America and southern Africa and their basin infills flanking the Gondwanides: du Toit revisited. In: Pankhurst, R. et al. (eds), Western Gondwana - the Ties that Bind. The Geological Society, London.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 107

Figure 6 – Paleogeology of SW Gondwana during the Late Jurassic.

Ramos, V.A., 1988, Late Proterozoic - Early Paleozoic of South America - a collisional history: Episodes, v. 11, n. 3, p. 168-174.

Ramos, V.A., 1990, Field guide to geology of the Central Andes (31o-33o SL): Buenos Aires, Universidad de Buenos Aires, Central Andes Field Seminar, 68 p.

Zalán, P.V., Wolff, S., Astolfi, M.A.M., Vieira, I.S., Conceição, J.C.J., Appi, V.T., Neto, E.V.S., Cerqueira, J.R., and Marques, A., 1990, The Paraná Basin, Brazil, in Leighton, M.W., Kolata, D.R., Oltz, D.F., and Eidel, J.J., eds., Interior cratonic basins: Tulsa, American Association of Petroleum Geologists Memoir 51, p. 681-708.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 108 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Cyclostratigraphy during the Carboniferous glaciations in central western Argentina: glacial ageism and tectonic framework

Dr. Pablo José Pazos Universidad de Buenos Aires, CONICET – Bs.As. – Argentina, [email protected]

Abstract The understanding of the tectonic, sedimentary and climatic evolution of Carboniferous basins of central-western Argentina experienced remarkable advances in the last decade. However, refined correlations are still debatable mainly due to mixing of lithostratigraphic and genetic schemes. The study of the sedimentary architecture of key basins (Paganzo/Rio Blanco and San Rafael) is crucial to for correlations intra and interbasinally. Unquestionably, climatic fluctuations controlled the internal architecture of sequences but tectonism is responsible for many unconformities. The recent discovery of two glacial horizons in a Mississippian unit indicates an ageism of the starting of cold conditions and contradicts some climatic interpretations suggesting temperate and humid conditions. Combining the analysis of paleotopographic and facies changes with glacial horizons it is posible to conclude icehouse conditions prevailed from the early Visean until Serpukhovian. Active tectonism with rapid subsidence increased across the Mississippian. The Precordillera, remained as a positive area during most of the Mississippian. In the beginning of the early Pennsylvanian it was a low exposed topography with fjord valleys draining waters from glacial centres located principally in the Pampean Ranges. Local subsidence was high associated to reactivation faults and alignments producing endorreic “trap” basins. However, in general high frequency sequences are developed during a time of gently postorogenic subsidence. The San Rafael Basin contains a Pennsylvanian record that exhibit an internal architecture not differentiable from the Paganzo Basin. The upper sequence is pre-latest Pennsylvanian-Sakmarian and shows reversion in paleocurrents indicating the eastern migration of the magmatic arch. This unconformity is also randomly observed in the Paganzo Basin but clear in the western flank of the Precordillera. In the San Rafael basin a period of uplifting is recorded by unroofing during the lower Permian until massive distribution of piroclastic-volcanic flows and aeolian dunes as part of the Cochicó Group. Conversely, in the Paganzo Basin red beds of fluvial systems and some aeolian intervals dominated the landscape in a broken foreland domain.

Introduction Gondwana experienced one of the most severe glaciations of the geological record during the late Paleozoic, with consequent sea level changes detectable in cyclothems of Euroamerica (e.g. Maynard, 1992; Aitken and Flint, 1994, Goldhammer et al., 1994;

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 109 Hampson et al., 1997). In Gondwana, are well known at least III glacial episodes (see Isbell et al., 2003) spanned from Late Devonian up to Early Permian. In Argentina is usual to accept that only the episode II is important and was in apogee during the early Pennsylvanian and of alpine type (Lopez Gamundi et al., 1994). Glacial centres were situated in the named Proto-Precordillera. Interestingly this idea is the result of the misunderstanding of a pioneer map by Salfity and Gorustovich (1983). In consequence, a mountain belt was suggested as a barrier separating the Paganzo and Rio Blanco- Calingasta basins during postglacial times. However this feature has been severely questioned as a big source of glacial-debris based on the absence of western provenance in the eastern depocentres and paleocurrent patterns. Moreover, many researchers claimed for the Protoprecordillera abolition as a division between the Rio Blanco and Paganzo basins (e.g, Perez et al., 1993; Fernandez and Seveso and Tankard, 1995; Gonzales Bonorino and Eyles, 1995; Pazos 2002b). Interestingly, many facies and ichnological models rest on the premise of this topographic barrier that converted the Paganzo Basin in a fjord-lake with punctuated marine incursions (e.g. Buatois and Mangano, 1995, Buatois et al., 2006). However, other interpretations suggested a brackish marine realm with enormous amount of melt water from glacial recess (Pazos, 2000a, 2000b, 2002b). The San Rafael Basin, situated further south (Figure 1) is an engulfment directly connected with the Paleopacific Ocean. It is a very good area to discern the influence of climatic and tectonic changes, facies models and also to propose stratigraphic correlations and comparative evolution between basins discussed in this contribution because of the simple stratigraphy and low level of tectonic overprint. Crucial information to understanding the geological evolution has been overviewed and is the aim of the present contribution to be used in correlations based on sequence stratigraphic principles rather than merely litostratigraphy. This contribution is mainly based on personal data published in the last decade (e.g. Pazos, 1996, 2000a, 2000b, 2002a, 2002b, Pazos et al., 2005a, 2005b) from basins of central western Argentina (Paganzo-Rio Blanco) and San Rafael Basin (Loss and Pazos, 2006; Pazos et al, in press). The understanding of the tectonic, climatic and geological evolution impacts in the correlations of this portion of W Gondwana but also modifies some previous schemes that prevailed during years. Some aspects like roll play of glacial onset and retreats, tectonism, glacioisostatic rebound and facies distribution are addressed in order to discuss a most accurate correlation between basins.

Glacial Episodes and Ageism Glacial record in western Gondwana has experienced ageism after the introduction of absolute dating. Argentina escaped to this tendency until were confirmed two glacial horizons in the Maliman Formation of late Tournaisian-early Visean age by Pazos et al. (2005a). It

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 110 contradicts humid and temperate interpretations traditionally offered by the early Mississippian (e.g. Lopez Gamundi et al., 1992) and it agrees with cold affinities found in plants and fossil invertebrates. The lower glaciomarine horizon correspond to the interval where has been mentioned Protocanites scalabrini. Previously, Carrizo and Azcuy, (1997) suggested a glacial onset during the early Carboniferous but without a precise stratigraphic position. On the other hand, late Visean glacial deposits were reported by Limarino et al., (1990) in the Cortaderas Formation in the Rio Blanco Basin, almost contemporary with the oldest record of NBG flora. Other record (Agua del Jagüel Formation) in the same basin has been correlated to the upper Cortaderas glacial deposits (Limarino et al., 2006) rather than the Malimán, and it looks like correct. However, the widespread glaciation in this area of Argentina is traditionally considered early Pennsylvanian (Namurian) based on palynomorphs distinguishable from the late Visean record according to Limarino et al., (2006). However, the first one may indicates the onset and advances (proglacial) while the last one represents the glacial retreat with the progressive glacioeustatic sea level rise. It implicates not strong tectonism between both records. The paleotopography progressively was reduced and isostatic rebound in the Pampean Ranges is responsible for forced regressions and remarkable progradation from the east. Ageism of the glacial record, in facts, permits to explain the cyclothems and punctuated transgressions observed in Euroamerica.

Figure - 1 Paleogeography of the Pennsylvanian and tectonic framework.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 111 Unconformities The Mississipian rests on an angular unconformity post mid Devonian and pre Late Tournaisian in age attributed to the Chanic Phase. Deposition started with, scarce mature conglomerates and meandering fluvial facies (Pazos et al., 2005b) and high frequency sea level fluctuations related to glacial advances and retreats (Pazos et al., 2005a). Paleoenvironmental evidence suggests moderate subsidence across the Malimán Formation. Abundant shallow marine, marginal marine and fine grained fluvial sandstones, with rare Lower Paleozoic clasts appears in scarce conglomerates indicate relative low gradients. Additionally, alternating sandstone provenance in the Malimán Formation includes green lithic types, related to glacial episodes and local source and light brown arkosic types coming from distal sources areas. However, dropstones in glaciomarine deposits indicate not Pampean provenance and a sea level glaciation. It clearly suggests that ice contact facies were situated in Precordillera or to the west (?), while glacial recess with reworking indicated for prograding deltas and fluvial system arrived from distal eastern areas after sea level rise. Accommodation space was reduced by glacial onset and transgressive-regressive cycles seem related to sea level fluctuations by glacial advances and retreats. After the second glacial episode in the Malimán Formation, the postglacial transgression was almost completely compensated by tectonic uplift indicating strong agradational to progradational wedges from the east. The tectonic uplift apogee in Precordillera is recorded in the angular unconformity that separates the Malimán and Cortaderas formations. This interpretation is based on the exposition and erosion of Devono-Carboniferous units recorded from clasts in a thick congloremeratic interval. Some of them contain palynomorphs only recorded in the lower section of the Malimán Formation (LateTournaisian?). Conglomerates were deposited by fluid flows of high energy. Angularity of the unconformity varies upward indicating progradation and downlap during reduced accommodation space. This unconformity is interpreted as tectonic in origin (Fig 2). The next unconformity is realistically in most places an artefact of the Andean tectonism (see Figure 2), it occurs between the Lower and Late Carboniferous and was interpreted like the onset and recess of a same glacial episode (see previous section). This unconformity is recognised in the base of the Guandacol Formation at the homonymous hill. Finally a pre-latest Pennsylvanian-Sakmarian unconformity is recorded in Paganzo and San Rafael Basins and constitutes the first evidence of the San Rafaelic tectonic phase (Figure 2), mainly developed in the Permian with the progressive reversion of the provenance.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 112

Figure - 2 Composed stratigraphic chart including glacial deposits (triangles). Black: glacial diamictites. Gray: reworked diamictites, mainly. Yellow: Novel glacial horizons. Unconformities marked in red (tectonic) or mixed (green).

Pennsylvanian Paleotopography and Paleoenvironments Glacial retreat exposed carved paleovalleys in the Paganzo Basin (Andreis et al., 1986; Kneller et al., 2004) descending from paleohighs situated in the Pampean Ranges. Glaciers moved following old tectonic alignments mainly southeast-northwest orientation. Some depressions (Valle Fértil alignment) acted like a sink that retained sediments deposited like many types of subaqueous gravity flows (Pazos, 1996). High variations in thickness agree with a local subsidence related to reactivation of old faults (Pazos, 2000) in a transtensional tectonic framework (Fernandez Seveso and Tankard, 1995). Postglacial transgressions extended further east indicating an irregular and twisted paleotopography with radical facies shifting in short distances (e.g. Pazos 2002a). Fjords with brackish waters in a disputable ichnological interpretation are frequent in the glacial to postglacial transition (see Pazos 2000b, 2002b, Pazos et al., in press and Buatois et al., 2006). A series of high frequency sequences glacioeustatically controlled (Pazos 2002a) are in agreement with a period of tectonic tensional phase mainly related to a transtensional type (Fernandez and Tankard, 19995; Limarino et al., 2006). Protoprecordillera almost disappeared for the latest Carboniferous and glacioisostatic rebound explain forced regression. The magmatic arch

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 113 was located to the west (Chile) and the Pampean Ranges to the east remained the main source during the Pennsylvanian and earliest Permian times. A comparison with the neighbour San Rafael Basin confirms the proposed evolution. There the glacial record starts with a highly irregular paleogeography and paleotopography (Loss y Pazos, 2006; Pazos et al., in press) like in the Paganzo Basin. Facies architecture is organised high frequency sequences (Loss and Pazos 2006). But the top one evidence reversion in paleocurrents and was considered the first evidence of the advance of the magmatic arch to the east (Espejo et al., 1996). This sequence starts previously to the marine transgression corresponding to the Carboniferous-Permian boundary in the Paganzo basin (see Limarino et al., 2006). late Lower Permian deposits strongly differ in both basins. In the San Rafael Basin unroofing is clear. Conversely, alluvial fans and thick volcaniclastic deposits typical of the Cochicó Group are absent in the Paganzo Basin that only contains red beds.

References

Aitken, J. and Flint, S. S. (1994). High-frecuency sequences and the nature of incised-valleys fill in fluvial systems of the Breathitt Group (Pennsylvanian), Apalachian foreland basin, easthern Kentucky. In: Incised- valley systems: Origin and Sedimentary Sequences, SEPM Special Publication, v. 51, pp 353-368. Andreis, R., Leguizamón, R. and Archangelsky S. (1986). El paleovalle de Malanzán: nuevos criterios para la estratigrafía del Neopaleozoico de la Sierra de Los Llanos, La Rioja, Argentina. Bol. Acad. Nac. Cs. Cba., v. 57, pp 3-119. Buatois, L. and Mángano, M. G. (1995). The paleoenvironmental and paleoecological significance of the lacustrine Mermia ichnofacies: an archetypical subaqueous nonmarine trace fossils assemblege, Ichnos, v. 4, pp 151-161. Buatois, L. R. G. Netto, M. G. Mangano and P. Balestieri (2006) Extreme freshwater release during the late Paleozoic Gondwana deglaciation an dits impact on coastal ecosystems. Geology 34(12): 1021-1024. Carrizo, H. A. and Azcuy, C. A. (1997). The Phytozones of the Early Carboniferous of Argentina and age of the discordances: A discussion. Rev. Univ. Guarhullos- Geociencias II (special), pp 19-27. Espejo, I., Andreis, R.R. and Mazzoni, M.M. (1996). Cuenca San Rafael, in Archangelsky, S., ed., El Sistema Pérmico en la República Argentina y la República Oriental del Uruguay: Córdoba, Academia Nacional de Ciencias, p.177-185. Fernández Seveso, F. and Tankard, A. (1995). Tectonics and Stratigraphy of the Late Paleozoic Paganzo Basin of Western Argentina and its Regional Implications, In: Tankard, A., Suárez, R. y Welsink, H (Eds), Petroleum Basins of South América Am. Ass. Petr. Geol., Mem. 62, 285-301. Goldhammer, R. K., Oswald, E.J. and Dunn, P. A. (1994). High-frequency, glacio-eustatic cyclicity in the Middle Pennsylvanian of the Paradox Basin: an evaluation of Milancovitch forcing. In Eds. P. L. de Boer and D. G. Smith: Orbital Forcing and Cyclic Sequences. Spec. Publs Int. Ass. Sediment., v. 19, pp 243-283. González Bonorino, G. and Eyles, N. (1995). Inverse relation between ice extent and the late Paleozoic glacial record of Gondwana. Geology, v. 23 (11), pp 1015-1018. Hampson, G. J., Elliott, T and Davis, S. (1997). The application of sequence stratigraphy to Upper Carboniferous fluvio-deltaic strata of the onshore UK and Ireland: implications for the southern North Sea. J. Geol. Soc. London, v. 154, pp 719-733. Isbell, J.L., Miller. M. F. and Wolf, K.L. (2003.) Timing of late Paleozoic glaciation in Gondwna: Was glaciation responsible for the development of northern hemisphere cyclothems?. In: Chan, M.A. and Archer, AA.W. eds. Extreme depositional environments: Mega end members in geological time: Geological Society of America Special Paper 370, P. 5-24. Kneller, B., Milana, J.P., Buckee, C., and al Ja'aidi, O., (2004). A depositional record of deglaciation in a paleofjord (Late Carboniferous [Pennsylvanian] of San Juan Province, Argentina): The role of catastrophic sedimentation: Geological Society of America Bulletin, v. 116, p. 348-367. Limarino, C., Page, R., and A. Caselli, (1993). Origen y significado estratigrafico de las diamictites del membro superior de la Formacion Cortaderas (Carbonífero). Precordillera de San Juan. XII Congreso Geológico Argentino.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 114 Limarino C., and S. Cesari (1993). Reubicación estratigrafica de la Formacion Cortaderas y definición del Grupo Angualasto (Carbonifero inferior, Precordillera de San Juan). Revista de la Asociación Geológica Argentina 47: 61-72. Limarino, C. A. Tripaldi, S. Marenssi and L. Fauqué (2006). Tectonic, sea-level and climatic controls on the Late Paleozoic sedimentation in the western basins of Argentina. Journal of South American Earth Sciences 22: 205- 226. López Gamundí O. and Martínez, M. (2000). Evidence of glacial abrasion in the Calingasta-Uspallata and western Paganzo basins, mid-Carboniferous of western Argentina. Paleaogeogr. Palaeoclim. Palaeoec., v. 159: 145-165. López Gamundí, O., Limarino, O. and Césari, S. N. (1992). Late Paleozoic paleoclimatology of central-west Argentina. Palaeogeog. Palaeoclim. Palaeoec., v. 91, pp 305-329. López Gamundí, O., Espejo I., Conaghan P., Mc. Powell, C. and Veevers, J. (1994). Southern South America. In: Veevers, J and Mc.Powell, C (Eds). Permian-Triassic Pangean Basins and Foldbelts along the Panthalassan Margin of Gondwanaland. Geol. Soc. Am. Mem. 184, pp 281-329. Loss, L. and P.J. Pazos (2006.). Arquitectura depositacional de la Formación El Imperial (Carbonífero superior), en el Cañon del Atuel, Cuenca San Rafael, Mendoza, Argentina. IV Congreso Latinoamericano de Sedimentología. Bariloche 11/2006. pp. 130. Maynard, J. R. (1992). Sequence Stratigraphy of the Upper Yaedonian of Northern England. Marine and Petroleum Geology, v. 9, pp 197-207. Pazos, P. J. (2000a). Analisis litofacial y trazas fosiles de las Formaciones Guandacol y Tupe en el ambito Occidental de la Cuenca Paganzo. Tesis Doctoral [PhD Thesis] Universidad de Buenos Aires, p 467. Pazos, P. J. (2000b). Trace fossils and facies in glacial to postglacial deposits from the Paganzo Basin, (late Carboniferous), central Precordillera, Argentina. Ameghiniana, v. 37 (1), pp 23-38. Pazos, P.J., (2002a). The late Carboniferous glacial to postglacial transition: facies and sequence stratigraphy, western Paganzo Basin, Argentina: Gondwana Research, v. 5, p. 467-487. Pazos, P.J., (2002b). Palaeoenvironmental framework of the Late Palaeozoic glacial to postglacial transition in the Paganzo-Calingasta Basin (South America) and Karoo-Kalahari Basins (southern Africa): Ichnological implications: Gondwana Research, v. 5, p. 619-640. Pazos, P. J., M. Di Pasquo, Rodríguez Amenabar, C. (in press). Trace Fossils in a Late Paleozoic glacial to postglacial transition. San Rafael Basin, Argentina. In: Ichnology at the crossroads: a multidimensional approach to the science of organism-substrate interactions (Bromley et al. eds.) SEPM, Special Publication on Ichnology. 34 pp. (2007) Pazos, P. J., R. Amenabar, C., di Pasquo, M. M. and C. L. Azcuy (2005). Are The Late Tournasian-Visean Sequences In Central Western Argentina Climatically Controlled? The Maliman Formation data set. Gondwana 12. “Geological and Biological Heritage of Gondwana. Mendoza. Pag. 284. Pazos P.J., M. di Pasquo y C. R. Amenazar (2005). La sección basal de la Formación Mal imán (Carbonífero inferior) en la quebrada Don Agustín, Provincia de San Juan, Argentina: rasgos sedimentarios y paleontología. XVI Congreso Geológico Argentino. Actas III. Pags, 145-152. Pazos, P.J. (1996): Facies de talud en el Carbonífero de Agua de la Pena. Sierra de Valle Fértil San Juan. VI Reunión Argentina de Sedimentología Actas I. (147-153) Pérez, M., Fernández Seveso F., Alvarez L. and Brisson, I. (1993). Análisis ambiental y estratigráfico del Paleozoico Superior en el área del anticlinal de Huaco, San Juan, Argentina. XII Int. Cong. of Geol. and Estr. Carb.-Perm., v. 2: 297-318. Salfity, J. and Gorustovich, S. (1983) Paleogeografía de la Cuenca del Grupo Paganzo (Paleoz. Sup.), Rev. Asoc. Geol. Arg., v. 58, pp. 437-453.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 115 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Silurian and Devonian world–wide flooding events in the Paraná Basin

Egberto Pereira1, Sérgio Bergamaschi1, René Rodrigues1, Maria Suzana Pessôa de Souza1 ¹ Faculdade de Geologia, UERJ, Rua São Francisco Xavier nº. 524, sala 2020, bloco A, 2º andar - Rio de Janeiro, Brasil, 20559-900, [email protected]

Introduction A high resolution Silurian and Devonian stratigraphy of the Paraná Basin, based on borehole data, has been achieved with an integrated study of stratigraphy, biostratigraphy and geochemistry. The data produced by this study allowed the characterization of the two most important flooding events (Llandovery and Late Givetian / Early Frasnian) observed in the Gondwana. The area studied in this work includes the three sub-basins of the Paraná Basin, characterized as pre-Carboniferous strata: the Alto Garças Sub-Basin, located in the northern portion of the Paraná Basin and the Apucarana Sub-Basin, in the southern portion of the same Basin (Ramos, 1970; Melo, 1988; Pereira et al., 1998) and the new compartment, named the Eastern Paraguay Sub-Basin, which was informally proposed by Grahn et al. (2000) based on the sedimentological and biostratigraphical aspects of the pre-Carboniferous sediments in Paraguay. The whole Paraná Basin covers approximately 1,600,000 km2 of South America, comprising areas in southern and central Brazil, eastern Paraguay, central Uruguay and northeastern Argentina. To characterize the flooding events were used the TOC (Total Organic Carbon) contents of the studied sections and Spectral gamma-ray data. In the wells studied in this paper, the TOC values increase continuously up to the world-wide flooding events, which are marked by a TOC peak. Following these peaks, the TOC values decrease quickly. In the context of the stratigraphy of sequences these events represent a second order Maximum Flooding Surface (MFS). This

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 116 second order cycle is composed of high frequency cycles which are marked by the abrupt shift of the TOC contents. These TOC cyclic variations allow the characterization of chemostratigraphy units, which correspond to the high-resolution 3rd order sequence. The TOC content preserved in the sediments has a direct relationship with the anoxic / disoxic conditions of the environments. Under normal conditions of oxygenated water, the organic matter degradation is relatively fast during its transportation from the surface of the water to the sediment-water interface (Rodrigues, 2005). Usually, only a very small percentage of the primarily produced organic matter is preserved in the sediment (Emerson & Hedges, 1988). The anomalous TOC values, more than 1%, preserved in the sedimentary record indicate that better preservation conditions took place in the sedimentary basin. These conditions occurred during times of heavy stratification of the water column, associated or not with high primary productivity. In this situation, the dissolved oxygen is depleted, and anoxic bottom conditions are developed. The anoxic conditions result from restricted vertical circulation of seawater and / or high biological productivity (Myers, 2004). In the geologic record these conditions are present in black shales and condensed section. These sediments are marked by very high concentrations of organic carbon (TOC>10%). Rodrigues (2005) describes that the black shales depositional conditions involve increasing primary productivity and inflow of organic matter, associated to improved organic matter preservation and limited clastic inflow. Recently, some authors (Myers, 2004; Sutton et al., 2004; Lüning et al., 2004; Rodrigues, 2005; and some references therein) described the relationship between TOC content and the stratigraphy of sequences. As discussed before, the organic matter preservation depends on many factors. Myers (2004) pointed out that the most important factors are the physiogeography of the basin, climate, terrestrial organic productivity, marine aquatic organic productivity, oceanic circulation, sedimentation rate and water depth. With the exception of climate and oceanic circulation, the other factors described by Myers (2004) are influenced by relative sea level change. In the low stand conditions, the high sediment influx product, the dilution of the marine organic matter content and the terrestrial organic matter is highly oxidized. In this context the organic matter content preserved in the sediments is low and the consequent TOC values are inexpressive. The U content, following the same model, shows a decrease in the absolute value, and there is a relative increase of the Th. In consequence, the GR profile shows low values. In the transgressive system tract, the fast increase in the relative sea level caused the shoreline to retreat landward. It resulted in a progressive extension of the shallow-marine shelf deposition and in a reduction of the clastic sediment supply. According to Lüning et al. (2004), the transgression has likely led to sediment starvation because the detritus became trapped in

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 117 the river mouths, thus preventing the dilution of the organic matter on the shelf. These conditions permitted, progressively, the expansion of the distal deep area of the basin until it reached the maximum flooding surface (MFS). There has occurred, in this basin, an extension of the anoxic conditions, which permitted an increase in the organic matter preservation. The retreat of the sediments in shoreline decreases the dilution of the organic matter depositional setting. In contrast with the low stand system tract, the TOC values increased, reaching the maximum in MFS. The TOC and U show a close relationship of the organic-rich strata to the organic matter type I/II. According to Lüning et al. (2004), this relationship is based on the fact that in the seawater U6+ is carried in solution and it is fixed under oxygen-depleted conditions during the deposition. Usually, the gamma-ray (RG) peak coincides with elevated TOC values that indicate the MFS. Therefore, Creaney & Passey (1993) noted that many marine source rocks are characterized by an initial abrupt upward increase in the organic content, in constrast with background values, and a subsequent gradual decrease in the organic content (Myers, 2004). These authors attributed this pattern to the control of the organic carbon contents by the clastic sedimentation rate under anoxic bottom water conditions. The rapid increment of TOC results from the retreat of the sediments in the source during the transgression. The subsequent gradual decrease in TOC reflects the increase in the clastic sediment supply and the dilution of organic carbon during highstand progradation. In this context, the MFS is positioned in the turning point of the TOC curve, being considered by the majority of the authors to be partly contained in the lower highstand, and partly in the upper transgressive systems tracts (Sutton et al., 2004; Posamentier & Allen, 1999).

Methods Total organic carbon measurements were conducted on samples collected from 50 to 50 cm in the wells studied. In the Paraguayan part of the Paraná Basin the 269-R1 and RD-116 wells better record the Llandovery event. In the Brazilian part of the Paraná Basin the RSP-1 better shows the Late Givetian / Early Frasnian flooding event. The samples were pulverized in an agate mortar and the organic carbon and sulphur contents were measured using a LECO CS-444 Carbon-Sulphur analyzer. The inorganic carbon content was carefully removed from samples by repetitive addition of 0.5 N HCL (under warming). Spectral gamma-ray measurements were carried out in the same samples analyzed for TOC using a gamma-ray spectrometer (model GRS 2000, manufactured by Geofyzika, Czech Republic). A 2-minute standard period of time was selected for each measurement. The natural radioactivity in rocks originates mostly from uranium, potassium and thorium. The spectrometer

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 118 differentiates between these elements by identifying typical peaks in the gamma-ray energy spectrum (Lüning & Kolonic, 2003).

Results The Silurian sediments studied (Vargas Peña Formation) crop out in a large area of the eastern Paraguay Sub-Basin. The sediments consist of a thick (ca. 200m) and monotonous succession of black shales and siltstones of Llandoverian age (Mauller et al., 2004). In the studied wells the flooding event (MFS) is characterized by an increase of the total organic carbon (TOC). The event shows that the TOC content is greater than 1.5%. Based on organic and inorganic geochemical data, these MFS were characterized as anoxic/disoxic depositional events. The Silurian record observed in the 269-R1 well can be subdivided into five chemostratigraphic units based on the distribution pattern of major elements. The MFS is characterized by a positive excursion of the Cu, Ni, Co, Zn, Pb, Cr, Fe and Mn values. (Pereira et al., 2003). This anomalous event, recorded in Paraguay, is also observed in the Brazilian part of the Paraná Basin (Vila Maria Formation) and in other Paleozoic basins of South America, as well in the West Gondwana. The Devonian strata recorded in the Brazilian part of the Paraná Basin can be subdivided in 3rd order sequences. In the sequences, the transgressive system tract is characterized by a continuous increase of the total organic carbon content (TOC). The TOC reaches maximum value in the maximum flooding surfaces (MFS). A peak in the TOC curve normally characterizes this surface. After this point, the TOC content decreases, showing a dilution of the organic matter in the sedimentary record associated to the increase of the influx sediment in the highstand context. This abrupt TOC shift defines the 3rd order sequence boundary. The Devonian sediments recorded in the Paraná Basin took place in a ramp basin context. In this situation the lowstand system tracts are rarely preserved. The sequence limit is understood as an amalgamated surface involving the sequence boundary (SB) and the transgressive surface (TS). In some sequences, lowstand wedge deltaic / upper shoreface deposits associated to forced regression, and limited upward by a transgressive surface, also are preserved. The transgressive process reworked the backstepping depositional geometries deposited in the shoreline, in lowstand context, while in the shelf area there is the organic matter concentration in function of the trapped sediment in shoreline. This observed framework is due to the contrast between the TOC value from highstand and transgressive tracts. In the 3rd order context, the TOC values start in the transgressive tract around 0.5 % and vary up to 2 %, approximately. In contrast, the distribution pattern of the TOC content in the highstand tract ranges from 0 to 1 %. The values decrease from a peak around the MFS, mostly over than 1%,

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 119 to very low values (0 to 0.5 %). In special, in the RSP-1 well, the abrupt contrast between highstand and trangressive TOC values is marked by clear surfaces. These surfaces are interpreted like sequence boundaries. The MFS values range from 1 to 2.7 %. The MFS values increase progressively from the base to the top of the Devonian sedimentary pile. The increase of TOC values is in accordance with the intense Devonian transgression, which is world-wide known (Johnson et al, 1985). The maximum value of the whole section is plotted in the early Frasnian age. Radioactive shales of Frasnian age occur in the most important Brazilian Paleozoic basins (Solimões, Amazonas, Parnaíba and Paraná), where they display high concentrations of organic matter. Rodrigues et al. (2005), described that the organic geochemical data in these shales suggest a dominant algal contribution to the organic matter, as well as an anoxic/disaerobic depositional environment. The radioactive shales correspond to a condensed section related to the Devonian maximum marine transgression.

Conclusions This study shows the influence of sea-level changes on the depositional setting over wide geographic areas. Thus, the apparent synchronism between the MFS in the Paraná Basin and other intracratonic basins in South America and West Gondwana suggests an important eustatic control during the sedimentary evolution in the Silurian and Devonian, especially during the Llandovery and Late Givetian / Early Frasnian interval.

References Creaney, S. & Passey, Q.R. 1993. Recurring patterns of total organic carbon and source rock quality within a sequence stratigraphic framework. AAPG Bulletin, 77: 386 - 401.

Emerson, S. & Hedges, J.I. 1988. Processes controlling the organic carbon content of open ocean sediments. Paleoceanography, 3: 621-634.

Grahn, Y., Pereira, E. & Bergamaschi, S., 2000, Silurian and Lower Devonian Chitinozoan biostratigraphy of the Paraná Basin in Brazil and Paraguay: Palynology, v. 24, p. 143-172.

Johnson, J.G., Klapper, G. & Sandberg, C.A. 1985. Devonian eustatic fluctuations in Euramerica. Geological Society of American Bulletin, 96: 567-587.

Lüning, S. & Kolonic, S. 2003. Uranium spectral gamma-ray response as a proxy for organic richeness in black shales: applicability and limitations. Journal of Petroleum Geology, 26: 153-147.

Lüning, S.; Wendt, J.; Belka, Z. and Kaufmann, B. 2004. Temporal-spatial reconstruction of the early Frasnian (Late Devonian) anoxia in NW Africa: new field data from the Ahnet Basin (Algeria). Sedimentary Geology, 163: 237- 264.

Mauller, P.M., Pereira, E., Grahn, Y. & Steemans, P., 2004, Análise bioestratigráfica do intervalo llandoveriano da Bacia do Paraná no Paraguai Oriental: Revista Brasileira de Paleontologia, v. 7, n. 2, p. 199-212.

Melo, J.H.G. The Malvinokraffric realm in the Devonian of Brazil. In: McMillean, N.J.; Embry, S.F.; Glass, D.J. (eds). Devonian of the World. Canadian Society of Petroleum. Geologist Memoir, 1(14): 669-703.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 120 Myers, K.J. 2004. Organic-rich facies and hydrocarbon source rock. In: D. Emery and K. J. Myers (eds), Sequence Stratigraphy, Blackwell Science: 238 – 257.

Pereira, E., Bergamaschi, S. & Rodrigues, M.A., 1998; Sedimantary Evolution of the Ordovician, Silurian and Devonian Sequences of the Paraná Basin in Brasil: Zentralblatt fur Geologie und Paläontologie Teil I, v. 1, n. 3-6, p. 779-792.

Pereira, E., Rodrigues, R. & Blazutti, D., 2003, Caracterização geoquímica da superfície de inundação llandoveriana na parte paraguaia da Bacia do Paraná: Geochimica Brasiliensis, 17 (1): 1-12.

Posamentier, H.W. & Allen, G.P. 1999. Siliciclastic sequence stratigraphy – Concepts and applications. SEPM Concepts in Sedimentology and Paleontology, 7, 210p.

Ramos, A.N. 1970. Aspectos paleo-estruturais da Bacia do Paraná e sua influência na sedimentação. Boletim de Geociências da Petrobrás, 13: 85-93.

Rodrigues, R. 2005. Chemostratigraphy. In: E. A. M. Koutsoukos (ed), Applied Stratigraphy, Springer: 165-178.

Rodrigues, R., Pereira, E. & Bergamaschi, S. 2005. Organic geochemical characterization of Frasnian petroleum source rocks of Brazilian Paleozoic basins. In: Gondwana 12 Meeting. Cdroon.

Sutton, S.J.; Ethridge, F.G.; Almon, W.R.; Dawson, W.C. & Edwards, K.K. 2004. Textural and sequence-stratigraphy controls on sealing capacity of Lower and Upper Cretaceous shales, Denver basin, Colorado. AAPG Bulletin, 88: 1185-1206.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 121 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Paleontological characterization of Supersequences in the Paraná Basin

Luiz Padilha de Quadros Petróleo Brasileiro S.A. – Petrobrás, Rio de Janeiro – RJ – Brazil, [email protected]

Abstract PETROBRAS (Petróleo Brasileiro S.A.) has conducted several projects in the Paraná Basin region aimed at defining precisely the stratigraphic occurrence of commercial hydrocarbons in the Paleozoic sequences. Based on radiometric data, Chitinozoa, marine microphytoplankton (Acritarcha), megafossils, spores and pollen, it proved possible to define biostratigraphically the paleoenvironment and age of these various sequences. The relationship between age, environment and lithostratigraphic data facilitates interpretive reconstruction of the evolving Paleozoic sediments in the Paraná Basin.

Introduction According to Milani (1997), the Paraná Basin comprises six, major, second-order allostratigraphic units or supersequences: Rio Ivaí (Caradoc-Llandovery), Paraná (Lochkovian-Frasnian), Gondwana I (Westphalian-Scythian), Gondwana II (Anisian-Norian), Gondwana III (Upper Jurassic-Berriasian), and Bauru (Aptian-Maastrichtian). The first three of these correspond to Paleozoic transgressive-regressive cycles, and the others are Mesozoic continental sedimentary packages with associated igneous rocks.

Rio Ivaí Supersequence The Rio Ivaí Supersequence begins with a basal sandy unit (Alto Garças Formation; unfossiliferous) up to 300 meters thick, followed by some tens of meters of diamictites (Iapó Formation; unfossiliferous) that record the Late Ordovician-Early Silurian glaciation of Gondwana. The uppermost unit of the supersequence, the Vila Maria Formation, consists of micaceous shales and fine-grained sandstones. The available phytoplankton evidence suggests a Silurian age for the Vila Maria Formation and is consistent with the early Llandoverian age indicated by the spore tetrads (Gray et al., 1985). Radiometric Rb-Sr isochron age also indicates Llandovery as the main Vila Maria depositional interval (Mizusaki et al., 2002). The chitinozoan faunas in the shales of the Vila Maria Formation and its

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 122 equivalents have species in common with the middle to upper LIandovery Vargas Peña Formation in Paraguay (Grahn et al., 2000). It is suggested that the Vila Maria Formation and its equivalents are of Llandovery age and coeval with the Itacurubí Group in eastern Paraguay (Grahn et al., 2000). Graptolites are common fossils and have a worldwide distribution. They are important index fossils for dating Palaeozoic rocks as they evolved rapidly through time, producing many different, short-ranged species. For instance, British geologists have divided their Ordovician and Silurian successions into numerous graptolite biozones; these are generally less than one million years in duration. According to Melo (1993), the Vargas Peña Shale records the maximum deepening of the Early Silurian sea in eastern Paraguay, and preserves the richest and most diverse biotas of the Silurian in the Paraná Basin. Several species do persist into the overlying regressive unit (Cariy Sandstone). The main components of the Silurian fauna are: trilobites (Calymene, Diacalymene, Makasspis, Guayakinites, Guaranites and Eohomalonotus); Malvinokaffric brachiopods (Anabaia paraguayensis); Mollusca, and some Conulariida. The fossil content of the Vila Maria Formation is poor in comparison with that of the correlative units in eastern Paraguay, probably reflecting littoral environments in the east and northeast of the basin. In effect, the few megafossil associations southwest of Goiás are typical of littoral environments, probably lagoonal with Plectonotus derbyi, Orbiculoidea sp., bivalves, etc., or indicative of high energy (sandstones with Arthrophycus harlani). The most important occurrences of graptolites are in eastern Paraguay (mainly from the Vargas Peña Shale). Due to the bad preservation of fossils: the Vargas Peña Shale is dated as Llandovery (Aeronian) on the basis of Monograptus lobiferus and M. aff. sedgwicki or more precisely as early Llandovery (Rhuddanian) from the presence of Diplograptus modestus.

Paraná Supersequence The Devonian package of the Paraná Basin (Paraná Supersequence) is composed of a blanket of coarse-grained, kaolinite-rich sandstones (Furnas Formation) succeeded by a shaly section with subordinate siltstones and deltaic sandstones (Ponta Grossa Formation). Based on paleogeographic considerations, the age of the lower Furnas Formation is possibly Lochkovian (Grahn et al., 2000).The upper Furnas Formation is possibly datable within the earliest Pragian, as suggested by Dino and Rodrigues (1990) and Loboziak et aI. (1995). In the "East Paraguay Sub-basin," the base of the Santa Elena Formation is probably contemporaneous with the base of the Furnas Formation and its equivalents. In its type section, the Ponta Grossa Formation (Jaguariaíva Member of Lange and

Petri, 1967) is of Pragian age. This part corresponds to unit D2a of Lange (1967). Dark shales in the Alto Garças Sub-basin are of same age. Late Emsian siltstones apparently are missing in the north-northwestem part of that sub-basin; and a gap in the sequence,

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 123 corresponding at least to the earIy Emsian, is identifiable in the Alto Garças Sub-basin (Grahn et al., 2000). The Devonian shales, particularly their lowermost section of Emsian- Eifelian age, host the endemic Malvinokaffric fauna of invertebrates. The Pragian age of the upper part of Santa Elena Formation implies correlation with the Ponta Grossa Formation s.s. and its equivalents. The upper Ponta Grossa Formation s.s.(= Unit D2b of Lange 1967) is dated as late Emsian. Rocks of this age remain unrecorded in eastern Paraguay. The tempestitic sandstones of the Tibaji Member of the Ponta Grossa Formation s.s. in the Apucarana Sub-Basin are also late Emsian in age, and have no known paleontologically dated equivalents in the Alto Garças Sub-Basin. The Tibaji Member of

Lange (his unit D3) probably corresponds to a predominantly sandy interval within the lower part of the São Domingos Formation and its equivalents. ln the Apucarana Sub-basin, the lowermost part of the São Domingos Formation is still considered late Emsian (Grahn et al., 2000). The Ponta Grossa Formation is characterized by high diversity of invertebrate fossils, including Conulariida, Brachiopoda (Lingula, Orbiculoidea, Australocoelia, Australospirifer), Ostracoda, Bivalvia (Modiolopsis, Nuculites, Palaeoneilo.), Gastropoda, Tentaculitoidea (Tentaculites), Trilobita (Calmonia, Tibagya, Burmeisteria, Paracalmonia, Pennaia, Metacryphaeus) and Crinoidea, together with abundant ichnofossils of the Zoophycus ichnofacies. The invertebrates are of particular paleobiogeographic importance principally because they include representative elements of the Malvinokaffric fauna (Bolzon et al., 1999). Plant remains such as Spongyophyton are also common in shales of the Ponta Grossa Formation.

Gondwana I Supersequence As originally defined (Milani, 1997), the Gondwana I Supersequence, among all second-order allostratigraphic units, includes the largest sedimentary volume of the Paraná Basin, with a thickness attaining a maximum of about 2,500 meters. The Gondwana I Supersequence consists of the Itararé, Guatá and Passa Dois Groups and it constitutes the thickest and stratigraphically most complete Gondwanan record of the Late Carboniferous through Late Permian. The biostratigraphy of the Upper Palaeozoic strata of the Brazilian Paraná Basin has been studied by several authors, who proposed different zonations based on plant megafossils (Glossopteris, Noeggerathiopsis, Sphenophyllum, Gangamopteris, Arberia, Asterotheca, Phyllotheca, Botrychiopsis, Nothorhacopteris, and Paranoclodus; Rösler, 1978, Rohn, 1994, 1997, Iannuzzi, 2000); invertebrates (Pinzonella, Australomya, Jacquesia, Heteropecten, and Schizodus); and palynomorphs (Potonieisporites, Lundbladispora, Vittatina, Weylandites, and Cannanoropollis). Palynology appears to be the most efficient

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 124 tool in providing biostratigraphic data for the Paraná Basin, because of the abundance, diversity and widespread distribution of spore-pollen assemblages (Souza and Marques- Toigo, 2003). The depositional environment of the Itararé Group (Gondwana I Supersequence) is an interesting subject of discussion because some lithologies previously interpreted as varvites are now considered as probable turbidites. The Itararé Group was possibly connected to an open seaway to the south (Eyles, Eyles and França, 1993). The occurrence of Tasmanites in the Campo Mourão Formation indicates a marine influence during its deposition. Other Carboniferous-Permian marine microphytoplankton, such as Deusilites, Micrhystridium and Dictyotidium, have been encountered in Supersequence I (Quadros, 2002). According to Daemon and Quadros (1970), the saccate pollen Potonieisporites trigonalis (P-490) and Cannanoropollis triangularis (P-501) are restricted to the G palynostratigraphic interval. These two species have been recorded in the Upper Itararé

Subgroup in Rio Grande do Sul State (Dias, 1993) from strata assignable to the H3-I interval. Additional information on the spore-pollen content of these intervals was given by Daemon and Quadros (1970) and Daemon (1981). Late Carboniferous palynofloras have been recognized in the Lower and Middle Itararé Subgroup in São Paulo and Paraná States, in the northeastern sector of the Paraná Basin. The Carboniferous ages assigned to these palynozones are based on the presence of diagnostic species, and on correlation between the main Gondwanan palynozones, especially from the Argentinian basins (e.g., Paganzo, San Rafael, Tarija and Chaco-Paraná basins), where the Carboniferous successions are more complete. Radiometric data are scarce in the Paraná Basin, and absent in its Carboniferous sequence, thus limiting accurate age calibration of the existing palynostratigraphic schemes (Souza and Marques-Toigo, 2003). According to Souza and Marques-Toigo (2000), the Vittatina costabilis Interval Zone includes the Upper ltararé Subgroup and part of the Rio Bonito Formation. No significant biostratigraphic differences have been recorded in these sections, despite lithological variations in the basin. The boundary between the Vittatina costabilis and Lueckisporites virkkiae Interval Zones is recorded in the Upper Rio Bonito Formation and the Lower Palermo Formation, and is related to the J/K intervals (Daemon and Quadros, 1970). The lrati Formation is part of the Passa Dois Group, which represents the regressive phase within the major Upper Paleozoic transgressive-regressive cycle of the Gondwana I Supersequence (Milani and Zalán, 1999). This formation contains a rich and diversified fossil biota, including plant megafossils, vertebrates, invertebrates (e.g., insects, crustaceans, foraminifera), palynomorphs (mainly pollen grains and spores) and ichnofossils (e.g., Mussa et al., 1980; Oelofsen and Araújo, 1983; Pinto and Adami-Rodrigues, 1996). Among these fossils, mesosaurs (Progranosauria) are the most common (genera Mesosaurus, Stereosternum and Brazilosaurus) and have been largely used for stratigraphic correlations

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 125 (e.g., Oelofsen and Araújo, 1983.) They occur in sediments of the Paraná Basin, Brazil, as well as in the Karoo System, in Africa. However, the paleoenvironmental reconstruction of the lrati Formation and its depositional setting are still controversial, being regarded either as a succession deposited in an epicontinental sea or as a continental sequence formed in a lagoon or gulf with varied salinity (Araújo, 2001). The palynomorphs from this leveI include key species of the Lueckisporites virkkiae Zone (such as L.virkkiae, L. densicorpus, L. stenotaeniatus, Weylandites lucifer, and Alisporites splendens). Typical species of this zone also occur in African and other South American (Argentina) Permian strata, thus enabling stratigraphic correlations between these regions. Based on pollen grains and their distribution in Euroamerican palynozones, Daemon and Quadros (1970) proposed a Kazanian age for this unit. However, the general lack of radiometric data and of cIear reference-Ievels with chronologically significant value impedes large-scale correlation and accurate age calibration among the available biostratigraphic schemes (Santos et al., 2006). According to Santos et al. (2006), the Irati Formation is one of the most important oil sources within the Paraná Basin; it consists mainly of siltstones, gray claystones, as well as organic-rich claystones intercalated with limestones. Zircon morphology based on cathodolumnescence images serves to characterize two distinct populations: a dominant category of euhedral, prismatic, elongate to acicular grains that are most likely related to ash- fall volcanism; and a group of rounded to large prismatic grains interpreted as detrital. SHRIMP (Sensitive High Resolution Ion Microprobe) analysis performed on the euhedral and prismatic grains revealed an age of ca. 278.4 + 2.2 Ma (7 points with 95% confidence) interpreted as the crystallization age of the volcanic eruption. Based on this new dating, the Irati Formation is assignable to the Early Permian (Cisuralian: specifically Artinskian); this represents a substantial modification of ages traditionally attributed to this unit. The other formations of the Passa Dois Group (Serra Alta, Teresina, Corumbataí and Rio do Rasto) are dated as Late Permian (Daemon and Quadros, 1970).

Gondwana II Supersequence Fluvial and lacustrine red beds of local occurrence constitute the Gondwana II Supersequence of the Paraná Basin. These contain abundant Middle to Late Triassic tetrapods (Thrinaxodon, Chiniquodon, Rhynchosauria, Dinodontosaurus, Prestosuchus, and Hyperodapedon).

Gondwana III Supersequence Milani (1997) inferred that huge seas and eolian dunes were responsible for deposition of the regionally widespread Botucatu Formation during Jurassic times in the

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 126 Paraná Basin. This scenario was followed in the Early Cretaceous by the predominantly basaltic magmatism (with some acid and intermediate contributions) that resulted in the Serra Geral Formation. These formations jointly compose the Gondwana III Supersequence. Brasilichnium elusivum tracks related to primitive mammals occur in the Botucatu Formation (Leonardi and Carvalho,2002).

Bauru Supersequence Finally, the Upper Cretaceous Bauru Supersequence (Milani, 1997) is a package of alluvial, fluvial and eolian sedimentary rocks that terminated the depositional history of the Paraná Basin. This post-basaltic section has a Senonian fossil content that includes genera representative of the Chelonia, Crocodilia and Dinosauria.

References Araújo, L.M., 2001, Análise da expressão estratigráfica dos parâmetros de geoquímica orgânica e inorgânica nas seqüências deposicionais Irati. PhD Thesis, Instituto de Geociências, Universidade Federal do Rio Grande do Sul, 2 v., 307 p.

Araújo, D.C. and Gonzaga T.D., 1980, Uma nova-espécie de Jachaleria (Therapsida, Dicynodontia) do Triássico do Brasil. In: Congreso Argentino de Paleontologia y Bioestratigrafia, 2, and Congreso Latinoamericano de Paleontologia, :1, Buenos Aires, Àctas,1, p.159-174.

Barberena, M.C.; Holz. M.; Schultz, C.L.; Scherer, C.M.S. 2002,Tetrápodes Triássicos do Rio Grande do Sul. Sítios geológicos e paleontológicos do Brasil. Instituto de Geociências, Universidade Federal do Rio Grande do Sul.SIGEP, Brasília, p. 11-22.

Daemon, R.F., 1981, Controle litobioestratigráfico preliminar do Devoniano, Carbonifero Superior e Permiano da bacia sedimentar do Paraná. 3 Simp. Reg. Geol. São Paulo, p. 124-132.

Daemon, R.F., Quadros, L.P., 1970, Bioestratigrafia do Neopaleozóico da Bacia.do Paraná. Anais 24° Congresso Brasileiro de Geologia. Brasília, p. 359-412.

Dino, R. and Rodrigues, M. A. C., 1995, Palinomorfos Eodevonianos da Formação Furnas - Bacia do Paraná. An. Acad. bras. Ci. 67 (1), p. 107-116.

Eyles, C.H., Eyles, N., França, A.B., 1993, Glaciation and tectonics in an active intracratonic basin: the Paleozoic ltararé Group, Paraná Basin, Brazil: Sedimentology, v.40, p.1-25.

Grahn, Y., Pereira, E. and Bergamaschi, S., 2000, Silurian and Lower Devonian chitinozoan biostratigraphy of the Paraná Basin in Brazil and Paraguay. Palynology v.24, p.143-172.

Gray, J.; Colbath, G. K.; Faria. A.; Boucot, A .J. and Rohr, D. M., 1985, Silurian-age fossils; from the Paleozoic Paraná Basin. southern Brazil Geology, v.13, p. 521-525.

Iannuzzi, R., 2000, Presença do gênero Stephanophyllites em estratos Eopermianos do Rio Grande do Sul, Brasil (Formação Rio bonito, Bacia do Paraná). Revista da Universidade de Guarulhos (Guarulhos) V, p.74- 77.

Lange, F.W, 1967, Subdivisão bioestratigráfica e revisão da coluna siluro devoniana da Bacia do Baixo Amazonas. Atas do Simpósio sobre a Biota Amazônica (Geociências), 1:p. 215-326.

Lange, F.W. and Petri, S. 1967, . The Devonian of the Paraná Basin. Boletim Paranaense de Geociências v.21/22, p. 5-55.

Leonardi,G.; Carvalho,I.S.,2002, Jazigo Icnofossilífero do Ouro (Araraquara), SP - Ricas pistas de tetrápodes do Jurássico. In: Schobbenhaus,C.; Campos,D.A. ; Queiroz,E.T.; Winge,M.; Berbert-Born,M.L.C. (Edits.) Sítios Geológicos e Paleontológicos do Brasil. 1. ed. Brasilia: DNPM/CPRM - Comissão Brasileira de Sítios Geológicos e Paleobiológicos (SIGEP), v. 1, p. 39-48

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 127 Loboziak, S., Melo, J.H.G., Steemans, P., and Barrilari, I.M.R.,1995, Miospore evidence for pre-Emsian and latest Famennian sedimentation in the Devonian of the Paraná Basin, south Brazil. Anais da Academia Brasileira de Ciências, v. 67, p. 391-392.

MeIo, J.H.G., 1988, The Malvinokaffric realm in the Devonian of Brazil. In: N.J. McMillan, A.F. Embry and D.J. Glass (eds.), Devonian of the World. Canadian Society of Petroleum Geologists Memoir, V. 1, n.14, p. 669- 703.

MeIo,, J.H.G., 1993, A paleontologia do Siluriano da Bacia do Paraná: estado da arte. In: SIMPÓSIO SOBRE CRONOESTRATIGRAFIA DA BACIA DO PARANÁ, 1, Rio Claro. Resumos... Rio Claro, UNESP/SBG, p. 6- 7. (Expanded abstract.)

Milani, E.J.and Zalán, P.V., 1999, An outline of the; geology and petroleum systems of the Paleozoic interior basins of South Ameríca. Episodes, v. 22, p.199-205.

Mizusaki, A. M. P.; Melo, J. H. G.; Vignol-Lelarge, M. L. and Steemans, P. 2002. Vila Maria Formation (Silurian, Paraná Basin, Brazil): integrated radiometric and palynological age determinations. Geol. Mag. 139 (4), p. 453-463.

Mussa, D., R. G. Carvalho and P. R. Santos. 1980. Estudo estratigráfico em ocorrências fossilíferas da Formação Irati de São Paulo, Brasil. Boi. IG-USP v. 11, p. 142-149.

Oelofsen, B. and D. C. Araújo, 1983, Paleoecological implications of the distribution of Mesosaurid reptiles in the Permian Irati sea (Paraná Basin), South America. Rev. Bras. Geoc. V.13, p. 1-6.

Pinto, I. D and K. Adami-Rodrigues, 1996, Pygocephalomorph crustácea. New data and interpretations, with emphasis on Brasilian and South African forms. Pesquisas v. 23, p. 41-50.

Quadros, L.P.,2002, Acritarcos e Tasmanites do Permo-Carbonífero da Bacia do Paraná. Revista do Instituto Geológico, São Paulo, 23(1), p; 39-50.

Rohn, R.,1997, First glossopterids of the Rio do Rasto Formation in the São Paulo State (Upper Permian, Paraná Basin), Brazil.. Revista da Universidade de Guarulhos, Guarulhos, v. 2, n. n° especial, p. 76-84.

Rösler, O. 1978. The Brazilian Eogondwanic floral succession. BoI.G-USP, v.9, p. 85-91

Santos, R. V., P. A. Souza, C. J .S. Alvarenga, E. L. Dantas, M. M. Pimentel, C. G. Oliveira and L. M. Araújo. 2006. Shrimp U-Pb Zircon and palinology of bentonitic layers from the Permian Irati Formation, Paraná Basin, Brazil. Gondwana Research , v.9, p. 456-463.

Souza, P.A and Marques-Toigo, M., 2000, Zona Vittatina: marco palinobioestratigráfico do Permiano Inferior da Bacia do Paraná. Ciência- Técnica-Petróleo, v. 20, p. 153-159.

Souza, P.A and Marques-Toigo, M., 2003, An overview on the palynostratigraphy of the Upper Paleozoic Brazilian Paraná Basin. Rev. Mus. Argent. Cienc Nat. v. 5, p. 205-214.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 128 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Early correlations between Cape Town and Ventania Systems: Keidel’s pioneer work and his influence on Wegener’s Continental drift and Du Toit’s ideas

Victor A. Ramos Laboratorio de Tectónica Andina, Universidad de Buenos Aires, [email protected]

Although different authors at different times advanced some ideas on continental drift (Bacon, 1620; Schneider, 1858; Taylor, 1910, etc), there was no doubt that, behind that there is something more than a fortuitous parallelism of the coasts on either side of the Atlantic. There is a robust consensus that Alfred Wegener (1912, 1915, 1920, 1922, 1929) is the father of the theory, because he made the first comprehensive multidisciplinary approach to postulate the displacement of the continents, and worked hard his entire life until his last days in trying to demonstrate his assertions. However, it is little known that when he went through the geological arguments to demonstrate the first correlations between South Africa and South America, the best proofs he had were the vigorous correlations developed by Keidel (1913, 1916). His chapter on geological correlations starts with the comparison between both sides of the South Atlantic using the recent data presented in the Compte Rendu of the XII° International Geological Congress of Toronto, Canada, by Keidel (1913). Wegener in his book textually stated following Keidel: “In the Sierras of Buenos Aires Province, particularly in the southern range, we find a succession of beds very like that of the Cape mountains of South Africa”. “There appears to be strong conformity among at least three cases: the lower sandstone of the Lower Devonian transgression, the fossil-bearing schists which mark the culmination of this transgression and a more recent and very characteristic structure, the glacial conglomerate of the Upper Paleozoic…both the sedimentary rocks of the Devonian transgression and the glacial conglomerate are strongly folded just as in the Cape

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 129 mountains; and here, as there, the direction of the folding movement is mainly a northerly one”. Keidel (1913) in his presentation described the structure of Sierras de la Ventana, their northern vergence, and particularly the glacial conglomerates (Figures 1 and 2). It is interesting to learn how much interest he put in these conglomerates, known since the early work of Hauthal (1892). Halle (1911) described from Hill Cove in the Malvinas Islands the late Paleozoic glacial deposits and correlated them with the Dwyka tillite of Cape Town. Keidel found a great similarity between the Malvinas tillite description and the Sauce Grande conglomerates, and therefore planned a visit to the Sierra de Pillahuincó in April, 1912. In that visit he found compelling evidence of the glacial origin of the conglomerates, where pentagonal shape clasts plenty of striae were common (Keidel, 1916). He was able to correlate the Sauce Grande conglomerates with the Dwyka tillite, based also in the description of Rogers (1905) and Stutzer (1911) of the South African counterparts.

Figure 1 - Striated clasts of Sauce Grande described by Keidel (1913).

Figure 2 - Cross section of the Sierra de Pillahuincó in the southern province of Buenos Aires, showing the northern vergence of the late Paleozoic deformation after Keidel (1913).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 130 On the other hand Du Toit in 1916, without knowing Wegener’s “displacement theory” was trying to demonstrate the continuity of the glaciation in the late Paleozoic among South Africa, Brazil, Uruguay, and Argentina. After his seminal paper on “The Carboniferous glaciation of South Africa” Du Toit (1921) wanted to compare these deposits with the South American counterparts. He received a grant from the Carnegie Institution of Washington and came in 1923 for five months to South America, spending two months visiting the Ventania system and Precordillera. This trip was the basis for his “A geological comparison of South America with South Africa” where for the first time a single geologist had the opportunity to visit the exposures of the same units at both sides of the Atlantic (Du Toit, 1927). During that visit he was acquainted with local geological work on the late Paleozoic from his colleagues of Brazil, Uruguay and Argentina. His previous claim that “papers on English language on the geology of any part of South America are relatively few” that avoid him a previous knowledge of the geology was then overcome.

Figura 3 - A. Du Toit with J.J. Nágera in San Juan (after Nágera, 1939).

In that visit, accompanied by Juan J. Nágera, Franco Pastore and Augusto Tapia, Du Toit learnt about the Gondwanides: a term coined by Keidel (1921) to describe the important mountain system uplifted by the late Paleozoic deformation in several parts of central and southern Argentina. Du Toit spent most of his academic life producing important contributions to the continental drift theory, mainly after the decease of Wegener with his master work on “Our wandering continents” that had an important influence in the English speaking community

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 131 (Du Toit, 1937). On the other hand, Professor Keidel taught during many decades to several generations of students in geology of the Buenos Aires University the hypothesis of continental drift and the facts that supported it, in spite of the discredit that the theory had in the northern hemisphere. Among his students outstands Dr. Horacio J. Harrington that continued his lectures and research on continental drift. As a final comment it is necessary to emphasize that prior to 1916, neither Keidel nor Du Toit knew about Wegener, and neither of them were supporters of the continental drift theory, but soon after the AAPG Symposium, both became ones of the most active supporters.

References Bacon, F., 1620. Novum Organum. Trnasl. A. Banfi (1943), Serie Collana di Testi Filosofici, Verona, 143 pp.

Du Toit, A.L., 1921. The Carboniferous glaciation of South Africa. Transactions of the Geological Society of South Africa, v. 24, p. 188-227.

Du Toit, A.L., 1927. A geological comparison of South America with South Africa. Publications Carnegie Institute v. 381, p. 1-157.

Du Toit, A.L., 1937. Our wandering continents. London, Oliver and Boyd, 366 p.

Halle, T.G., 1911. On the geological structure and history of the Falkland Islands. Bulletin of the Geological Institute of Uppsala, v. 11, p.626-629.

Hauthal, R., 1892. La Sierra de la Ventana (Provincia de Buenos Aires). Revista del Museo de La Plata v. 3, p. 3- 11.

Keidel, J., 1913. Über das Alter, die Verbreitung und die gegenseitigen Beziehungen der verschiedenen tektonischen strukturen in den argentinischen Gebirgen. XII° Session du Congrés Géologique International, Compte Rendu p. 671-687, Toronto.

Keidel, J., 1916. La geología de las Sierras de la Provincia de Buenos Aires y sus relaciones con las montañas de Sudáfrica y Los Andes. Ministerio de Agricultura de La Nación, Sección Geología, Mineralogía y Minería, Anales v. XI(3), p. 1-78.

Keidel, J., 1921. Sobre la distribución de los depósitos glaciares del Pérmico conocidos en la Argentina y su significa¬ción para la estratigrafía de la serie del Gondwana y la paleogeografía del Hemisferio Austral. Academia Nacional de Ciencias, Boletín 25: 239 368.

Nágera, J.J., 1939. Geografía Física de la República Argentina. En O. Manito y J.J. Nágera, Geografía Física de las Américas y de la República Argentina. Editorial Kapeluz, 232 p., Buenos Aires.

Rogers, A.W., 1905. The glacial conglomerate in the Table Mountain Series near Clanwilliam. Transactions Southafrican Philosophical Society, v. 16, p. 1-8.

Snider-Pellegrini,A., 1858. La Creation et ses mystéres dévoilés. Franvk & Dentu, Paris.

Stutzer, O., 1911. Ueber Dwykakonglomerat im Lande Katanga. Zeitrschift der Deutsche geologische Gessellschaft, v. 43, p. 626-629.

Taylor, F.B., 1910. Bearing of the Tertiary mountain belt on the origin of the Earth’s plan. Bulletin of the Geological Society of America, v. 21(2), p. 179-226.

Wegener, A., 1912. Die Entstehung der Kontinente. Petermanns Geographische Mitteilungen, 58 I, 185–195, 253–256, 305–309. Transl. from German by W.R. Jacoby (2001), Journal of Geodynamics, v. 32 , p. 29–63.

Wegener, A., 1915. Die Entstehung der Kontinente und Ozeane. Sammlung Vieweg v. 23, 94 pp.; 2nd. Ed. (1920) Die Wissenschaft v. 66, 135 pp.; 3rd. Edition (1922)

Wegener, A. , 1929. The origin of continents and oceans. Dover Publication, (Translation of the 4th. Ed.) 231 pp., New York.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 132 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The Ventania System: Tectonic constraints in its Paleozoic evolution and the Patagonia accretion

Victor A. Ramos Laboratorio de Tectónica Andina, Universidad de Buenos Aires, [email protected]

The Ventania system was a keystone in the early interpretations of continental drift theory due to its complete Paleozoic stratigraphy correlated with the Cape Town system, its structural trend perpendicular to the continental margin, and its northern vergence. These facts were taken in consideration by Wegener (1915 and subsequent papers), together with the presence of glacial deposits both in Ventania and Cape Town (Keidel, 1913 and Du Toit, 1921). The fact that it was perpendicular to the margin, led Harrington (1970) to interpret this mountain system as an aulacogen. However, in the early years of the plate tectonics theory, the Cape Town system called the attention of Dewey and Bird (1970), due to its location along the southern continental margin. Some years later, after the pioneer work of Monger et al. (1982) on suspect terranes, new interpretations arose in order to explain the northern vergence and the associated dynamic metamorphism (Ramos, 1984, 1986; Winter, 1984, 1986).

The Ventania System The structural style of the early and late Paleozoic sequences was controversial. For many authors, it was a typical example of pure folding with only minor faults (Harrington, 1947, 1970, Amos, 1995). Some alternative hypothesis has been advanced by Cobbold et al. (1991) who interpreted the strain as result of a transpressive regime. Others postulate the existence of an important rotational deformation with first and second order shear and thrust zones, which repeated the sequence (von Gosen et al., 1990; Tomezzoli and Cristallini, 1998). Most of the authors agree in the significant strain of the Paleozoic rocks reaching up to 20-24% (see Japas, 1989). The fold and thrust belt has imbrications of the basement

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 133 towards the southwest, which is preserved in anchizonal to greenschist facies metamorphism (von Gosen et al., 1990, 1991). New ages on the basement confirm its Early Cambrian age and the Early to Middle Permian deformation of the rocks (Rapela and Kostadinoff, 2005). The age of deformation is constrained by paleomagnetic data (Tomezzoli, 1999, 2001; Tomezzoli and Vilas, 1999; Tomezzoli and Cristallini, 1998) and by growth strata in the Las Tunas Formation of Early Permian age (López Gamundi et al., 1995).

Figure 1 - The Claromecó Basin, the foredeep of the Ventania fold and thrust belt based on gravimetric data of Ramos and Kostadinoff (2005).

Claromecó Basin A foreland basin developed as result of the intense Permian deformation, reaching a depth of more than 10,500 meters (Figure 1)(Ramos and Kostadinoff, 2005). That thickness is formed by imbrication in thrust sheets of the early Paleozoic rocks, plus the gentle folding of the molasse deposits of late Päleozoic age. New seismic data and oil exploration drilling have confirmed an extension over 40,000 km2 and the composition of the basin. There is an angular unconformity between the early Paleozoic rocks and the Carboniferous-Permian sequences, already known from outcrops (Massabie and Rossello, 1984).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 134 Patagonian Massif To understand the geologic evolution and the origin of the deformation in the Ventania system it is required to discuss the geologic constitution of the Somun Cura (or northern Patagonian) Massif (Ramos, 2004). This region conserved the relics of a magmatic arc developed during Late Paleozoic times, presently preserved as orthogneisses and micaschists with Carboniferous and Early Permian ages (see review in Pankhurst et al., 2006). These magmatic rocks have a northwestern trend parallel to the Ventania system. Several authors described the important shear zones affecting the metamorphic rocks of the Somun Cura Massif. Recent detailed structural and kinematic analyses of the ductile deformation showed important thrust indicating top to the north and northeast (von Gosen, 2003). These deformed rocks are intruded by Middle to Late Permian granites, which postdate the main deformation.

Tectonic setting The relationships among the Ventania fold and thrust belt, the Claromecó foredeep and the metamorphic and igneous rocks exposed in the Somun Cura Massif have been explained by different hypotheses. The autochthonous hypotheses invoke the lateral accretion of most of the Patagonian rocks during Paleozoic time (Forsythe, 1982; Uliana et al., 1986; Caminos et al., 1988). These hypotheses assume a magmatic arc oblique to the entire Patagonia that crosses the Somun Cura and the Desaedo Massifs as proposed by de Wit (1977), Forsythe (1982) and Uliana et al. (1986). These hypotheses do not explain the almost orthogonal deformation of the Ventania system and the magmatic arc. The interpretation of Lock (1980) of flat subduction, modified by Dalziel et al. (2000) by the interaction of a hot spot and the flat- subduction, does not satisfactory explain the more than 1,500 km between the subduction and the magmatic arc. The allochthonous hypothesis proposed by Ramos (1984, 1986) assumed that Patagonia was a terrane that docked against Gondwana during Early Permian times, in a similar way as proposed by Winter (1984, 1986) in the Cape Town system. This hypothesis has been revisited by Panckhurst et al. (2006), but there is no consensus on the location of the suture between Gondwana and Patagonia. As a concluding remark it should be emphasized that the sedimentary evolution of Ventania; the development of a foreland basin with a deep trough as the Claromecó foredeep and the ductile deformation along both the northern and southern boundaries of Patagonia, require an extraordinary episode of deformation. That episode is best explained by an important collision of part of Patagonia against the old continental margin of Gondwana.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 135 References Amos, A.J., 1995. Charles Darwin y la cadena plegada de Sierra de la Ventana. Revista de la Asociación Geológica Argentina, v. 50(1-4), p. 266-269.

Caminos, R., Llambías, E.J., Rapela, C.W. and Parica, C.A., 1988. Late Paleozoic Early Triassic magmatic activity of Argentina and the significance of new Rb/Sr ages from Northern Patagonia. Journal South American Earth Sciences, v. 1, p. 137-146.

Cobbold, P.R., Gapais, D. and Rossello, E.A., 1991. Partitioning of transpressive motions within a sigmoidal foldbelt: the Variscan Sierras Australes, Argentina, Journal of Structural Geology, v. 13 (7), p. 743-758.

Dalziel, I.W.D., Lawver, L.A. and Murphy, J.B., 2000. Plumes, orogenesis, and supercontinental fragmentation. Earth and Planetary Science Letters, v. 178, p. 1-11.

Dewey, J.F. and J. Bird, 1970. Mountain belts and the new global tectonics, Journal Geophysical Research, v. 75, p. 2625-2647.

Du Toit, A.L., 1921. The Carboniferous glaciation of South Africa. Transactions of the Geological Society of South Africa, v. 24, p. 188-227.

Forsythe, R., 1982. The Late Paleozoic to Early Mesozoic evolution of Southern South America: a plate tectonic interpretation. Journal Geological Society, v. 139, p. 671-682.

Harrington, H.J., 1947. Explicación de las Hojas Geológicas 33m y 34m, Sierras de Curamalal y de la Ventana, Provincia de Buenos Aires. Servicio Nacional de Minería y Geología, Boletín, v. 61, p. 1-43.

Harrington, H.J, 1970. Las Sierras Australes de Buenos Aires, República Argentina: Cadena Aulacogénica. Revista de la Asociación Geológica Argentina, v. 25(2), p. 151-181.

Japas, M.S., 1989. La deformación de la cadena plegada de las Sierras Australes de la provincia de Buenos Aires. Anales de la Academia Nacional de Ciencias Exactas Físicas y Naturales, v. 40, p. 193-215.

Lock, B.E., 1980. Flat plate subduction of the Cape Fold Belt of South Africa. Geology, v. 8, p. 35-39.

López Gamundi, O.R., Conaghan, P.J., Rosello, E.A. and Cobbold, P.R., 1995. The Tunas Formation (Permian) in the Sierras Australes Foldbelt, east central Argentina: evidence for syntectonic sedimentation in a foreland basin. Journal of South American Earth Sciences, v. 8(2), p. 129-142.

Massabie, A. and Rossello, E., 1984. La discordancia pre-Sauce Grande y su entorno estratigráfico. IX° Congreso Geológico Argentino, Actas, v. 1, p. 337-352.

Monger, J.H.W., Price, R.A., Tempelman-Kluit, D.J.,1982. Tectonic accretion and the origin of two major metamorphic and plutonic welts in the Canadian Cordillera. Geology, v. 10, p. 70-75.

Pankhurst, R.J., Rapela, C.W., Fanning, C.M. and Márquez, M., 2006. Gondwanide continental collision and the origin of Patagonia. Earth Science Reviews, v. 76, p. 235-257.

Ramos, V.A., 1984. Patagonia: un continente paleozoico a la deriva?, 9º Congreso Geológico Argentino (S. C. Bariloche), Actas v. 2, p. 311-325.

Ramos, V.A., 1986. Tectonostratigraphy, as applied to analysis of South African Phanerozoic Basins by H. de la R. Winter, discussion. Transactions Geological Society of South Africa, v. 87(2), p. 169-179.

Ramos, V.A., 2004. La Plataforma Patagónica y sus relaciones con la Plataforma Brasilera. In Mantesso-Neto, V., Bartorelli, A., Ré Carneiro, C.D., Brito Neves, B.B. (eds.): Geologia do Continente Sul-Americano, p. 371- 381, Sao Paulo.

Ramos, V.A. and Kostadinoff, J., 2005. La cuenca de Claromecó. In de Barrio, R.E., Etcheverry, R.O., Caballé, M.F., and Llambías E. (eds.): Geología y recursos minerales de la Provincia de Buenos Aires. 16° Congreso Geológico Argentino, Relatorio, p. 473-480.

Rapela, C.W. and Kostadinoff, J., 2005. El basamento de Sierra de la Ventana: historia tectonomagmática. In de Barrio, R.E., Etcheverry, R.O., Caballé, M.F., and Llambías E. (eds.): Geología y recursos minerales de la Provincia de Buenos Aires. 16° Congreso Geológico Argentino, Relatorio, p. 69-84.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 136 Tomezzoli, R.N, 1999. Edad de la sedimentación y deformación de la Formación Tunas en las Sierras Australes de la provincia de Buenos Aires (37º-39ºS - 61º-63ºW). Revista de la Asociación Geológica Argentina, v. 54 (3), p. 220-228.

Tomezzoli, R.N., 2001. Further palaeomagnetic results from the Sierras Australes fold and thrust belt, Argentina. Geophysical Journal International, v. 147, p. 356-366.

Tomezzoli, R.N. and Cristallini, E.O., 1998. Nuevas evidencias sobre la importancia del fallamiento en la estructura de las Sierras Australes de la Provincia de Buenos Aires. Revista de la Asociación Geológica Argentina, v. 53(1), p. 117-129.

Tomezzoli, R.N. and Vilas, J.F., 1999. Paleomagnetic constraints on age of deformation of the Sierras Australes thrust and fold belt, Argentina. Geophysical Journal International, v. 138, p. 857-870.

Uliana, M.A., Biddle, K.T., Phelps, D.W. and Gust, D.A., 1986. Significado del vulcanismo y extensión mesojurásicos en el extremo meridional de Sudamérica. Revista de la Asociación Geológica Argentina, v. 40(3-4), p. 231-253.

Von Gosen, W., 2003. Thrust tectonics in the North Patagonian Massif (Argentina): implication for a Patagonian plate. Tectonics, v. 22(1), p. 1005, doi: 10.1029/2001ITC901039.

Von Gosen, W., Buggisch, W. and Dimieri, L.V., 1990. Structural and metamorphic evolution of the Sierras Australes (Buenos Aires Province / Argentina). Geologische Rundschau, v. 79(3), p. 797-821.

Von Gosen, W., Buggisch, W. and Krumm, S., 1991. Metamorphic and deformation mechanisms in the Sierras Australes fold thrust belt (Buenos Aires, Province, Argentina). Tectonophysics, v. 185, p. 335-356.

Wegener, A., 1915. Die Entstehung der Kontinente und Ozeane. Sammlung Vieweg v. 23, 94 pp.; 2nd. Ed. (1920) Die Wissenschaft v. 66, 135 pp.; 3rd. Edition (1922).

Winter, H. de la R., 1984. Tectonostratigraphy, as applied to analysis of South African Phanerozoic basins. Transactions of Geological Society South Africa, v. 87, p. 169-179.

Winter, H. de la R., 1986. Reply to discussion. Transactions of Geological Society South Africa, v. 88, p. 184-185.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 137 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The Stratigraphic Nature of the Ordovician Graafwater/Peninsula Formational Contact in the Table Mountain Group, Cape Town, South Africa: Diastem or Discordance?

Dr John Rogers and Ms Zandile Dlamini (BSc) Department of Geological Sciences, University of Cape Town, South Africa

Abstract This study examines the contact between the Ordovician Graafwater and Peninsula formations of the Table Mountain Group in Cape Town, South Africa. It aims to determine whether or not there is an angular unconformity between the two formations. Fieldwork was done throughout most of the Cape Peninsula and particular attention was paid to the upper one to two metres of the Graafwater Formation and the basal one to two metres of the overlying Peninsula Formation. The key places visited were Table Mountain, Devil’s Peak, Ou Kaapse Weg, the western slopes of Constantiaberg above Hout Bay, Judas’ Peak at Smitswinkel Bay and Cape Maclear, all within the Table Mountain National Park. The angular unconformity was not found at outcrop scale, but there is strong evidence, in photographs taken from a few hundred metres, that the boundary between the Graafwater and the Peninsula formations is unconformable. Previous work by Rust (1967) recorded a minor angular unconformity in Lekkranskloof, Heerenlogement, northwest of Clanwilliam. Folding of the Graafwater Formation, beneath the relatively undeformed Peninsula Formation, is one of the new findings that have been made in this study. In addition, an excellent new exposure of paired trilobite tracks has been discovered, just below the Graafwater/ Peninsula contact near Platteklip Gorge on the face of Table Mountain itself.

Introduction General In this study, the contact between the Ordovician Graafwater and Peninsula Formations of the Table Mountain Group is examined, to determine whether there is an angular unconformity between the two formations. Geological Setting

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 138 The Graafwater and Peninsula formations are part of the Table Mountain Group (TMG), which is part of the Cape Supergroup. The Graafwater Formation was deposited (520 million years (Ma) ago), unconformably on an erosional surface of the Neoproterozoic Malmesbury Group (560Ma) and the Cambrian Cape Granite (540 Ma) (Armstrong et al., 1998). The Malmesbury Group was folded during the Saldanian orogeny and subsequently intruded in the Cambrian epoch (540 Ma) by the Peninsula Pluton of the Cape Granite Suite. Uplift and peneplanation of the Malmesbury Group and the Cape Granite Suite occurred during regional peneplanation at about 520 Ma. Previous Research Earlier studies of the TMG have been completed by Rust (1967), Tankard and Hobday (1977), Hobday and Tankard (1978) and Turner (1990). There is some controversy relating to the depositional model of the Graafwater and the Peninsula formations. Tankard and Hobday (1977) argue for a tide-dominated back-barrier sedimentation scenario for the Graafwater Formation. They focus on “…complex primary structures (including herringbone cross-stratification and reactivation surfaces), small-scale textural alternations, desiccation cracks, and biogenic features…” (Tankard and Hobday, 1978, p. 1735). An idealized facies sequence illustration is presented from their pooled data and appears in Tankard and Hobday (1977). In a later paper, Hobday and Tankard (1978) attribute the deposition of the Peninsula Formation to a transgressive-barrier and shallow-shelf environment. An opposing model that has been proposed by Turner (1990). He concludes that the Graafwater and the Peninsula formations were deposited in a major braided-plain system that prograded (moved seawards) into a marine depository (Turner, 1990). Aim of the study The aim of this project is to determine whether there is an angular unconformity between the Graafwater and Peninsula Formations. The only reference in the local geological literature that has suggested an angular unconformity between the Graafwater and Peninsula Formations in the Western Cape is that by Rust (1967). This was observed at Lekkranskloof, Heerenlogement, between Graafwater and Klawer, NW of Clanwilliam. It appears in Plate 2F of Rust (1967) (Figure 1). Scope and limitations Stratigraphically, the scope of this study is from the upper one to two metres of the Graafwater Formation across the boundary into the basal metre or two of the overlying Peninsula Formation. Geographically, Table Mountain, Devil’s Peak, Ou Kaapse Weg (OKW) the western slopes of Constantiaberg (CB) above Hout Bay and Cape Maclear (CMC) beside the Cape of Good Hope, were visited during this study (Figure 2).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 139

Figure 1 - Low-angle minor unconformity between the Graafwater and Peninsula Formations at Lekkranskloof, Heerenlogement, NW of Clanwilliam (Rust, 1967, Plate 2F).

The Graafwater Formation/Peninsula Formation Contact Stratigraphic columns Stratigraphic columns were compiled from the measurements gathered from the six measured sections namely Cape Maclear (CM), Ou Kaapse Weg (OKW), Constantiaberg (CB), Upper Waterfall (UWF), Silver Stream Cave (SSC) and Platteklip Gorge West (PKGW), the latter three being along Tafelberg Road (Figure 3).

Cape Maclear (CM) The measurements for this stratigraphic column were collected on the east face of Cape Maclear, west of Dias Beach (Figures 2, 3 and 4) ) (GPS Stop 1: 34o 21’ 20.7” S, 18o 28’ 46.6”, 6 m elevation). At this locality, the Graafwater Formation sandstone has a yellow-brown colour and there are minor maroon mudstone beds that are about 6 m below the contact (Figures 6 and 7). The basal sandstone bed of the Peninsula Formation was greater than 100cm thick and therefore is described as a very thick bed, according to McKee and Weir (1953). The contact is defined by an undulating erosional surface (Figure 7) with a pebble-sandstone conglomerate at the base of the Peninsula Formation (Figure 8). It was found that the sandstone beds of the Peninsula Formation were harder than those of the Graafwater Formation and more orange on their weathered surfaces.

Ou Kaapse Weg (OKW) The stratigraphic column for Ou Kaapse Weg (GPS Stop 2: 34o 5’ 20.1” S, 18o 25’ 52” E, 73 m elevation) (Figures 2, 5 and 7) was found to be similar to that at Cape Maclear, in that the

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 140 top part of the Graafwater Formation is dominated by sandstones with mudstones lower down in the formation. The sandstones of the Graafwater Formation ranged from crossbedded to horizontally laminated to homogeneous. Some had ripple-marked upper bounding planes and some had erosional lower bounding planes. The Peninsula Formation is defined by thick (30-100cm) and very thick beds (>100cm) of sandstone. The basal bed of the Peninsula Formation was very thick and homogeneous to faintly horizontally laminated towards the top. The basal contact was clearly erosional, with the erosional surface, below the basal thick bed of the Peninsula Formation, undercutting into the beds of the Graafwater Formation (Figures 5 and 6).

Figure 2 - Cape Peninsula map showing the location of study sites. (Source: The Map, 5th Edition. Cartography by Peter Slingsby).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 141

Figure 3 - View, to the west, of Dias Beach and Cape Maclear. The Cape Maclear measured section is just below the label CMC. (Photo: John Rogers).

Figure 4 - Cape Maclear, west of Cape Point and Dias Beach and east of the Cape of Good Hope. Undulating erosional contact between the underlying Graafwater Formation and the overlying Peninsula Formation. Figure (2 m) for scale beside the site of the measured section. (Photo: Zandile Dlamini).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 142

Figure 5 - The contact between the underlying, thinner-bedded Graafwater Formation and the overlying, thicker- bedded Peninsula Formation at the top of Ou Kaapse Weg (OKW), viewed to the south. (Photo: John Rogers).

Figure 6 - The erosional contact between the Graafwater and Peninsula formations observed at the top, southern end of Ou Kaapse Weg, viewed to the west. Two-metre survey rods for scale. (Photo: Zandile Dlamini).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 143 Constantiaberg (CB) The measurements for this stratigraphic column were taken on the western slopes of Constantiaberg (Figures 2, 7 and 8) on an outcrop beside the weir at the top of Blackburn Kloof, where the footpath crosses the stream (Figure 8). It was found that the Graafwater Formation had thinner beds there than at other localities and that the mudstones were found closer to the contact, compared to other locations. The mudstones were also very micaceous. The sedimentary features found in the Graafwater Formation were mud-intraclasts and crossbedding. The top of the Graafwater Formation was defined by a thin bed (11 cm) of homogeneous mudstone, which was horizontally laminated. The Graafwater/Peninsula contact showed no signs of either erosion or of undercutting into the underlyng Graafwater Formation. This basal bed of the Peninsula Formation is a thick bed (74 cm) of cross-bedded sandstone. The contact is conformable.

Table Mountain Four localities, Platteklip Gorge West, Silverstream Cave, the Upper Waterfall and Twelve O’Clock Hole were studied (Figure 9), but only the first and last are described here.

Platteklip Gorge West (PKGW) The stratigraphic column, Platteklip Gorge West (PKGW) was measured on the Contour Path, west of Platteklip Gorge (Figure 14) and just east of a waterfall. The GPS coordinates of PKGW, GPS Stop 9, were 33o 57’ 22.1” S, 18o 24’ 50.3” E. 525 m elevation. A strongly undulating erosional contact, similar to the one at Cape Maclear (Figure 4), was observed (Figure 11). At this locality there was more alternation between the sandstones and mudstones of the Graafwater Formation and there were more siltstones observed than at other locations. Soft- sediment deformation (e.g. pseudonodules), was more pronounced, although it was also observed on the floor of Silverstream Cave, east of Platteklip Gorge. The top bed of the Graafwater Formation was found to be a horizontally laminated siltstone. The Peninsula Formation’s basal sandstone overlies an erosional surface (Figure 11), revealing erosion of the sandstones and siltstones at the top of the Graafwater Formation. Crossbedding was noted on this basal bed and it was overlain by a massive homogeneous very thick sandstone bed (Figure 11).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 144

Figure 7 - Western slopes of Constantiaberg. (Photo: John Rogers).

Figure 8 - Hout Bay and Constantiaberg Map. (Source: The Map, 5th edition. Cartography by Peter Slingsby).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 145

Figure 9 - Table Mountain location map. (Source: The Map, (Seventh Edition) on Table Mountain by Peter Slingsby).

Figure 10 - Site of measured section, PKGW (Platteklip Gorge West) on the face of Table Mountain, beside the Contour Path, west of Platteklip Gorge. (Photo: John Rogers).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 146

Figure 11 - View, towards the southwest, of a strongly undulating erosional contact between the underlying Graafwater Formation and the overlying Peninsula Formation, beside the Contour Path, west of Platteklip Gorge and east of the Cable Station. Note the great contrast, at this locality, between the thinness of the beds of sandstones in the Graafwater Formation and the much greater thickness of the beds of sandstone in the overlying Peninsula Formation. The individual sandstone beds of the Graafwater Formation fine upwards to equally thin beds of maroon- coloured siltstones. (Photo: John Rogers. Co-author, Zandile Dlamini for scale).

Folding and erosion in the Graafwater Formation Folding on the slopes of Devil’s Peak The gentle syncline at the top of the Graafwater Formation, on the western slopes of Devil’s Peak (12 O’Clock Hole), (Figure 12) was investigated further during this study. However, because of the dense vegetation and the overhang of the outcrop no readings were obtained. This means that stereographic projections could not be used to determine whether there is true folding or whether it is an illusion caused by the angle of one’s photograph.

Folding in Constantiaberg Folding was observed in the Graafwater Formation above East Fort and north of Blackburn Kloof, on the eastern shore of Hout Bay, on the western flanks of Constantiaberg (Figures 8 and 13). The Graafwater Formation is folded into an anticline above a nonconformity with the underlying Cape Granite. On the northern limb of the anticline, parasitic “Z” folds were observed (GPS Stop 3: 34o 3’ 25.2” S, 18o 22’ 22.7” E, 217 m elevation) (Figures 8, 13 and 14), showing that the anticline closure was to the east. Dip/dip-direction readings were plotted on an equal-area stereonet (Figure 15).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 147

Figure 12 - View, to the east, of vertical beds of very thin-bedded siltsones of the Malmesbury Group’s Tygerberg Formation, below an angular unconformity with the horizontal sandstones of the Table Mountain Group’s Graafwater Formation. The upper part of the overhang of Twelve O’Clock Hole consists of thin-bedded sandstones and siltstones of the Graafwater Formation, which appear to be folded into a gentle syncline (U-fold) beneath horizontal, thick beds of the Peninsula Formation. Vegetation obscures any actual contact between the two formations on the limbs of this fold. Photo: John Rogers).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 148

Figure 13 - Anticline in the Graafwater Formation on Constantiaberg above a nonconformity with the underlying Cape Granite and an angular unconformity with the overlying Peninsula Formation. (Photo: J. Rogers).

Discussion and Conclusions Viewed from a distance of from 100 to 500 m, the authors are confident that there is clear evidence of an angular unconformity at the top of Twelve O’Clock Hole on the western slopes of Devil’s Peak, where there is a gentle syncline in sandstones at the top of the Graafwater Formation below horizontal sandstones of the Peninsula Formation (Figure 12). In addition, viewed from a distance of 500 to 1000 m, there is even clearer evidence of an anticline in the Graafwater Formation on the western slopes of Constantiaberg, below horizontal sandstones of the Peninsula Formation (Figure 13).

This study has therefore revealed that the contact between the Graafwater and Peninsula formations is not simply conformable, as previously thought. Rust (1967) did record a minor angular unconformity, at outcrop scale, between the two formations in Lekkranskloof, Heerenlogement, (Figure 1), northeast of Cape Town and northwest of Clanwilliam. It is therefore possible that a similar, well-exposed, angular unconformity may yet be found in the Cape Peninsula and elsewhere in the Cedarberg.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 149

Figure 14 - Parasitic folding in the Graafwater Formation on the northern limb of the anticline north of Blackburn Kloof above a nonconformity with the Cape Granite. (Photo: Zandile Dlamini).

Figure 15 - Equal-area stereonet plot. The plot shows the two limbs of the anticline. The northern limb is dipping towards the northwest and the southern limb towards the southeast and the fold axis trends northeast/southwest.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 150 References Armstrong, R., de Wit, M.J., Reid, D., York, D. and Zartman, R. 1998. Table Mountain reveals rapid Pan-African upift of its basement rocks. Journal of African Earth Sciences, 2: 10-11. Hobday, D.K. and Tankard, A.J. 1978. Transgressive-barrier and shallow-shelf interpretation of the lower Paleozoic Peninsula Formation, South Africa. Bulletin Geological Society of America, 89: 1733-1744. McKee, E.D. and Weir, G.W. 1953. Terminology for stratification and cross-stratification in sedimentary rocks. Bulletin Geological Society of America, 64: 381-390. Rust, I.C. 1967. On the Sedimentation of the Table Mountain Group in the Western Cape Province. Unpublished DSc Thesis, Department of Geology, University of Stellenbosch, 110pp, 123 figures, 21 plates. Tankard, A.J., and Hobday, D,K., 1977, Tide dominated back-barrier sedimentation, Early Ordovician Cape Basin, Cape Peninsula, South Africa: Sedimentary Geology, 18: 135-159 Turner , B.R. 1990. Continental sediments in South Africa. Journal of African Earth Sciences, 10: 139-149.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 151 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The Passa Dois Group (Paraná Basin, Permian): investigations in progress

Rosemarie Rohn UNESP – Rio Claro, CP 178, CEP 13506-900, Rio Claro-SP–Brazil; Research grant FAPESP-05/55027-4, [email protected]

Abstract The main characteristics of the Passa Dois Group added to brief new data are presented. The following aspects are focused: a) the thickness variations of the Taquaral and Assistência members of the Irati Formation in the Ponta Grossa Arch area; b) the siliciclastic cyclic successions and the carbonatic intercalations of the Teresina Formation; c) the carbonatic rocks of the Serrinha Member; d) the paleocurrent data of the Morro Pelado Member and implications for environmental interpretations; e) boundaries between the stratigraphic units, intraformational unconformities and some discussions about the sequence stratigraphy; f) the age of some new fossils.

Introduction The Passa Dois Group and its equivalents are widely distributed in the intracratonic Paraná Basin and its maximum thickness is over 1400 m. It is divided into the Irati, Serra Alta, Teresina and Rio do Rasto formations, as well as the Corumbataí Formation in the northern part of the basin (approximately equivalent to the Serra Alta and Teresina formations in southern Brazil). It is often referred to as the “continentalization” interval of the Paraná Basin, but many questions remain in open especially about: the depositional environments, the chronostratigraphic range and the tentative application of “sequence stratigraphy models”. This paper presents a revision and some new stratigraphic and paleontological results, mostly of drill-core analyses, from the eastern and northeastern border of the Paraná Basin. The new data have been obtained by the author and her students at the São Paulo State University - UNESP-Rio Claro (Lourenço, 2002; Lages, 2004; Meglhioratti, 2005, 2006; Tavares, 2007; Ferreira-Oliveira, 2007; Neregato, 2007; available, in part, at

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 152 www.rc.unesp.br/biblioteca/). Part of the most important bibliographic references, maps and correlation charts may also be found in Rohn & Rösler (2000) and Rohn (2001).

Irati Formation The 35-45 m thick Irati Formation is divided into the (lower) Taquaral and (upper) Assistência members. Absolute age determination of the Assistência Member (278.4±2.2 Ma, Santos et al., 2006) indicates a late Artinskian age. A pebbly sandstone with abundant fragmented fish remains that characterizes the lower boundary of the formation in northeastern Paraná Basin is probably a transgressive lag of a third order sequence formed by the reworking of the distal parts of a NE-SW fan delta conglomeratic wedge (Ibicatu Facies). The Taquaral Member (5-25 m) is commonly described as a package of dark homogeneous claystones. Drill-cores from the eastern basin show that it has about five discrete coarsening upward successions, grading from dark shales to coarser siltstones and sometimes to thin fine sandstones and/or to small bivalve coquinas. They correspond to shallowing-up cycles of the depositional environment. The Assistência Member (15-30 m) varies from the eastern to the northeastern Paraná Basin, but in almost all studied drill-cores, the lower part has thin to thick dolomite breccias with anhydrite (or indirect evidences of evaporites). The overlying interval shows mainly a thick bituminous shale superposed by a siltstone, which grade towards northeast to dolomitic micrites and very fine bituminous shale/micritic dolomitic ritmites. These dolomites apparently contain the first mesosaurs (also found in the Whitehill Formation of Karoo and Huab basins). Storm deposited bone beds and crustacean coquinas are common. In the upper part of the Assistência Member, the ritmites are more widespread, but only in the northeastern Paraná Basin, they continue almost to the formation’s top. Rare acritarchs found in some intervals may indicate a limited connection of the aquatic environment with the ocean. Shale and carbonate/evaporite are respectively attributed to humid and dry climate deposition in a very shallow restricted sea. The generally abrupt facies changes probably indicate sedimentation gaps. The Irati Formation is commonly used as a datum in the basin because of its large geographic distribution and relatively small and uniform thickness. However the geometry of the two members varies (Figure 1): the Taquaral Member is thickest and the Assistência Member is thinnest at the Ponta Grossa Arch region; the opposite occurs in neighboring regions to southwest and especially to northeast. These differences suggest that the Ponta Grossa Arch region had different subsidence rates during the deposition of each member; therefore these units are separated by an unconformity and do not belong to the same third order stratigraphic sequence. In a ≥40 km wide area of northeastern Paraná Basin (Ipeúna-Limeira), a very typical 1-1,5 m thick fine sandstone with parallel to hummocky cross stratification separates the Irati

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 153 Formation from the Corumbataí Formation. This fine sandstone should be considered the transgressive lag of the new sequence above the Irati Formation, but it has some articulated mesosaurid bones in addition to abundant crustaceans. Therefore, although this sandstone layer shows a great change in the depositional environment, it was deposited before the extinction of the organisms. At the most northeastern outcrop region of the Passa Dois Group, mapped as Corumbataí Formation, large oriented stromatolites are associated to articulated mesosaurid bones. This paleotological data as well as the lack of the typical Irati Formation sediments in this region suggests that the stromatolites grew in the most marginaly preserved part of the basin during the deposition of the Assistência Member. A similar situation is found in the Whitehill Formation in Namibia.

Figure 1 – Correlation section of the Irati Formation.

Serra Alta Formation This usually 60-90 m thick, very homogeneous formation is composed of dark shales with rare thin intercalations of fish bone beds and micrites with bivalves. The thickness of this unit decreses very quickly in the eastern and northeastern regions of the basins (towards the paleomargin). This may be due to lower subsidence rates, or possibly there are gaps at the base and/or in other intervals. Palynological content is very poor and no acritarchs are found.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 154 Teresina Formation This 280-330 m thick formation has traditionally been described as a monotonous unit predominantely composed of grey siltstones, interlaminated shales and fine sandstones with flaser-wavy-lenticular-bedding, sometimes with mudcracks, some fine sandstones and thin calcareous intercalations. It was believed that sandstones and mudcracks would increase upwards along the unit and towards the northeast (where the formation has red colours and is regionally named Corumbataí Formation). However, borehole cores show a diferent picture. The formation is composed of several siliciclastic coarsening upward successions – “SCUS” (~parasequences or high frequency sequences), from dark shales to the interlaminated flaser-wavy-lenticular-bedded shales/fine sandstones and sandstones at the top, which indicate shallowing up cycles of the depositional environments. Mudcracks are common in almost the whole formation. The carbonate rocks are generally submetric, separated from the SCUS by an erosional surface. They are composed of shelly micrites, oosparites, stromatolites and/or carbonate breccias with complex vertical arrangement, usually mixed with quartz of even feldspar grains, sometimes with very dark thin shales. A thick, distinctly sandier interval (including sandstones with hummocky cross stratification) occurs in the lower part of the Teresina Formation in the eastern/northeastern Paraná Basin. In a sequence stratigraphic approach, this sandier interval is at the end of a third order sequence that began at the base of the Serra Alta Formation. The remaining Teresina Formation would contain two more sequences, but it is very difficult to distinguish their boundaries, and their transgressive and highstand intervals. The top of the second possible sequence is positioned at the level of an apparent unconformity that seems to coincide with the boundary between the Pinzonella illusa and Pinzonella neotropica bivalve zones, where this intraformational unconformity increases towards the northeast of the basin (Figure 2). The third sequence finishes at the boundary between the Teresina and Rio do Rasto formations or encompasses a part of the upper formation. All sequences thin towards the basin’s paleomargin at northeast and the third one is particularly less thick due to erosion before the sedimentation of the overlying São Bento Group. The Teresina Formation was probably deposited in an interior epeiric sea influenced by storm winds and climatic variations. No formally described fossil may be considered as a true marine representant. The relatively flat bottom of the basin caused similar sedimentation in wide areas, without clear differences between onshore and offshore settings. In marginal areas, wave energy was minimal, dampened by friction along a broad shallow sea floor. In more distal aras, storm waves may have produced gentle shoals. The occurrence of oosparites very far from the basin’s margins suggests that the ooids were possibly formed in upper parts of the shoals. During rainier intervals, the sea expanded, waters were less salty and the sedimentation was predominantly siliciclastic; in dryer intervals, there occurred greater water evaporation, the salinity increased, less siliciclastics entered the basin, more

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 155 carbonatic sediments were deposited and large areas of the basin were exposed. The boundaries between the SCUS probably represent significative gaps. An intriguing fact is that palynological analyses of supposely distal deposits of the formation revealed predominance of fern spores, which could not have been widely transported. Perhaps, some shoals produced during storms were occasionally exposed and transformed into islands that could have been inhabited by plants. Fern fronds show clear xeromorphic characteristics, which may be caused by water deficit.

Rio do Rasto Formation This formation is subdivided into the Serrinha and Morro Pelado members. The boundary between the Teresina and Rio do Rasto formations is not easily determined in outcrop because the rock’s granulation remains predominantly fine, the beds continue tabular and strong weathering obliterated many important structures. The most conspicuous evidence of an environmental change is the appearence of conchostracans, which are fresh water crustaceans. There is also an abrupt change of the bivalve and plant assemblages. Lycopod plants disappear and other plants as glossopterids, ferns and sphenopsids become much more abundant. The bivalve Leinzia similis, which is typical of the lower part of the Serrinha Member, was also recently found in the Gai-As Formation in Namibia, some meters below 265,5 ± 2,2 Ma old volcanic ashes (~Wordian-Capitanian boundary). Only one borehole (SP-23-PR; Figure 2), from eastern-northeastern Paraná Basin, has cores suitable to show the boundary between the formations. Therefore, sequence stratigraphy interpretations are very preliminary. The 150-250 m thick Serrinha Member is characterized by an abrupt appearance of submetric to metric repetitive fine sandstones that present coarser grains at the base and fine upwards slightly. Mudstones and siltstones still occur, but they rarely show fine interlaminations and flaser-wavy-lenticular bedding. Mud cracks are not common. Some parts show SCUS, as in the Teresina Formation, but usually it is much more difficult to recognize a repetive organization of the lithofacies. In the mentioned borehole and in the outcrop area slightly further northeast, about 40 meters above the base, the Serrinha Member presents a predominantly carbonatic interval (~25 m) characterized by thick and thin micrites (with bivalves and ostracodes) and coquinas, including oncoids. Furthur northeast, the member is absent due to erosion. In the eastern basin, as in the type locality of the Serrinha Member, thick carbonatic rocks apparently are missing. In general, the member may correspond to a lacustrine environment, with variable contribution of river discharges. The most carbonatic area and interval are probably related to less direct fluvial influence. The Morro Pelado Member (250-300 m thick at outcrops) is mainly characterized by lenticular, wedged, lobate and sigmoidal fine sandstones, commonly with cross stratification, and red pelitic intercalations with abundant conchostracans, some bivalves, fish scales and

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 156 relatively frequent, but poorly diversified stems and leaves. Breccias composed of intraformational pelitic clasts are rare, but in the Ponta Grossa Arch area they may reach over 1 m thickness. In the southeastern regions of the Paraná Basin, the Rio do Rasto Formation is sandier and the first lenticular bodies apparently appear in lower stratigraphic positions in comparison with the northern regions, but the correlations are not very precise, even when borehole gama ray and resistivity logs are used. For a long time, the Morro Pelado Member was interpreted as a product of meandering fluvial systems or deltaic plain systems. However, some sandstones have large trough cross stratification that probably correspond to eolian dunes; other sandstones have climbing-ripple cross-laminations that are typical of hyperpicnal fluxes that entered a low energy water body. These facies supported the re-interpretation that the Morro Pelado Member deposited in a complex semiarid lacustrine system surrounded by eolian dunes and intermittent fluvial channels. The lenticular sandstones with ripple drift cross laminations are interpreted as mouth bars facies deposited in the lakes during brief, high flood discharges caused by sporadic intense storms. However, recent studies indicated that the cross stratifications or laminations of the sandstones (except the eolic sandstones) generally indicate a mean sediment transport northwards and not in centripetal orientations as expected in lakes. A new alternative is being investigated with the colleague Dr. Mario Assine. There could have existed an alluvial source area at south and the observed facies associations may represent crevasse splay deposits and inundites, laterally associated with occasional shallow fluvial channel deposits formed in laterally extensive coalescent fluvial plains. Frequent fluvial-eolian interactions and desiccation cracks point to the alternation of short-lived flood events and periods of widespread dry conditions. Paleocurrent measurements of the Morro Pelado Member eolian sandstones at its eastern outcrop belt indicate a predominant wind blowing southwards. The upper part of the Rio do Rasto Formation in the outcrop area, already interpreted as Triassic by some authors, may be surely positioned in the Permian. Tetrapods (including Endothiodon) and conchostracans (as Hemicycloleaia mitchelli in the formation’s upper portion) are not younger than Wuchiapingian. Plants as Schizoneura gondwanensis corroborate this interpretation.

References Rohn, R., 2001, A estratigrafia da Formação Teresina (Permiano, Bacia do Paraná) de acordo com furos de sondagem entre Anhembi (SP) e Ortigueira (PR). Ciência-Técnica-Petróleo. Seção:Exploração de Petróleo, v.20, p.209-218.

Rohn, R. and Rösler, O., 2000, Middle to Upper Permian Phytostratigraphy of the Eastern Paraná Basin. Revista Universidade Guarulhos, Geociências, v.5 (no especial), p.69-73.

Santos, R.V., Souza, P.A., Alvarenga, C.J.S., Dantas, E.L., Pimentel, M.M., Oliveira, C.G. and Araujo, L.M., 2006, Shrimp U-Pb zircon dating and palynology of bentonitic layers from the Permian Irati Formation, Paraná Basin, Brazil. Gondwana Research, v.9, n.4, p.456-463.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 157

Figure 2 – A: Correlation section of the Passa Dois Group in the E/NE Paraná Basin; B: Detail of “A”; α and ß: examples of lithologic profiles (located in “B”).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 158 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Karoo tetrapod biostratigraphy: Relevance to understanding gondwanan development

Bruce Rubidge Bernard Price Institute for Palaeontological Research, School of Geosciences, University of the Witwatersrand, Johannesburg, South Africa, [email protected]

Abstract The Permian to Jurassic fossil tetrapod wealth of the Karoo of South Africa has greatly enhanced understanding of the evolution of important tetrapod lineages. Because these fossils provide the best record of continental Permian to Jurassic faunal biodiversity they have enabled biostratigraphic subdivision of the Karoo which has international applicability. Recent enhanced biostratigraphic resolution and has refined basin models and a better understanding of tetrapod biogeography and biodiversity across Pangaea. Recent discovery of dateable ash beds together with biozone defining fossils will enhances age determination for vertebrate biozones, and thus the possibility of ascertaining rates of evolution in tetrapod lineages and the timing of events for basin modelling.

Introduction The Karoo Basin of South Africa contains a near continuous sequence of late Carboniferous to the early Jurassic sedimentary rocks hosting a wealth of tetrapod fossils which have enabled biostratigraphic subdivision of the Beaufort Group (Rubidge, 1995) and Elliot and Clarens formations (Kitching and Raath, 1984). Recent work has refined biostratigraphic divisions and enhanced correlation (Botha and Smith 2007, Botha et al 2007, Hancox and Rubidge, 1996; Hancox et al.., 2000; Modesto et al. 2001, Neveling et al.., 2000, 2005, Rubidge 2005). Tetrapod faunas of “Karoo” age are known from around the world (eg. Rubidge 2005), but the main Karoo Basin of South Africa remains the most time transgressive succession.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 159 Faunal biodiversity patterns Because of its relatively complete fossil tetrapod record the Karoo serves as the reference section for Permian –Jurassic tetrapod biostratigraphic correlation. The following summary indicates correlatives of Karoo tetrapods with other successions.

Wordian (288 – 265.8 Ma) -The fauna of the Eodicynodon Assemblage Zone of South Africa (Rubidge 1995) has been correlated with North China (Cheng and Li, 1996; Li et al., 1996), and Russia. Dicynodonts, gorgonopsians and therocephalians were already present in South Africa (Rubidge, 1995b) but had not yet appeared in Laurasian fauna (Battail, 2000).

Capitanian (265.8 – 260.4 Ma) - The fauna of the Tapinocephalus Assemblage Zone (Boonstra, 1969, Keyser and Smith 1977/78, Smith and Keyser 1995) has been correlated with Russia (Modesto and Rybczynski, 2001), Brazil (Langer, 2000), Zimbabwe (Lepper et al., 2000; Munyikwa, 2001). Present in South Africa, but not known from other countries, are dicynodonts, gorgonopsians, and the parareptile Eunotosaurus.

Wuchiapingian (260.4 – 253.8 Ma) - This period most likely corresponds with the Tropidostoma and Cistecephalus Assemblage Zones (Modesto and Smith, 2001; Smith and Keyser, 1995). Correlatable faunas are known from Zimbabwe (Bond, 1973), Zambia (Lee et al., 1997), Malawi (Haughton, 1926), Tanzania (Gay and Cruickshank, 1999), Mocambique (Latimer et al., 1995), India (Ray, 1999, 2000) (Barbarena and Araujo, 1975; Battail 2000) and China (Li and Cheng, 1995).

Changhsingian (Late Tatarian) (253.8 – 251 Ma) - The fauna of the Dicynodon Assemblage Zone (Kitching, 1995, Smith and Ward, 2001) correlates with Zambia, Tanzania (Battail, 2000), Madagascar (Mazin and King, 1991, Smith, 2000), China (Li and Cheng, 1995), England (Sues and Boy, 1988), Germany (Sues and Boy, 1988), Laos (Battail, 2000), Russia (Battail, 2000, Battail and Surkov, 2000) and Scotland (Benton and Walker, 1985). Dicynodonts, gorgonopsians and pareiasaurs were present in southern Africa since the Wordian, but appeared as newcomers in the Laurasian record in the Changhsingian (Battail, 2000).

Induan (251 – 249.7 Ma) - The lower part of the Lystrosaurus Assemblage Zone of South Africa has been correlated with the Induan (Neveling 2004; Shishkin et al., 1995). The fauna has been described (Groenewald and Kitching, 1995, Neveling 2004; Smith and Ward, 2001). Tetrapod-bearing Induan rocks are present in Antarctica (Colbert and Kitching, 1981; Cosgriff et

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 160 al., 1982), Australia (Cosgriff, 1965; Warren 1991), Brazil (Cisneros and Schultz, 2002; Lavina, 1983), China (Cheng et al., 1996), Greenland (Shishkin et al., 1996; Schoch and Milner, 2000), Madagascar (Schoch and Milner, 2000); India (Bandyopadhyay et al., 2002; Schoch and Milner, 2000; Shishkin et al., 1996), Pakistan (Damiani and Welman, 2001; Schoch and Milner, 2000) and Russia (Damiani and Welman, 2001; Shishkin et al., 2000).

Battail (1993) considers Lystrosaurus to be restricted to the base of the Induan, but Neveling (2004) considers its range to extend from the Changhsingian to the Olenekian.

Olenekian (249.7 – 245 Ma) - Both the Lystrosaurus Assemblage Zone (Neveling, 2004) and Subzone A of the Cynognathus Assemblage Zone (Shishkin, et al., 1995; Hancox, 2000) are considered to be Olenekian. Olenekian tetrapod bearing rocks are present in Antarctica (Colbert and Cosgriff, 1974), Australia (Damiani, 1999, 2001a), India (Bandyopadhyay, et al., 2002), China (Li, 1983), Germany (Schoch, 1999b), Russia (Battail and Surkov, 2000; Benton and Allen, 1997; Shishkin et al., 2000; Gower and Sennikov, 2000), Spitsbergen (Schoch and Milner, 2000)

Anisian (245 -237 Ma) - Subzones B (Hancox, et al., 1995, Shishkin, et al., 1995) and C (Hancox and Rubidge, 1996, 1997 of the Cynognathus Assemblage Zone are respectively considered Early and Late Anisian.

Subzone B faunal elements correlate with Namibia (Keyser, 1973) Tanzania, Zambia (Hancox, 1988), Antarctica (Hammer, 1995) Argentina (Abdala, 1996,) India. (Kutty et al., 1987; (Bandyopadhyay and Sengupta, 1999), China (Renaut, 2001), Germany (Schoch, 1999a), Russia (Gower and Sennikov, 2000, Renaut, 2001). England (Walker, 1969), United States of America (Damiani, 2001; Schoch and Milner, 2000)

Faunal correlatives of Subzone C occur in Argentina (Neveling, 2004), Australia (Damiani, 2001), India (Damiani, 2001b; Kutty, et al., 1987), Namibia (Keyser, 1973), Zambia (Damiani, et al., 2000, King, 1988, Hancox, 1998), Tanzania (Cox and Li, 1983; King, 1988; Crompton, 1955), China (Chen and Li, 1995; Hancox, 1998).

Ladinian (237 – 228 Ma)- The post Beaufort depositional hiatus covered most of the Ladinian (see Visser, 1984, 1991; Hancox, 1998).

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 161 Carnian (228 – 216.5 Ma)- The only tetrapod remains from the Molteno Formation are footprints (Raath et al., 1992). Hancox (2000) considers at least the lower half of the Molteno Formation to be Carnian.

Norian-Rhaetian (216.5 – 199.6 Ma) - The age of the lowermost part of the Elliot Formation (“Euskelesaurus” Range Zone) can be Carnian-Rhaetian, but is more likely to be Norian (Hancox, 2000). The contact between the Euskelosaurus and Massospondylus Range Zones is marked by a subaerial unconformity (Bordy, et al., 2004b, Kitching and Raath, 1984) and considered the Triassic-Jurassic boundary (Olson and Galton, 1984).

Correlatives are Argentina (Busbey and Gow, 1984, Yates 2003), Argentina (Baez and Marsicano, 1998) Zimbabwe (Cooper, 1980), Britain (Fraser, 1997; Rauhut and Hungerbuhler, 1998), Greenland (Jenkins, et al., 1994), Germany (Rauhut and Hungerbuhler, 1998), United States of America (Colbert, 1989)

Hettangian, Sinnemurian and Pliensbachian (199.6 – 183 Ma) - The Massospondylus Assemblage Zone encompasses the Upper Elliot and Clarens formations (Kitching and Raath 1984). Faunal correlatives with other countries are: Australia (Schoch and Milner, 2000), India (Bandyopadhyay, et al., 2002, Kutty, et al., 1987) Namibia (Raath, pers comm.), Zimbabwe (Gow and Raath, 1977; Raath 1977), Britain (Evans and Kermack, 1997; Fraser, 1997), Canada (Sues, et al., 1997, Shubin, et al., 1997), China (Crompton and Luo, 1993; Luo and Wu, 1997; Luo, et al., 2001) Germany (Sues, 1985b), Mexico (Clark, et al., 1997), United States of America (Crompton and Lou, 1993, Sues, et al., 1997, Lucas and Heckert, 2001)

Basin development Enhanced biostratigraphic resolution has improved time resolution of basinal depositional events for the subaerial continental deposits of the basin (Hancox and Rubidge, 1997, 2002). This enabled Catuneanu et al. (1998) to model the Karoo Basin in terms of first and second order sequences, and suggesting that the Karoo behaved as a partitioned basin with reciprocal proximal (foredeep) and distal (forebulge) deposition across the basin hingeline. Geological work around the Ecca-Beaufort contact demonstrated shown that while the lithological shoreline transition from the Ecca to the Beaufort is the same all around the basin (Rubidge, et al., 2000), fossil representation becomes progressively younger in a northerly direction (Rubidge et al., 1999; Welman, et al., 2001). Thus, while Middle to Late Permian Beaufort Group continental fluvial deposition took place in the southern part of the basin (proximal sector), contemporaneous deposition of

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 162 subaqueous Ecca Group rocks (Volksrust Formation) occurred in the northern distal sector. By the end of the Permian, the Ecca “sea” had been completely filled, and continental fluvial deposition occurred in both sectors (Rubidge and Hancox, 2002) giving rise to the upper Beaufort Rocks which were also deposited in the proximal and distal parts of the basin at different times (Neveling 2004) with the lowermost and uppermost Katberg Formation being limited to the proximal sector (Hancox et al. 2002, Neveling 2004) while the the progressively smaller surface areas of the Mid-Triassic Cynognathus B and C subzones of the Burgersdorp Formation indicate a receding basin (Hancox, 1998). The overlying Molteno, Elliot and Clarens formations formed in the distal sector of the Karoo Basin with earliest “Stormberg” deposition limited to the southern part of the basin (Hancox, 1998). The Molteno basin continued to expand northwards to reach its maximum extent when the Indwe Member was deposited in the Ladinian whereafter the basin again receeded (Hancox 1998). The lower Elliot Formation was deposited by meandering fluvial systems, while the upper part of the Formation resulted from ephemeral fluvial processes (Bordy et al., 2004a). Continental fluvial deposition continued in the Early Jurassic in an increasingly arid environment, leading ultimately to the expansive erg in represented by the Clarens Formation in South Africa.

Conclusion The rocks of the Karoo are globally recognised for the wealth of fossil tetrapods which span a stratigraphic record from the Mid-Permian to the Early Jurassic. Refined stratigraphic and geographic documentation in the collecting of fossils has enhanced Karoo biostratigraphic resolution, global correlation of Permian to Jurassic vertebrate faunas, and basin development models. Detailed studies addressing fine-tuned tetrapod phylogenetic, biostratigraphic and geographic resolution is now essential. South Africa occupied a central position in Gondwana and because of its extensive record, both in time and geography, the unravelling of the unique fossil record from the South African Karoo is essential for further understanding of the development of the Gondwanan world.

References Abdala, F., 1996, Redescripión del cráneo y reconsideración del la validez de Cynognathus minor (Eucynodontia- Cynognathidae) del Triásico inferior de Mendoza: Ameginiana, v. 33, p. 115-126.

Bandyopadhyay, S. and Sengupta, D.P. 1999. Middle Triassic vertebrates of India: Journal of African Earth Sciences, v. 29, p. 233-241.

Bandyopadhyay S., RoyChowdhury T.K. and Sengupta D.P., 2002, Taphonomy of some Gondwana vertebrate assemblages of India. Sedimentary Geology, v. 147, p. 219-245.

Baez, A.M. and Marsicano, C.A. 1998. A heterodontosaurid ornithischian in the Upper Triasic of southern Patagonia? Journal of African Earth Sciences, v. 27 (1A), p. 14-15.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 163

Barbarena, M.C. and Araujo, D.C., 1975, Tetrapodes fosiles de Sudamerica y deriva continental: Actas I Congreso Argentino de Paleontologia y Bioestratigrafia (Tucuman), v. 1, p. 497-504.

Battail, B., 1993, On the Biostratigraphy of Triassic Therapsid-Bearing formations: in Lucas, S.G. and Morales, M., eds., The NonmarineTriassic. New Mexico Museum of Natural History and Science Bulletin, v. 3, p. 31-35.

Battail, B., 2000, A comparison of Late Permian Gondwanan and Laurasian amniote faunas: Journal of African Earth Sciences, v 31, p. 165-174.

Battail, B and Surkov, M.,2000. Mammal-like reptiles from Russia: in Benton, M.J., Shishkin, M.A., Unwin, D.M., and Kurochkin, E.N., eds., The age of dinosaurs in Russia and Mongolia, Cambridge University Press, Cambridge, p. 86- 119.

Benton, M.J. and Allen, J.L., 1997, Boreopricea from the Lower Triassic of Russia and the relationships of the prolacertiform reptiles: Palaeontology, v. 40, p. 931-953.

Benton, M.J and Walker, A.D., 1985, Palaeoecology, taphonomy and dating of Permo-Triassic reptiles from Elgin, northeast Scotland: Palaeontology, v. 28, p. 207-234.

Bond, G., 1973, The Palaeontology of Rhodesia: Bulletin of the Geological Survey of Rhodesia, v. 70, p. 1-121.

Boonstra, L.D., 1969, The fauna of the Tapinocephalus Zone (Beaufort beds of the Karoo): Annals of the South African Museum, v. 56, p. 1-73.

Bordy, E.M., Hancox, P.J. and Rubidge, B.S. (2004a). Fluvial style variations in the Late Triassic-Early Jurassic Elliot Formation, main Karoo Basin, South Africa: Journal of African Earth Sciences, v.38, p. 383-400.

Bordy, E.M., Hancox, P.J. and Rubidge, B.S. (2004b).Main Karoo Basin development during the deposition of the Elliot Formation (Late Triassic- Early Jurassic), Karoo Supergroup, South Africa. South African Journal of Geology, v. 107, p. 395-410.

Botha, J. and Smith, R.M.H., 2007, Lystrosaurus species composition across the Permo-Triassic boundary in the Karoo Basin of South Africa: Lethaia, v.40, p. 125-137.

Botha, J. Modesto, S.P. and Smith, R.M.H. 2007, Extended procolophonoid reptile survivorship after the end-Permian extinction: South African Journal of Science, v.103, p.54-55.

Busbey, A.B. and Gow, C., 1984, A new protosuchian crocodile from the Upper Triassic Elliot Formation of South Africa: Palaeontologia africana, v. 25, p. 127-149.

Catuneanu, O., Hancox, P.J., Rubidge, B.S., 1998. Reciprocal flexural behaviour and contrasting stratigraphies, a new Basin development model for the Karoo retroarc foreland system, South Africa. Basin Research, v. 10, p. 417-439.

Cheng, Z. and Li, J., 1996, First record of a primitive anteosaurid dinocephalian from the Upper Permian of Gansu, China: Vertebrata PalAsiatica, v. 34, p. 123-134.

Cheng Zhenwu, Wu Shaozhu and Fang Xiaosi.,1996, The Permian-Triassic sequences in the southern margin of the Junggar Basin and the Turpan Basin, Xinjiang, China: in Field Trip Guide 30th International Geological Congress 4-14 August 1996, Beijing, China. Geological Publishing House Beijing, China, p.1-25.

Cisneros, J.C. and Schultz, C.L., 2002. Procolophon brasiliensis n. sp., a new procolophonid reptile from the lower Triassic of southern Brazil: Neues Jahrbuch für Geologie und Palaontologie, Monatschefte, v. 2002, p. 641-648.

Colbert, E.H., 1989, The Triassic dinosaur Coelophysis: Museum of Northern Arizona Bulletin, v. 57, p. 1-160.

Colbert, E.H. and Cosgriff, J.W., 1974, Labyrinthodont amphibians from Antarctica: American Museum Novitates, v. 2552, p. 1-20.

Colbert, E. H. and Kitching, J.W., 1981, Scaloposaurian reptiles from the Triassic of Antarctica: American Museum Novitates, v. 2709, p. 1-22.

Cosgriff, J.W., 1965, A new genus of temnospondyli from the Triassic of Western Australia: Journal of the Royal Society of Western Australia, v. 48, p. 65-90.

Cosgriff, J.W., Hammer, W.R. and Ryan, W.J., 1982, The Pangean reptile Lystrosaurus macaigi, in the Lower Triassic of Antarctica: Journal of Palaeontology, v. 56, p. 371-385.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 164

Cox, C.B. and Li, J., 1983, A new genus of Triassic dicynodont from East Africa and its classification. Palaeontology, v. 26, p. 389-406.

Crompton, A.W., 1955, On some Triassic cynodonts from Tanganyika: Proceedings of the Zoological Society of London, v. 125, p. 617-669.

Crompton, A.W and Lou, Z., 1993, Relationships of the Liassic mammals, Sinocodon, Morganucodon oehleri and Dinnetherium: in Szalay F.S., Novacek, M.J., and Mc Kenna, M.C., eds., Mammal Phylogeny: Springer-Verlag, Berlin, 30-40.

Damiani, R.J., 1999. Parotosuchus (Amphibia, Temnospondyli) in Gondwana, biostratigraphic and palaeobiogeographic implications: South African Journal of Science, v. 95, p. 458-460.

Damiani, R.J., 2001a, Parotosuchus (Amphibia, Temnospondyli) from the Cynognathus Assemblage Zone of South Africa, cranial morphology and relationships: Alcheringa, v. 25, p. 351-379.

Damiani, R.J. (2001b). A systematic revision and phylogenetic analysis of Triassic mastodonsauroids (Temnospondyli, Stereospondyli): Zoological Journal Linnaean Society, v. 133, p. 379-482.

Damiani, R., and Welman, J., 2001, A long snouted trematosaurid amphibian from the Early Triassic of South Africa: South African Journal of Science, v. 97, p. 318-320.

Damiani, R.J., Neveling, J., Hancox, P.J., and Rubidge, B.S., 2000, First trematosaurid (Amphibia, Temnospondyli) from the Lystrosaurus Assemblage Zone and its biostratigraphic implications: Geological Magazine, v. 137, p. 659-665.

Evans, S.E. and Kermack, K.A., 1997. Assemblages of small tetrapods from the Early Jurassic of Britain: in Fraser, N.C. and Sues, H.D., eds., In the shadow of the dinosaurs – early Mesozoic tetrapods: Cambridge University Press, Cambridge, p. 271-283.

Fraser, N.C.,1997, Assemblages of small tetrapods from British Late Triassic fissure deposits: in Fraser, N.C. and Sues, H.D., eds., In the shadow of the dinosaurs – early Mesozoic tetrapodsP Cambridge University Press, Cambridge, p. 214-226.

Gay, S.A. and Cruickshank, A.R.I., 999, Biostratigraphy of the Permian tetrapod faunas from the Ruhuhu Valley, Tanzania: Journal of African Earth Sciences, v. 29, p. 195-210.

Gow, C.E. and Raath, M.A., 1977. Fossil vertebrate studies in Rhodesia: sphenodontid remains from the Upper Trias of Rhodesia: Palaeontologia Africana, v. 20, p. 121-122.

Gower, D.J. and Sennikov, A.G., 2000, Early archosaurs from Russia: in Benton, M.J., Shishkin, M.A., Unwin, D.M. and Jurochkin, E.N., eds., The age of dinosaurs in Russian and Mongolia: Cambridge University Press, p. 140-159.

Groenewald and Kitching (1995). Biostratigraphy of the Lystrosaurus Assemblage Zone: In: Rubidge,B.S., ed., Biostratigraphy of the Beaufort Group (Karoo Supergroup): SACS Biostratigraphic Series No. 1, Council for Geoscience, Pretoria, p. 35-39.

Hammer, W.R., 995, New therapsids from the Upper Fremouw Formation (Triassic) of Antarctica: Journal of Vertebrate Paleontology, v. 15, p. 105-112.

Hancox, P.J., 1998, A Stratigraphic, Sedimentological and Palaeonvironmental synthesis of the Beaufort-Molteno contact in the Karoo Basin: Unpublished Ph.D. thesis, University of the Witwatersrand, Johannesburg, South Africa,381p.

Hancox, P.J., 2000, The Continental Triassic of South Africa: Zentralblatt für Geologie und Paläontologie, Teil I, 1998, p. 1285-1324.

Hancox, P.J., and Rubidge, B.S., 1996, The first specimen of the Mid-Triassic dicynodont Angonisaurus from the Karoo of South Africa, implications for the dating and biostratigraphy of the Cynognathus Assemblage Zone, Upper Beaufort Group: South African Journal of Science, v. 92, p. 391-392.

Hancox, P.J. and Rubidge, B.S.,1997, The role of fossils in interpreting the development of the Karoo Basin: Palaeontologia africana, v. 33, p. 41-54.

Hancox, P.J., and Rubidge, B.S., 2001, Breakthoughs in the biodiversity, biostratigraphy, biogeography and basin analysis of the Beaufort Group: Journal of African Earth Sciences, v. 33, p. 563-577.

Hancox, P.J., Shishkin, M.A., Rubidge, B.S., and Kitching, J.W., 1995, A threefold subdivision of the Cynognathus Assemblage Zone (Beaufort Group, South Africa) and its palaeogeographical implications: South African Journal of Science, v. 91, p. 143-144.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 165

Hancox, P.J., Damiani, R.J., and Rubidge, B.S., 2000, First occurrence of Paracyclotosaurus (Temnospondyli, Capitosauridae) in the Karoo Basin of South Africa and its biostratigraphic significance: South African Journal of Science, v. 96, p. 135-137.

Hancox, P.J., Brandt, D., Reimold, W.U., Koeberl, C., and Neveling, J., 2002, Permian-Triassic boundary in the northwest Karoo Basin: Current stratigraphic placement, implications for basin development models, and the search for evidence of impact: in Koeberl, C. and Mac Leod, K.G., eds., Catastrophic Events and Mass Extinctions: Impacts and Beyond. Boulder, Colorado, Geological Society of America Special Paper, 356, p. 429-444.

Haughton, S.H., 1926, On Karoo vertebrates from Nyasaland: Transactions of the Geological Society of South Africa, v. 29, p. 69-83.

Jenkins, F.A., Shubin, N.H., Amaral, W.W., Gatesy, S.M., Schaff, C.R., Clemmensen, L.B., Downs, W.R., Davidson, A.R., Bonde, N., and Osbaeck, F., 1994, Late Triassic continental vertebrates and depositional environments of the Flemming Fjord Formation, Jameson Land, East Greenland: Meddelelser om Grønland, Geoscience, v. 32, p. 1-25.

Keyser, A.W., 1973. A new Triassic vertebrate fauna from South West Africa: Palaeontologia africana, v. 16, p. 1-15.

Keyser, A.W., and Smith, R.M.H.,1977-78, Vertebrate biozonation of the Beaufort Group with special reference to the Western Karoo Basin. Annals of the Geological Survey of South Africa, v. 12, p. 1-36.

King, G.M., 1988, Anomodontia:. in Wellnhofer, P., Ed., Encyclopedia of Paleoherpetology 17C. Stuttgart, Fischer, 174 p.

Kitching, J.W., 1977, The distribution of the Karroo vertebrate fauna: Memoir Bernard Price Institute Palaeontological Research, University Witswatersrand, v. 1, p. 1-131.

Kitching, J.W., 1995, Biostratigraphy of the Dicynodon Assemblage Zone: in Rubidge,B.S., Ed., SACS Biostraigraphic Series No. 1, Council for Geoscience, Pretoria, p. 29-34.

Kitching, J.W., and Raath, M.A., 1984, Fossils from the Elliot and Clarens formations (Karoo sequence) of the northeastern Cape, Orange Free State and Lesotho, and suggested Biozonation based on Tetrapods. Palaeontologia africana, v. 25, p. 111-125.

Kutty, T.S., Jain, S.L., and RoyChowdhury, T., 1987, Gondwana sequence of the Northern Pranhita-Godavari valley: its stratigraphy and vertebrate faunas: Palaeobotanist, v. 36, p. 214-229.

Langer, M.C., 2000. The first record of dinocephalians in South America, Late Permian (Rio do Rasto Formation) of the Paran‡ Basin, Brazil: Neues Jahrbuch für Geologie und Palaontologie, Abhandlungen,v. 215, p. 69-95

Latimer, E.M, Gow, C.E., and Rubidge, B.S., 1995,. Dentition and feeding niche of Endothiodon (Synapsida; Anomodontia): Palaeontologia africana, v. 32, 75-82.

Lavina, E.L., 1983, Procolophon pricei sp. n. um novo ŕeptil procolofonídeo do Triássico do Rio Grande do Sul: Iheringia (Geol.), v. 9, p. 51-78.

Lee, M.S.Y., Gow, C.E., and Kitching, J.W.,1997, Anatomy and relationships of the pareiasaur Pareiasuchus nasicornis from the Upper Permian of Zambia: Palaeontology, v. 40, p. 307-335.

Lepper, J., Raath, M.A., and Rubidge, B.S., 2000, A diverse dinocephalian fauna from Zimbabwe: South African Journal of Science, v. 96, p. 403-405.

Li, J., 1983, Tooth replacement in a new genus of procolophonid from the Early Triassic of China: Palaeontology, v. 26, p. 567-583.

Li, J., and Cheng, Z., 1995, A new Late Permian vertebrate fauna from Dashankou, Gansu with comments on Permian and Triassic vertebrate assemblage zones of China: in Sun AL, Wang YQ., Eds., Short papers of Sixth Symposium on MesozoicTerrestrial Ecosystems and Biota. Beijing, China Ocean Press. P. 33-7.

Li J., Rubidge, B.S., and Cheng, Z., 1996, A primitive anteosaurid dinocephalian from China - implications for the distribution of earliest therapsid faunas: South African Journal of Science, v. 92, p. 252-253

Luo, Z., and Wu X-C., 1997, The small tetrapods of the Lower Lufeng Formation, Yunnan, China: in Fraser, N.C., and Sues, H.D., Eds., In the shadow of the dinosaurs – early Mesozoic tetrapods. Cambridge University Press, p. 251-270.

Lucas, S.G., and Heckert, A.B., 2001, Theropod dinosaurs and the Early Jurassic age of he Moenave Formation, Arizona- Utah, USA: Neues Jahrbuch für Geologie und Palaontologie, Monatshefte, v. 2001, p. 435-448.

Luo, Z-X., Crompton, A.W., and Sun, A.-L., 2001, A new mammaliaform from the Early Jurassic and evolution of mammalian characters: Science, v. 292, p. 1535-1540.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 166

Mazin, J.M., and King, G.M., 1991, The first dicynodont from the Late Permian of Malagasy: Palaeontology, v. 34, p. 837- 842.

Modesto, S.P., and Rybczynski, N., 2000, The amniote faunas of the Russian Permian, implications for Late Permian terrestrial vertebrate biogeography: in Benton, M.J., Kurochkin, E.N., Shishkin, M.A., Unwin, D.M., Eds., The age of dinosaurs in Russia and Mongolia: New York, Cambridge University Press, p.17-34.

Modesto, S., and Smith, R.M.H., 2001, A new Late Permian captorhinid reptile: a first record from the South African Karoo: Journal of Vertebrate Paleontology, v. 21, p. 405-409.

Modesto, S.P., Rubidge, B.S., de Klerk, W.J., and Welman, J., 2001, A new dinocephalian therapsid fauna on the Ecca- Beaufort contact in the Eastern Cape Province, South Africa. South African Journal of Science, v. 97, p. 161-163.

Munyikwa, D., 2001. Cranial morphology of a primitive dinocephalian (Amniota, Therapsida) from the Madumabisa Mudstone Formation, Zimbabwe: Unpublished MSc. thesis, University of Witwatersrand, Johannesburg, South Africa, 75 p.

Neveling, J., 2004. Stratigraphic and sedimentological investigation of the contact between the Lystrosaurus and Cynognathus Assemblage Zones (Beaufort Group: Karoo Supergroup): Bulletin of the Council for Geoscience of South Africa, v. 137, p. 1-165.

Neveling, J., Hancox, P.J., Damiani, R.J., and Rubidge, B.S., 2000, Early Triassic faunas from the Lystrosaurus and Cynognathus Assemblage Zones, South Africa: Journal of Vertebrate Paleontology, v. 20, p. 60A.

Neveling, J., Hancox, P.J. and Rubidge, B.S., 2005, Biostratigraphy of the lower Burgersdorp Formation (Beaufort Group; Karoo Supergroup) of South Africa – implications for the stratigraphic ranges of early Triassic tetrapods: Palaeontologia Africana, v. 41, p. 81-88.

Olson, P.E., and Galton, P., 1984, A review of the reptile and amphibian assemblages from the Upper Triassic of Lesotho: Annals of the South African Museum, v. 67, p. 227-329.

Raath, M.A., 1977, The anatomy of the Triassic theropod Syntarsus shodesiensis (Saurischia: Podokosauridae) and a consideration of its biology: Unpublished PhD thesis, Rhodes University, Grahamstown, 233 p.

Raath, M.A., Kitching, J.W., Shone, R.W., and Rossouw, G.W., 1990, Dinosaur tracks in Triassic Molteno sediments: the earliest evidence of dinosaurs in South Africa? Palaeontologia africana, v. 27, p. 89-95.

Raath, M.A., Oesterlen, P.M., and Kitching, J.W., 1992, First record of Triassic Rhynchosauria (Reptilia: Diapsida) from the Lower Zambezi Valley, Zimbabwe: Palaeontologia africana,v. 29, p. 1-10.

Rauhut, O.W.M., and Hungerbuhler, A., 1998, A review of European Triassic theropods. Gaia, v. 15, p. 75-88.

Ray, S., 1999, Permian reptilian fauna from the Kundaram Formation, Pranhita-Godavari Valley, India: Journal of African Earth Sciences, v. 29, p. 211-218.

Ray, S., 2000, Endothiodont dicynodonts from the Late Permian Kundaram Formation, India: Palaeontology, v. 43, p. 375- 404.

Renaut, A.J., 2001, A re-evaluation of the cranial morphology and taxonomy of the Triassic dicynodont genus Kannemeyeri: Unpublished PhD. thesis, University of Witwatersrand, Johannesburg,191 p.

Rubidge, B.S., 1995, Biostratigraphy of the Eodicynodon Assemblage Zone: in Rubidge, B.S., Ed., Biostratigraphy of the Beaufort Group (Karoo Supergroup). SACS Biostratigraphic Series 1, 3-7.

Rubidge, B.S, 2005, Re-uniting lost continents – Fossil reptiles from the ancient Karoo and their wanderlust. South African Journal of Geology, v. 108, p. 135-172.

Rubidge, B.S., and Hancox, P.J., 2002, Reptile fossils, sequence analysis, and the nature of the Ecca-Beaufort contact: implications for the Permian development of the main Karoo Basin, South Africa. Abstracts Gondwana 11, correlationas and connections. University of Canterbury, Christchurch, New Zealand.

Rubidge, B.S., Modesto, S.P., Sidor, C., and Welman, J., 1999, Eunotosaurus africanus from the Ecca-Beaufort contact in Northern Cape Province, South Africa – implications for Karoo Basin development: South African Journal of Science, v. 95, p. 553-555.

Rubidge, B.S., Hancox, P.J., and Catuneanu, O., 2000, Sequence analysis of the Ecca-Beaufort contact in the southern Karoo of South Africa: South African Journal of Geology, v. 103, p. 81-96.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 167

Schoch, R.R., 1999a, Comparative osteology of Mastodonsaurus giganteus (Jaeger, 1828) from the Middle Triassic (Lettenkeuper, Longobardian) of Germany (Baden-Wurtemberg, Bayern, Thuringen): Stuttgarter Beitrage zur Naturkunde Seri B (Geologie und Palaontologie), v. 278, p. 1-175.

Schoch, R.R., 1999b, The Triassic labyrinthodonts from Germany. Zbl. Geol. Palaont. Teil I, 1998, heft, 7-8, p. 629-650.

Schoch, R.R., and Milner, A.R., 2000, Stereospondyli. Stem-stereospondyli, Rhinesuchidae, Rhytidostea, Trematosauroidea, Capitosauroidea:.in Wellnhofer, P., Ed., Encyclopedia of Paleoherpetology 3B. Stuttgart, Fischer, 203 p.

Shishkin, M.A., Ochev, V.G., Lovzovsky, V.R., and Novikov, I.V., 2000, Tetrapod biostratigraphy of the Triassic of Eastern Europe: in Benton, M.J., Shishkin, M.A., Unwin, D.M., and Kurochkin, E.N., Eds., The age of dinosaurs in Russia and Mongolia: Cambridge University Press, p. 120-139.

Shishkin, M.A., Rubidge, B.S., and Hancox, P.J., 1995, Vertebrate biozonation of the Upper Beaufort Series of South Africa - A new look on correlation of the Triassic biotic events in Euramerica and southern Gondwana: Sixth Symposium on Mesozoic Terrestrial Ecosystems, China Press Beijing, p. 39-41.

Shishkin, M.A., Rubidge, B.S., and Kitching, J.W., 1996, A new Lydekkerinid (Amphibia, Temnospondyli) from the Lower Triassic of South Africa – implications for evolution of early capitosauroid cranial pattern: Philosophical Transactions of the Royal Society of London B, v. 351, p.1635-1659.

Shubin, N.H., Olsen, P.E. and Sues, H.D., 1997, Early Jurassic small tetrapods from the Mc Coy Brook Formation of Nova Scotia, Canada: in Fraser, N.C. and Sues, H.D., Eds., In the shadow of the dinosaurs – early Mesozoic tetrapods. Cambridge University Press, p. 251-270.

Smith, R.M.H., and Keyser, A.W., 1995, Biostratigraphy of the Tapinocephalus Assemblage Zone: in Rubidge,B.S., Ed., SACS Biostraigraphic Series No. 1, Council for Geoscience, Pretoria, p. 8-17.

Smith, R.M.H., 2000, Sedimentology and taphonomy of Late Permian vertebrate fossil localities in southwestern Madagascar: Palaeontologia africana, v. 36, p. 25-41.

Smith, R.M.H.and Ward, P.D., 2001, Pattern of vertebrate extinctions across an event bed at the Permian-Triassic boundary in the Karoo Basin of South Africa: Geology, v. 29, p.1147-1150.

Sues, H-D., and Boy, J.A., 1988, A orocynosuchid cynodont from central Europe: Nature, v. 331, p. 523-524.

Sues, H-D., Clark, J.M., and Jenkins, F.A., 1997, A review of the Early Jurassic tetrapods from the Glen Canyon Group of the American Southwest: in Fraser, N.C., and Sues, H.D., Ed., In the shadow of the dinosaurs – early Mesozoic tetrapods. Cambridge University Press, p. 284-294.

Visser, J.N.J., 1984. A review of the Stormberg Group and Drakensberg volcanics in southern Africa. Palaeontologia africana, v. 25, p. 5-27.

Visser J.N.J., 1991, Geography and climatology of the Late Carboniferous to Jurassic Karoo basin in south–western Gondwana: Annals of the South African Museum, v. 99, p. 415-431.

Walker, A.D., 1969, The reptile fauna of the Lower Keuper Sandstone: Geological Magazine, v. 106, p. 470-476.

Warren, A.A., 1991, Australian fossil amphibians: in Vickers-Rich, P., Monaghan, J.M., Baird, R.F., and Rich, T.H., Eds., Vertebrate Paleontology in Austalasia. Melbourne, Monash University, p. 569-590.

Welman, J., Loock, J.C., and Rubidge, B.S., 2001, New evidence for diachroneity of the Ecca-Beaufort contact (Karoo Supergroup, South Africa): South African Journal of Science, v. 97, p. 320-322.

Yates, A.M., 2003, The first definite dinosaur from the Lower Elliot Formation (Norian: Upper Triassic): Palaeontologia africana, v. 39, p. 63-68.

Yates A.M., and Kitching J.W., 2003, The earliest known sauropod dinosaur and the first steps towards sauropod locomotion: Proceedings of the Royal Society of London B, v. 270, p. 1753-1758.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 168 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Stratigraphic architecture of aeolian strata of the Botucatu and Serra Geral formations (latest Jurassic – earliest Cretaceous), Paraná Basin, Brazil

Claiton M. S. Scherer Instituto de Geociências, UFRGS, P.O. Box 15001, CEP 91501-970, Porto Alegre - RS, Brazil, E-mail: [email protected]

The latest Jurassic - earliest Cretaceous geological record of the intracratonic Paraná Basin comprises a thick succession of aeolian sandstones and volcanic rocks. The intercalation between aeolian sandstone and volcanic floods allowed the preservation of distinct aeolian genetic units. Each genetic unit represents an accumulation episode and it is bounded by supersurfaces that coincide with the base of lava floods which covered the aeolian deposits. The entire package can be subdivided into a lower genetic unit, that corresponds to aeolian sandstones preserved below initial lava flows (Botucatu Formation), and an upper set of genetic units, which comprises interlayered aeolian deposits and lava floods (Serra Geral Formation). The lower genetic unit reaches up to 400 metres in thickness. Its base is composed of ephemeral stream and sand sheet deposits that are overlain by cross-bedded sandstone ascribed to aeolian dunes. Aeolian accumulation of the lower unit was possible due to the existence of a wide topographic basin, which caused wind deceleration, and a large sedimentary inflow. The direction of sediment transport in the aeolian sandstones of the Botucatu Formation) has been interpreted on the basis of cross-strata dip directions, which allowed the reconstruction of regional wind patterns in midwestern Gondwana. Regionally, cross-strata dip directions indicate variations of palaeowind directions across the outcrop area of the Botucatu palaeoerg. The northern portion of the palaeoerg was characterized by palaeowinds blowing from the NNE, whereas the southern portion was under the influence of palaeowinds coming from the SW. A

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 169 wind convergence zone occurred at palaeolatitudes around 24° to 26°S. The identified wind pattern agrees with general circulation models for the Late Jurassic, which predict that, during the summer months (December through February), strong monsoonal winds to the S existed at low latitudes in Gondwana. Southerly winds converged at the ITCZ (at around 23° S) with winds blowing toward the northeast. The upper genetic units comprise isolated sandy bodies which occur in two different ways: (a) thin lenses (< 2 metres thick) formed by aeolian sand sheets and (b) thick sand lenses (3-15 metres) comprising cross-bedded cossets generated by migration and climbing of aeolian dunes. Accumulation of the aeolian strata was associated with wind deceleration on depressions of the irregular upper surface of lava floods. The interruption of sedimentation in the Lower and Upper Genetic Units, and related development of supersurfaces, occurred due to widespread effusions of basaltic lava. Preservation of both wind-rippled topset deposits of the aeolian dunes and pahoehoe lava imprints indicates that lava floods covered active aeolian dunes and hence protected the aeolian deposits from erosion, thus preserving the genetic units.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 170 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

The search for the P/Tr and Tr/J boundaries in South Brazil, Argentina and Uruguay

Cesar Leandro Schultz Instituto de Geociências, Universidade Federal do Rio Grande do Sul, Brazil, PPGGeo/UFRGS, [email protected]

The P/Tr boundary is related to the greatest mass extinction whenever occurred on Earth. One of the main causes of such gigantic biotic crisis was the tectonic activity that culminated with the formation of the Pangaea. Therefore, the study of the rocks around this boundary is important as for tectonic as for paleontological reasons. Regarding to tectonics, at this epoch, in South America, the orogenic Gondwanides system generated great structural changes in the sedimentary basins of southern South America and Africa, which, by its turn leaded to changes in the depositional sedimentary sequences of these basins, with the occurrence (or not) of unconformities. Besides, huge faunistic and floristic changes are observed in these sequences, but not exactly coeval among the different basins. The Tr/J limit is equally globally marked by great changes in the floral and faunal contents of the fossil record, specially regarding to the terrestrial tetrapods, with the beginning of the Age of the Dinosaurs. The apparent temporal coincidence between such faunal changes and the Pangaea break-apart, as occurred in the case of P/Tr boundary, brings to light the cause-effect relationship between these events. Nevertheless, neither the P/Tr nor the Tr/Jr boundaries are precisely delimited, even in bio or chronostratigraphic terms, in south Brazil, Uruguay or Argentina. So, it is necessary, at first, a cooperative effort in order to obtain reliable chronostratigraphic data from all rock units around these boundaries in southern South American basins, as well as an improvement of biostratigraphic data, in order to establish – or not – a direct correlation between

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 171 tectonic, sedimentary and fossiliferous changes, as well as their temporal correspondence – or not – among such changes in the different basins. Only after these steps it would be possible to establish the precise position of P/Tr and Tr/J limits in southern South America and then correlate them with global data regarding to such limits, in order to know if all them fit in time or are influenced – and time-avereged – by regional tectonics.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 172 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Environmental impact of the End-Permian Extinction on the Karoo Basin of South Africa

Roger M H Smith Iziko South African Museum, Cape Town, South África, [email protected]

Abstract The extinction of vertebrates around the Permo-Triassic boundary, approximately 251 Mya, has long been regarded as a gradual event occurring over hundreds of thousands to millions of years. Our new field investigations of fluvial strata in the central and southern Karoo Basin of South Africa have revealed the presence of an event bed coinciding with a mass extinction of terrestrial fauna and flora over a much shorter period. This event bed is a laminated red mudrock within a sedimentary sequence that reflects a rapid transition from high to low sinuosity river channel systems. Not only is the pattern of species extinction and recovery similar in all studied sections, but it occurs in a similar sequence of sedimentary facies. The changes in fluvial style from meandering to low-sinuosity braided channels, combined with the onset of rubification of the floodplain soils and changes in taphonomic style of the vertebrate fossils are interpreted as the result of climatic drying causing de-vegetation and a corresponding decrease in bank strength. Thus, it is proposed that the extinction represents a faunal changeover from wet floodplain with Glossopteris flora supporting a Dicynodon -dominated fauna to dry floodplain with drought-tolerant horsetails supporting communities dominated by Lystrosaurus. This changeover is consistent with a relatively sudden, possibly catastrophic drought event of 50,000-100 000 year duration at the end of the Permian that was most probably caused by atmospheric pollution emanating from the Siberian flood basalts.

Introduction Isotope and palaeomagnetic stratigraphy have been used to position Permo-Triassic boundary (PTB) in the main Karoo Basin of South Africa and correlate the study sections with terrestrial PTB sequences in other parts of the world (Ward et al., 2005, Figure 1). The Karoo PTB coincides with a major extinction episode recorded in the fossil record of the Glossopteris flora (Retallack et al., 2004) and the therapsid-dominated tetrapod fauna (Smith and Ward, 2001). This study uses field observations of the sedimentary facies, and taphonomy of in situ fossils of two well-exposed PTB sequences in the southern Karoo Basin

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 173 to provide evidence of environmental aridification that may have brought about the breakdown of the Late Permian terrestrial ecosystems in southern Gondwana.

Figure 1 - A. Map of South Africa, showing the location of Bethulie and Graaff-Reinet districts where detailed lithostratigraphical sections were conducted. B. Summary of the litho- and biostratigraphy for the southern and central Karoo Basin indicating the stratigraphic positions of the PTB described in this report. Abbreviations: C.T., Cape Town; D, Durban; JHB, Johannesburg; SACS, South African Committee for Stratigraphy.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 174 Pattern of tetrapod extinctions at the Karoo PTB. A total of 348 identifiable in situ vertebrate fossils were collected from 52 discrete stratigraphic levels spanning the uppermost Permian and lowermost Triassic strata in two widely separated study sections (Figure 2). These have so far yielded a minimum of 23 species, 9 species found in Permian strata, and 12 from Triassic strata and. Only 4 have been proven to cross the boundary however ghost lineages of some of the early Triassic recovery fauna should incude more survivors. In general the extinction could be described as “ecologically stepped” with the disappearance of small herbivores (Diictodon, Pristerodon) and small carnivores (Cyonosaurus) before the larger forms Rubidgea, Theriognathus, Pelanomodon and Dicynodon. The End-Permian extinction in the central Karoo basin lasted some 100 000 years (Smith and Ward, 2001) during which time there was a gradual extinction of small herbivorous dicynodonts and their gorgonopsian predators before the main pulse of extinction that involved medium and large herbivores and carnivores. This pattern suggests that the smaller ground foraging animals, feeding on the undergrowth of ferns and clubmosses, disappeared before larger browsing fauna feeding on Glossopteris shrubs and trees. This is in keeping with the interpreted onset of drought conditions. The fact that the medium- and large- sized Lystrosaurus arrived in the basin and seemingly flourished as the Dicynodon fauna began to fade, indicates that it was somehow pre-adapted to survive the worsening drought conditions. Field data demonstrates a 69% mass extinction of Late Permian terrestrial vertebrates lasting some 300Ky terminating at the PTB, followed by a lesser extinction event (31%) approximately 160Ky later involving four survivor taxa that crossed the PTB (Figure 2). The Early Triassic recovery fauna is fully developed within 15 stratigraphic metres of the boundary and omprises proterosuchian archosauromorphs (Proterosuchus), small amphibians (Micropholis, Lydekkerina), small procolophonoids (“Owenetta” kitchingorum, Procolophon), medium-sized dicynodonts (Lystrosaurus) and small insectivorous cynodonts (Progalesaurus, Galesaurus, Thrinaxodon). Detailed biostratigraphic analysis shows that Lystrosaurus curvatus and L. maccaigi lived together on the Karoo floodplains immediately before the extinction event. L. maccaigi did not survive into the Triassic in South Africa. L. curvatus survived, but did not flourish and soon became extinct. Two new species of Lystrosaurus, L. murrayi and L. declivis appeared in the Early Triassic. It is possible that L. murrayi and L. declivis occupied different niches to L. maccaigi and L. curvatus, and had special adaptations that were advantageous in an Early Triassic environment. We suggest that L. maccaigi may be used as a biostratigraphic marker to indicate latest Permian strata in South Africa and that, in support of previous proposals, the genus Lystrosaurus should not be used as a sole indicator of Triassic-aged strata. Our field data also shows that L. curvatus may be regarded as a biostratigraphic indicator of the P—T boundary interval.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 175

Figure 2 - Range chart of terrestrial vertebrate fossils found in situ in the PTB sequences of the southern and central Karoo Basin of South Africa and grouped into discrete extinction, survival and recovery faunas. The conversion of stratigraphic metres to the time estimates used in the text are based on floodplain accretion rate of 2mm/y with an average compaction ratio of 50% and a compensation factor of x5 to account for periods of non- deposition (pedogenesis) and erosion.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 176 The characteristic spade-shaped maxilla with shock- resistant sutures was most likely the key to the survival of Lystrosaurus (Botha and Smith, 2007) It had the ability to continue grazing throughout the drought, probably on the tough equisetalian “reeds” that remained growing in the moist areas around ponds and along the increasingly ephemeral floodplain channels. It appears that Dicynodon and associated herbivores were unable to digest the fibrous stems and leaf whorls of the horsetails efficiently enough to ensure their survival.

Karoo PTB taphonomy and palaeoenvironments. The largest diversity of Late Permian taxa is found in a massive grey mudrock facies composed of stacked metre-thick tabular beds of structureless bluish-grey and greenish-grey siltstone with interbedded erosionally-based, fine-grained sandstone bodies up to 6m-thick. This facies typifies all but the upper few tens of meters of the 200- to 300m thick Dicynodon lacerticeps zone (Figure 3). Vertebrate fossils are relatively rare, disarticulated post-cranial bones, lower jaws and isolated skulls mainly of the medium and large dicynodonts Dicynodon and Pelanomodon. The small dicynodont Pristerodon disappears at the top of this unit, thus well below the P/T boundary. The traditional Triassic indicator fossil, Lystrosaurus spp, first appears in this facies some 41m below the P/T boundary. The disappearance of most of the Late Permian taxa (Diictodon, Cyonosaurus, Rubidgea, Pelanomodon, Theriognathus) occurs in the succeeding massive maroon/grey mudrock facies. Varying from 12 to 20m in thickness this succession of rubified, pedogenically modified siltstone contains interbedded, thin tabular sandstone bodies displaying distinctively "gullied" basal contacts. This facies is terminated by a metre-thick maroon mudrock capped by a single "horizon" of large brown-weathering calcareous nodules that coincides with the last occurrence of Dicynodon lacerticeps, the index fossil of the latest Permian biozone. In the Karoo Basin, this layer of brown-weathering nodules is a lithological expression of the palaeontologically defined P/T boundary. The P/T boundary is immediately overlain by 3-5m of distinctively laminated maroon mudrock made up of thinly-bedded dark reddish-brown and olive-grey siltstone/mudstone couplets (Figure 3). This laminated facies is a lithologically distinctive unit in all measured sections. Intensive prospecting has failed to locate any vertebrate fossils in this interval although sub horizontal siltstone filled cylinders, resembling callianassid shrimp burrow casts, make their first appearance in this facies. In the Bethulie section, this facies coincides with a stable carbon isotope anomaly that has been identified as spanning the P/T boundary and it may therefore be described as an event bed associated with the End Permian mass extinction.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 177

Figure 3 - Summary sedimentological log of the PTB sequence compiled from the study sections in the Bethulie District with interpreted palaeolandscape reconstructions. Vertical scale in metres.

The Early Triassic faunal recovery takes place in the overlying facies composed of massively-bedded maroon siltstone with minor sandstone sheets. These thin (<2m) vertically-accreted sandstone bodies have distinctive "gullied" basal scours. Horizons of sand-filled mudcracks and large brown calcareous nodules occur within the maroon siltstone beds. Abundant curled-up Lystrosaurus skeletons, thickly enveloped in dark reddish brown nodular material occur in this interval. In our study sections most of the Triassic fauna made an appearance within 50m above the boundary. They appeared in the following sequence: Proterosuchus (10m), Micropholis (14m), Galesaurus (22m), Thrinaxodon (31m), Lydekkerina (51m) and Procolophon (94m). The highest facies sampled in our study comprised the basal portion of the Katberg Formation, a multistoried conglomeratic

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 178 sandstone composed of numerous vertically-stacked units of olive-grey, fine-grained sandstone containing numerous scoured disconformities. Many of these erosion surfaces are lined with lenses and stringers of intraformational mud pebble and pedogenic nodule conglomerate. The taphonomic signature of this facies is again dominated by numerous articulated adult Lystrosaurus . However, here they are accompanied by a few monospecific bonebed "pockets" each measuring approximately 2.5x1.5m and composed of several completely disarticulated and jumbled-up sub-adult Lystrosaurus skeletons strength leading to increased floodplain sedimentation rates. The Karoo PTB facies sequence is interpreted as a relatively rapid change in fluvial landscape from an alluvial plain traversed by a few large highly meandering rivers with expansive lowland floodplains (massive dark grey mudrock) through a transitional stage when the rivers straightened and widened and branched into a distributary channel network that scoured the now abandoned floodplains (massive maroon siltstone). As the sediment load increased these channels continued to widen and formed in-channel bars that eventually separated the flow into a braidplain of interconnected sand-dominated ephemeral channels (conglomeratic sandstone). During deposition of the massive maroon siltstone facies there was an apparently synchronous depositional event that co-incided with the extinction of Dicynodon- the last of the Permian dicynodonts to disappear from the Karoo basin. This resulted in the accumulation of up to 5 metres of red laminated mudrocks (maroon laminites) that show evidence of shallow standing water with periodic sub-aerial exposure and desiccation. This is interpreted as an interval when soil formation almost ceased over large parts of the Karoo floodplains. Periodic flooding deposited sand/mud couplets that show little post depositional colonization by either animal or plant life except for a calliannassid-like burrowing arthropod.

Evidence for drought at the Karoo PTB. The calcic palaeosols reflect not only a change in rainfall regime but also an increase in mean annual temperature in the early Triassic resulting in widespread reddening of the floodplain mudrocks. The conclusion drawn from sedimentological and taphonomic evidence of waterhole bone accumulations is that for a period following the disappearance of Dicynodon, the central Karoo basin was subject to an increasingly more unreliable and stormy, possibly monsoonal, rainfall regime combined with an increase in mean annual temperature which effectively allowed only drought-tolerant flora, and their dependant fauna, to survive into the Early Triassic.

Possible causes of the End-Permian mass extinction The changes in fluvial style in the Karoo Basin at the end of the Permian were initially thought to be triggered by a pulse of thrusting in the southerly source area, which brought

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 179 about rapid progradation of a large sandy braided fan system (the Katberg Fm.) into the central parts of the basin (Hiller and Stavrakis, 1984). Ward et al. 2000) suggested an alternative that better fits the observed palaeoclimatic changes. They proposed that the switch in fluvial regime was a consequence of the extinction event rather than the cause. The global aridification and concomitant de-vegetation of the continental interiors caused unstable channel banks and increased run-off that is reflected by the onset of gulleying of the floodplains and the switch from high to low sinuosity channel pattern. Stable isotopic analyses of pedogenic nodules and teeth taken from embedded fossils through the extinction interval at the Bethulie section revealed an anomalous negative excursion within the boundary laminated mudstones (Macleod et al. 2000, Ward et al. 2005, Figure 4). This anomaly is interpreted as having been caused by a relatively rapid release of methane or CO2 into the atmosphere over a period of some 150Ky. The greenhouse effect from these emissions, whether they originated from the Siberian volcanism (Renne et al 1995,) or oceanic overturn (Morante 1996,Knoll et al 1996), could account for the observed global aridification of continental interiors.

References Botha, J. and Smith, R.M.H., 2007, Lystrosaurus species composition across the Permian/Triassic boundary of South Africa: Lethaia, v. 40, p. 125-137

Hiller, N., and Stavrakis, N, 1984, Permo-Triassic fluvial systems in the southeastern Karoo Basin, South Africa: Palaeogeog., Palaeoclimatol., Palaeoecol.: v. 45, p.1-22.

Knoll, A.H., Bambach, R.K, Canfield, D.E., and Grotzinger, J.F., 1996, Comparative Earth history and the late Permian mass extinction: Science, v. 273, p. 452-457.

Macleod, G. K., Smith, R. M. H., Koch, P. L., and Ward P. D.,2000, Timing of mammal-like reptile extinctions across the Permian Triassic boundary in South Africa: Geology, v. 28, p. 227- 230

Morante, R., 1996, Permian and Early Triassic isotopic records of carbon and strontium in Australia and a scenario of events about the Permian-Triassic boundary: Historical Biology, v. 11, p. 289-310.

Renne, P.R., Zheng, Z.C., Richards, M.A., Black, M.T., and Basu, A.R., 1995, Synchrony and causal relations between Permian-Triassic boundary crisis and Siberian flood volcanism: Science, v. 269, p. 1413-1416.

Retallack, G.J., Smith, R.M.H., and Ward, P.D., 2003), Ecosystem extinction across the Permian-Triassic boundary in the Karoo Basin of South Africa: Bulletin of the Geological Society of America, v. 115, p. 1133- 1152.

Smith, R.M.H..and Ward, P.D., 2001, Pattern of Vertebrate Extinctions across an Event Bed at the Permian/Triassic Boundary in the Main Karoo Basin of South Africa: Geology, v. 29, p. 1147-1150.

Ward, P.D., Botha, J., Buick, R., de Kock , M.O., Erwin, D.H. Garrison, G.H., Kirschvink, J.L., Smith, R.M.H., 2005, Abrupt and Gradual Extinction Among Late Permian Land Vertebrates in the Karoo Basin, South Africa: Science, v. 307, p.709-714

Ward, P.D.,. Montgomery, D.R, Smith, R.M.H., 2000, Altered river morphology in South Africa related to the Permian-Triassic Extinction: Science, v. 289, p.1741-1743.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 180

Figure.4 - The combined soil nodule δ13C carb record and individual lithologic delts δ13 C org records for 4 sections in the Karoo Basin. The two sections at left have been

strongly affected by contact metamorphism of dolerite intrusions. Carlton Heights and Lootsberg Pass, however , have similar negative δ13 C org excursions within unit II, followed by

an increase in delta δ13 C org values in Units III and IV. The curve at the far right is combined δ13C carb data from pedogenic nodules from Carlton Heights and Lootsberg locatities.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 181 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

An integrative analysis on advances and perspectives of the Pennsylvanian and Permian palynostratigraphy in the Paraná/Chacoparaná Basin (Brazil, Argentina and Uruguay)

Paulo A. Souza¹, Maria M. Vergel², Ángeles Beri³ ¹ Universidade Federal do Rio Grande do Sul, Porto Alegre – RS – Brazil, [email protected] ² Insugeo-CONICET, Universidad Nacional de Tucumán, Tucumán – Argentina, [email protected] ³ Facultad de Ciencias, Universidad de la República, Montevideo – Uruguay, [email protected]

Abstract Pennsylvanian and Permian palyniferous sedimentary beds of the Paraná/Chacoparaná Basin are known from Argentina, Brazil and Uruguay, and have allowed distinct biostratigraphical zonal schemes. Pennsylvanian palynozones are recognized in the northeastern portion of the Brazilian Paraná Basin (Ahrensisporites cristatus and Crucisaccites monoletus palynozones) and within the Chacoparaná Basin in Argentina (Potonieisporites- Lundbladispora playnozone). Pteridophytic spores and monosaccate pollen grains related to the Cordaitales and Coniferales are dominant in these palynozones. The Permian Vittatina costabilis and Lueckisporites virkkiae palynozones of the Paraná Basin are adequately correlated with the Cristatisporites and Striatites zones in Argentina, as well as with the Cristatisporites inconstans-Vittatina subsaccata and the Striatoabieites anaverrucosus-Staurosaccites cordubensis zones of the Paraná Basin in Uruguay, respectively, although differences be found. A greater palynological diversity characterizes the older Permian palynozones (Vittatina costabilis, Cristatisporites and Cristatisporites inconstans-Vittatina subsaccata), marked by the presence of several taeniate pollen grains related to the incoming of the Glossopterids, as well as by a variety of other groups (bisaccate, polyplicate, pollen grains). Among the spores, cingulizonate species are very common and Converrucosisporites confluens seems to be a significant marker for correlation. The younger Permian palynozones (Lueckisporites virkkiae, Striatites and Striatoabieites anaverrucosus-Staurosaccites cordubensis) show similarities on the frequency of palynologic groups, such as dominance of taeniate pollen grains, which reach up to 80% of certain assemblages, and the lower frequency of spores, which are rare or scarce within certain levels. Radiometric data have been obtained in these last years to these deposits, and have contributed for the age calibration of these palynozones with the international geological scale. These palynological,

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 182 radiometric data and their significance within the Occidental Gondwana context are summarized and analysed herein.

Introduction The Paraná/Chacoparaná Basin is a large intracratonic basin comprising a thick and widespread sedimentary-magmatic sequence, which covers about 1,700,000 km2 in Central- Eastern of South America, mainly in Brazil, where is known as “Paraná” Basin, with ca. 1,400,000 km2. This basin is also distributed in Argentina, Paraguay, and Uruguay, where it is known as “Chacoparaná” (or Chacoparanense) Basin (Figure 1). While this basin exhibits expressive exposures (outcrops) in Brazil and Uruguay, all strata in Argentina occur in subsurface. Fossil animals, plants and palynomorphs are recorded at distinct stratigraphical levels within its Pennsylvanian/Permian interval. However, good guide-fossils, such as cosmopolitan marine invertebrates, are scarce, preventing correlation with the international stratigraphic scale. Furthermore, few radiometric data are available. In this context, palynology seems to be the most efficient biostratigraphic tool for this interval in this basin, because of the abundance, diversity and widespread distribution of spore-pollen assemblages. The main purpose of this work is to show the palynostratigraphic schemes proposed for this basin, to discuss its similarities and discrepancies, and its geochronological significance.

Figure 1 – Distribution of the Paraná/Chacoparaná Basin in South America.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 183 The Brazilian palynostratigraphy Four interval zones characterize the palynological succession of the Pennsylvanian/Permian strata of the Paraná Basin in Brazil, which are namely (in ascending order): Ahrensisporites cristatus (AcZ), Crucisaccites monoletus (CmZ), Vittatina costabilis (VcZ), and Lueckisporites virkkiae (LvZ) Zones. These palynozones are effect of improvements from formal palynostratigraphic proposals erected in different portions of the basin (Souza and Marques-Toigo, 2003, 2005; Souza, 2006), with close correlation to the informal and classical intervals proposed by Daemon & Quadros (1970). The oldest Carboniferous biozone (AcZ) occurs within the basal portion of the Itararé Subgroup, and was assigned as late Bashkirian to Kasimovian in age. The overlying CmZ biozone is dated as late Pennsylvanian (Kasimovian to Gzhelian), and ranges approximately from the top of the lower portion to the middle portion of the Itararé Subgroup. The VcZ extends from the top of the Itararé Subgroup into the post-glacial interval (uppermost Rio Bonito/lowermost Palermo formations), and is subdivided in two subzones (Protohaploxypinus goraiensis and Hamiapollenites karrooensis Subzones). The lower limit of the LvZ occurs in the uppermost Rio Bonito/lowermost Palermo formations, and its upper limit is marked by the last occurrences of Lueckisporites, within basal portions of the Rio do Rastro Formation. The VcZ and LvZ are regarded as Permian in age, related to the Asselian/Artinskian (VcZ) and the Artinskian/Wuachiapingian (LvZ).

The Argentinean palynostratigraphy According to Russo et al. (1980) and Vergel (1993), three assemblage zones are defined to the Pennsylvanian/Permian sequences of the Chacoparaná Basin in Argentina. In ascending order they are named Potonieisporites-Lundbladispora (P-L), Cristatisporites (C) and Striatites (St) Zones. The oldest miospore assemblages assigned to the P-L biozone are from the Sachayoj and Charata formations (in the Alhuampa Sub-Basin), and from the basal Ordóñez Formation (San Cristobal-Las Breñas Oriental Sub-Basin). In the Alhuampa Sub- Basin, P-L biozone has a basal level (lower interval) with miospores common to the western Argentinean Paganzo Basin, and is assigned as Pennsylvanian (late Bashkirian to Gzhelian) in age. The C biozone, regarded as Asselian to Artinskian in age, is subdivided into three intervals, namely lower, middle and upper. The lower interval occurs in the Charata and Chacabuco formations (Alhuampa Sub-Basin), while the others ones occur within the Ordóñez Formation (San Cristobal-Las Breñas Oriental Sub-Basin). The youngest biozone (St) is recorded only in the San Cristobal-Las Breñas Oriental Sub-Basin (upper portion of the Ordóñez and Victoriano Rodríguez formations), and is assigned to the late Cisuralian- Guadalupian.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 184 The Uruguayan palynostratigraphy Beri et al. (2004) proposed an informal zonation for the Paraná/Chacoparaná Basin in Uruguay based on palynological assemblages assigned to the Early to Late Permian. They were named as (in ascending order): Cristatisporites inconstans-Vittatina subsaccata (IS) Zone, which occurs from the San Gregorio Formation, Tres Islas and Frayle Muerto formations, and the Striatoabieites anaverrucosus-Staurosaccites cordubensis (AC) Zone, recorded from the Mangrullo Formation. However, taking into account new radiometric data those biozones could be assigned an age between the Cissuralian and the Guadalupian, as discussed below. Pennsylvanian palynological assemblages seem to be absent in Uruguay.

Comparison and general features of the palynozones According to Stephenson et al. (2003, p. 473-474) the accuracy in palynological correlations outside the study region and with the international stratigraphical stages is hampered by several factors. These problems are aggravated in Gondwana areas due the phytogeographical provincialism (mainly in Permian times), and because most of the global stratotypes are defined taking into account invertebrates of Euroamerican Realm affinity, which are generally absent in Gondwanan basins. Besides, there are scarce radiometrical datings available for Gondwanan regions. However, a tentative of correlation is made herein, as an initial contribution to future detailed investigations. Pennsylvanian palynozones are recognized in the northeastern portion of the Brazilian Paraná Basin (AcZ and CmZ palynozones) and within the Chacoparaná Basin in Argentina (P-L). Pteridophytic spores and monosaccate pollen grains related to the Cordaitales and Coniferales are dominant in these palynozones, while taeniate pollen grains are scarce (e.g., Protohaploxypinus, Vittatina). The Permian palynozones were erected based on different criteria, reflecting variation on the distribution of taxa and limitations on the sampling (cutting, cores). However, both Permian palynozones of the Paraná Basin, VcZ and LvZ, are adequately correlated with the C and St Zones of the Chacoparaná Basin in Argentina, as well as, with the IS Zone and the AC Zone of the Paraná Basin in Uruguay, although few differences be found. A greater palynological diversity characterizes the early Permian palynozones (VcZ, C, IS), marked by the presence of several taeniate pollen grains related to the incoming of the Glossopterids, as well as by a variety of other groups (bisaccate, polyplicate, pollen grains). Among the spores, cingulizonate species are very common and Converrucosisporites confluens seems to be a significant marker for correlation. The LvZ, St and AC zones show similarities on the frequency of palynologic groups, such as dominance of taeniate pollen grains, which reach up to 80% of the assemblages, and the lower frequency of spores, which are rare or scarce within certain levels. Species of Lueckisporites, Striatopodocarpites, Striatoabieites, Protohaploxypinus, Staurosaccites,

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 185 Vittatina and Weylandites are very common, as well as those linked to bisaccate pollen grains, such as Alisporites and Limitisporites. These data reinforce correlations previously proposed (e.g., Vergel, 1993) and are presented in the Figure 2, which summarizes their main features.

Ages and calibration with radiometric data Radiometrical datings are scarce for the Gondwanan Paleozoic strata. Besides, marine faunal assemblages with index species suitable for intercontinental correlation are missing. Then, only the meagre data on absolute ages are discussed herein, reinforcing remarks presented by Césari (2007). None radiometrical data is available from the AcZ and CmZ (Paraná Basin) and the P-L (Chacoparaná Basin) palynozones. Then, their ages are supported by palynological correlation with late Carboniferous strata in Argentina (e.g., Paganzo Basin) and Australia.

Figure 2 – Correlation of the Pennsylvanian and Permian palynozones of the Paraná/Chacoparaná Basin (geochronology is according to Gradstein et al., 2004).

In Australia, Gondwanan palynostratigraphic schemes have been established and partly calibrated with marine faunas. There, the Pseudoreticulatispora confluens and Striatopodocarpites fusus Zones were assigned to the early Early Permian:

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 186 Asselian/Tastubian (Archbold et al., 2004; Césari, 2007). These palynozones are equivalent, in the studied basins, to the VcZ (Paraná Basin) and the C Zone (Chacoparaná Basin), taking into account the stratigraphic ranges of the two mentioned species. Then, the underlying biozones (AcZ, CMZ and P-L palynozones) must be older than the Permian. Furthermore, absolute ages obtained from the Dwyka Group in Africa (Bangert et al., 1999), the regional equivalent of the Itararé Subgroup, where the AcZ and CmZ were recognized, support a Pennsylvanian age for most of the glacial event in Western Gondwana, represented by the Itararé Subgroup in Brazil, which bears the AcZ and the CmZ palynozones. The Fusacolpites fusus-Vittatina subsaccata Assemblage Biozone of western Argentina occurs within a sedimentary sequence equivalent to a sedimentary one bearing interbedded basalts dated as 302 ± 6 Ma and 288 ± 7 Ma. This biozone is tentatively correlated with the lower C biozone of the Argentinean Chacoparaná Basin and the VcZ of the Paraná Basin, corroborating an Early Permian age for both palynozones (see Césari, 2007). An absolute age of 267.1 ± 3.4 Ma was obtained from a tonstein interbedded in the upper coal seams of the Candiota Coalfield (Rio Bonito Formation, Brazilian Paraná Basin) by Matos et al. (2001), where the VcZ was recognized. However, a new zircon U/Pb SHRIMP dating for the same basin, has revealed an older age for the Irati Formation (278 ± 2.2 Ma), as stated by Santos et al. (2006). Hence, the datings so far obtained from both units are controversial, since the Rio Bonito Formation is stratigraphically lower than the Irati Formation. Geochronological analysis based on Gondwana sections in Africa reveals significant results. An absolute age of 270 Ma was obtained from tuff beds of the Collingham Formation, in South Africa, which overlies the Whitehill Formation. The latter is the regional equivalent of the Irati Formation in the Paraná Basin, which bears the Lueckisporites virkkiae Interval Zone. Considering that there is no marked diachronism between these two lithostratigraphical units, the Irati Formation should likewise be regarded as older than 270 Ma, reinforcing the absolute age obtained to this unit by Santos et al. (2006). These data corroborate intercontinental correlations established from the “Eurydesma” marine fauna throughout Gondwana, regarded as Early Permian (Asselian/Sakmarian) in age, on the top of the glacial sequence (Itararé Subgroup). In the Brazilian Paraná Basin, equivalent marine elements occur in the upper Itararé Subgroup, where the VcZ occurs. Recent datings preliminarily presented to the Paraná Basin in southmost Brazil (Guerra-Sommer et al., 2005) and in Uruguay (Rocha-Campos et al., 2006) deserve more detailed analysis when fully published.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 187 Conclusions Paleogeographic reconstructions show an equivalent latitudinal position to these basins during the Pennsylvanian and part of the Permian times, reflecting common paleoclimate conditions and floral composition. As result, the general features of each studied palynozone, established on different portions of the Paraná and the Chacoparaná basins, are relatively common, allowing reasonable correlation. Discrepancies are also verified, and must be interpreted as local paleoecological variations, as well as due the geological record, which is more complete in certain parts of these basins. Generally, for Gondwana areas, polyplicate pollen grains are more common from the Permian strata. However, Vittatina subsaccata occurs in the Pennsylvanian P-L palynozone, together with Hamiapollenites fusiformis, which are recorded only from the Permian VcZ of the Brazilian Paraná Basin. Furthermore, species of Lueckisporites are recorded within uppermost parts of the Uruguayan IS Zone. These appearances must be related to the first records of these genera, due few specimens are found under the limit between the Pennsylvanian P-L palynozone and the Permian C Zone (Argentina), as well as the IS and the AC Zone (Uruguay), respectively. The limit between the palynozones studied has geological importance and suggests paleoclimate changes, from the cold conditions of the Pennsylvanian/lowermost Permian to higher temperatures, which characterize the Middle and Upper Permian deposits. In general, Pennsylvanian palynozones are characterized by dominance of spores and monosaccate pollen grains, while taeniate pollen grains are infrequent or absent. Taeniate pollen and polyplicate pollen grains are more common from the Permian palynozones, especially from the Artinskian, when they became dominant. In the Permian, the palynofloras show great changes, most part due the appearance of Glossopterids allied pollen grains, besides the disappearance of Pennsylvanian taxa. New radiometric data have been obtained in these last years to these basins, and have contributed for the age calibration of these palynozones with the international geological scale. However, controversial results are verified, showing the necessity of systematic works in this theme.

Acknowledgements This work was partially supported by research grants from CNPq (Project 474153/2004-5).

References Archbold, N.W., Cisterna, G.A. and Simanauskas, T. 2004. The Gondwana Carboniferous-Permian boundary revisited: new data from Australia and Argentina: Gondwana Research, v. 7, n. 1, p. 125-133.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 188 Beri, A., Gutiérrez,P., Cernuschi, F. and Balarino, L., 2004, Palinoestratigrafía del Pérmico de la Cuenca Paraná, Uruguay: XI Reunião de Paleobotânicos e Palinólogos, Gramado, Brasil. Boletim de Resumos, p. 31.

Césari, S.N., 2007,. Palynological biozones and radiometric data at the Carboniferous-Permian boundary in western Gondwna: Gondwana Research, v. 11, n. 4, p. 529-536.

Daemon, R.F.and Quadros, L.P., 1970, Bioestratigrafia do Neopaleozóico da Bacia do Paraná: Congresso Brasileiro de Geologia, 24, Brasília, 1970. Anais, SBG, p. 359-412.

Gradstein, F. M. et al. (plus 38 authors), 2004, A geologic time scale 2004: Geological Survey of Canada, Miscellaneous Report 86, 1 chart.

Matos, S.L.F., Yamamoto, J.K., Riccomini, C., Hachiro, J. And Tassinari, C.C.G., 2001, Absolute dating of Permian ash-fall in the Rio Bonito Formation, Paraná Basin, Brazil: Gondwana Research, v. 4, p. 421–426.

Rocha-Campos, A.C., Basei, M.A.S., Nutman, A.P. and Santos, P.R., 2006, Shrimp U-Pb Zircon geochronological calibration of the Late Paleozoic Supersequence, Paraná Basin, Brazil: V South American Symposium on Isotopic Geology, Punta del Este, Uruguay. Short Papers, p. 298-301.

Russo, A., Archangelsky, S. and Gamerro, J.C., 1980, Los depósitos suprapaleozoicos en el subsuelo de la llanura Chaco-Pampeana, Argentina: II Congreso Argentino Paleontología y Bioestratigrafía y I Congreso Latinoamericano Paleontología, Buenos Aires, 1978. Actas, v. 4, p. 157-173.

Souza, P.A., 2006, Late Carboniferous palynostratigraphy of the Itararé Subgroup, northeastern Paraná Basin, Brazil: Review of Palaeobotany and Palynology, v. 138, p. 9-29.

Souza, P.A. and Marques-Toigo, M., 2003, An overview on the palynostratigraphy of the Upper Paleozoic strata of the Brazilian Paraná Basin: Revista del Museo Argentino de Ciencias Naturales, nueva serie, v. 5, p. 205- 214.

Souza, P.A. and Marques-Toigo, M., 2005, Progress on the palynostratigraphy of the Permian strata in Rio Grande do Sul State, Paraná Basin, Brazil: Anais da Academia Brasileira de Ciências, v. 77, p. 353-365.

Stephenson, M.H., Osterloff, P.L. and Filatoff, J., 2003, Palynological biozonation of the Permian of Oman and Saudi Arabia: progress and challenges: GeoArabia, v. 8, p. 467-496.

Vergel, M.M., 1993, Palinoestratigrafía de la secuencia neopaleozoica en la Cuenca Chacoparanense, Argentina: XII Congrès International de la Stratigraphie et Géologie du Carbonifère et Permien. Buenos Aires, 1991, Comptes Rendus, v. 1, p. 201-212.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 189 I Workshop - PROBLEMS IN WESTERN GONDWANA GEOLOGY

Serra Geral magmatism in the Paraná Basin – a new stratigraphic proposal, chemical stratigraphy and geological structures

Wilson Wildner1, Léo Afraneo Hartmann2 and Ricardo da Cunha Lopes1 1- CPRM – Geological Survey of Brazil, Rua Banco da Província, 105; 90840-030 Porto Alegre, Rio Grande do Sul, Brazil - [email protected] 2- Instituto de Geociências, Universidade Federal do Rio Grande do Sul, Avenida Bento Gonçalves, 9500; 91501-970 Porto Alegre, Rio Grande do Sul, Brazil - [email protected]

Introduction The intracratonic Paraná - Huab Basins and Serra Geral - Etendeka Large Igneous Provinces, central-eastern South America and minor Namibia remnants, contain a thick sequence of continental lava flows on the top and intrusions of Jurassic age, designated as Serra Geral Magmatic Province (e.g. White, 1908; Piccirillo and Melfi, 1988). This huge volume of magma is estimated at 790 000 km3 (Bellieni et al. 1984) and was generated in a comparatively short period of time, 137 to 127 Ma (Turner et al. 1994), linked to upwelling of the deep, hot Tristan mantle plume (Morgan, 1971; Richards et al. 1989; White and McKenzie, 1989; Ernst and Buchan, 2001), although the characterization of the plume is under debate (Marques et al. 1999; Comin-Chiaramonti et al. 2004). The lava thickness exceeds 1500 m in Pontal do Paranapanema (western São Paulo state), and cover a continental area estimated at 1.2x106 km2 (Cordani and Valdoros, 1967). Mafic and ultramafic intrusions are identified as part of a feeder conduit for the flood system. In the Atlantic Ocean continental platform, the volcanism is around 600 m thick, as observed in cores of drilling carried out 200 km east of Rio de Janeiro (Mizusaki et al., 1992). Over the past 15 years, numerous papers were published on geochemistry and geochronology settings of the extensive Paraná lava field, involving Brazilian, Italian, British and American collaboration (e.g. Piccirillo and Melfi, 1988; Peate, 1997; Turner et al. 1994). Despite this extensive geochemical investigation and the corresponding interpretations made on the lava sequence, the understanding of the three-dimensional structures, the

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 190 stratigraphy, and the geocronological evolution of the magmatic events still requires additional efforts. This paper integrates field mapping by the Geological Survey of Brazil in Rio Grande do Sul, western Santa Catarina and southwestern Paraná states with previous investigations. New data include lava chemical composition in vertical sections sampled flow by flow and in some borehole samples, geophysical gamma-ray data from drill holes and gravimetric field survey, geocronological Ar-Ar and U-Pb data.

Geological background A distinct bimodal basic-acid composition is observed in both the Paraná and

Etendeka flood basalts. Rocks are basalt and dacite to rhyolite, bimodality seen in the SiO2 gap between 60 and 64 wt%. The basalt lavas are divided (e.g., Peate, 1997) into high-Ti (Pitanga, Paranapanema and Urubici) and low-Ti types (Ribeira, Gramado and Esmeralda). Basic high-Ti type lavas occur in the north of the Paraná Basin, and the low-Ti type basalt and rhyolitic lavas occur from the center to the southeast of the basin; exception is the high- Ti Urubici type because it occurs in the southern part. Both types were affected by fractional crystallization and the low-Ti lavas were markedly affected by crustal contamination. Along with our improved knowledge of the characteristics of Serra Geral flows, we now know more about the regional stratigraphy, and recent regional studies reveal that the internal structure of the lava pile is not composed of a single sequence of different magma types. It was previously considered that the sequence started at the base with a high-Ti type and finished at the top with a low-Ti type. The sequential development shows an interfingering of distinct magma compositions, with the re-occurrence of the same magma type a few times during the geological record. This is critical to the understanding of how magmatic feeder systems and processes varied during the evolution of the province, because different chemical types were extracted at the same time, in different areas, building a complex stratigraphyc sequence. Any shift in the principal locus of magmatism can thus be linked to regional tectonic processes. The repeated occurrence of the same composition in different stratigraphic levels of the lava pile indicates that basalt magma types are not restricted to one stratigraphic unit but are very helpful in the characterization of stratigraphic units. The Paraná magma types distinguished by Peate et al. (1992) are defined on the basis of compositional characteristics irrespective of their position in the different stratigraphic levels. Stratigraphic subdivision has been made in several other basalt flood provinces (e.g., Siberian, Deccan) where continuous stratigraphic lava sections are common and can be correlated on a regional scale.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 191 Sandstone inside the volcanic pile At the base of the volcanic sequence, the intercalation of ergs and volcanic flows allowed the preservation of aeolian units, which can be subdivided into units up to 100 m thick (Scherer, 2002). These desert sandstones of Botucatu Formation interfingered with the Gramado type lavas. This is the beginning of Gondwana III supersequence (Milani, 1997), but sedimentary layers can be identified higher in the stratigraphy of the volcanic sequence. Sedimentary horizons vary between 0.5 to 25 m and may occur in the entire volcanic association such as along the lower contact of the Caxias type acid flows and at the top of the volcanic pile near the depocenter of the basin (Cordilheira Alta Formation). The sediments in the upper part of the Serra Geral Group were not deposited in a desert system, because cross-beds, ripple marks, rain drop marks and contraction fractures register the presence of water on the depositional system. These sedimentary deposits are evidence of non-volcanic hiatus, registering breaks in volcanism and allowing sufficient time for erosion and sedimentation between lava outpourings.

The triple junction model and the volcanic connection The plume-related doming of the Paraná continental area initiated the formation of a triple junction. This junction formed a rift system which led to the continental rupture.

Moreover, the evolution of magmas from CO2-alkali rich to tholeiitic depends on several factors, including the depth of melting, the reaction between the melt and the crust, the evolution of the magmatic chamber, the feeder system, and the extent of fractional crystallization. This process in the Paraná was similar to the Keweennawan province, which is composed of alkaline rocks and tholeiitic flows, connected with a rift system and an aulacogen, the remaining two arms correspond to the direction of the axis of the rift and connected to the hot spot action. Flood basalt is distributed in the axis of this arm, considered to be the center of the magma activities (Figure 1). The rift systems in the Paraná basin are associated with the Colorado and Salado Basins and the Rio Grande, Ponta Grossa, Campo Grande and Tietê Arches, in addition to the Torres-Posadas, Rio Alonzo and Guarapiara transform lateral fault systems. These are all expressions of sectors of these failed arms projected into the continent, and operated as centers of volcanic activity over 600 km in extension.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 192

Figure 1 - Tectono-magmatic features related to the South Atlantic volcanic margins. The western seaward- dipping wedge north of the Torres Arch refers to drilled basalts. Modified from: Gladzenko, et al. (1997)

The root of this magmatic system was connected to a mantle plume, over which Gondwana continent moved from south to north. The thinned continental crust ruptured and formed the two continents, opening from south to north and spreading to form the South Atlantic Ocean. This process has an important connection with the types of magma generated, which started high-Ti (Pitanga) in the northern part of the basin, under conditions of a thick crust and a low volume of magma extraction. The magmas extracted evolved to a highly fractionated and contaminated magmatic system on the southern part of the basin (Gramado magma type). Turner et al. (1994) and Stewart et al. (1996) suggest that different types of magma extruded in distinct places at the same time; e.g., the Paranapanema type in the north extruded at the same time as the Gramado and Urubici types in the south. Based on these characteristics we know that the different types of magma were not generated by fractionation from a single mantle source, but was generated from different sources of magma. Moreover, it is suggested that the magma was generated in a wider area than the distribution of the lavas. The different types of magma were formed by reaction between the plume and lithosphere. The various chemical types are thought to have erupted in different places dischronously, Figure 2. Taking in consideration the Gondwana continent tectonic movement over the Tristan da Cunha hot spot, particularly regarding volcanic activities, we know that: a- the magmatic activities started in the northern part of the basin, with a high-Ti dominant magma type; b- fractionation and contamination became more effective with the continent movement and the plume ascension under the lithosphere; c- the melt and mixing with the crust started the acid

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 193 volcanic activity, that is present from São Paulo state (Ourinhos magma type) and increased to the south in Rio Grande do Sul state (Caxias magma type); d- the magma modification was not sharp from north to south, because interfingering of distinct magma types is identified in the geological sequence; e- the same magma type can occur in distinct stratigraphic levels, so care has to be taken in the stratigraphic designation of the lavas; f- the chemical composition is a major step in the understanding of magmatic processes, but the positioning of each flow in the stratigraphic unit requires additional characterization; g- geochronological techniques are very useful in the studies, but the present precision of geochronological techniques, particularly the Ar-Ar, are not sufficient to solve stratigraphyc problems in detail.

Figure 2 - Sketch map showing the transition of the Paraná flood basalts magmatism. The center of the magmatism shifted from northwest to southeast, according to the relative movement of the plume.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 194 The Serra Geral Group and its constituent Formations Geological investigations have identified a sequence of 1700 m thick of flood basalt, that outcrops from Uruguay to central Brazil in a 2500 km-long and 700 km-wide exposure. Geological sections, field survey and geochemical investigations show that this volcanic pile can be grouped into 16 units. These 16 units follow the definition of Formation by any code (e.g., National Commission on Stratigraphic Nomenclature) and are included within Gondwana III Supersequence, and modify the status of the Serra Geral Formation into Serra Geral Group. This proposal is open to improvement by further investigations in the geographic area of occurrence of the Serra Geral Group, because stratigraphic knowledge is strongly dependent on the continued dissemination of data. We selected a stratigraphic section that runs across the center of the basin (Figure 3) to understand the internal stratigraphy of the Serra Geral Group, which enabled to make a consistent proposal for the vertical organization of the distinct lava flow units, distinguished as Formations. This cross-section is based on the integration of a large volume of data, which includes data from the literature and new data generated along the research projects that were carried out by CPRM and UFRGS groups. From the literature, we obtained particularly data on borehole core geochemistry. We undertook extensive field work for stratigraphic units identifications and rock sample collection, in addition to new chemical analyses on cores, and Ar-Ar and U-Pb geochronological data. Gamma ray geophysical information was available from Petrobras investigations and from our new work. For a large part of Rio Grande do Sul state, we compiled and interpreted the gravimetric information available. When chemical compositions are combined with gamma ray, lithology, and stratigraphic position, we are able to subdivide the Serra Geral Magmatic Sequence, which can be mapped and recognizable across the Paraná Basin. The sedimentary layers present in the volcanic pile do not mark the contact between different formations, as here defined. One exception is at the base of the Caxias Formation, which has a layer of <10 m of coarse-grained sandstone below the base. The sedimentary layers are included in the volcanic formations, intensifying to the top of the volcanic sequence, near the depocenter of the basin. In the lower part of the Serra Geral Group and restricted to the Gramado Formation, the desert-generated sandstone intertrapps are included in the Botucatu Formation. Higher in the volcanic stratigraphy, the water-related sedimentary rocks have no stratigraphic name proposed. In the following listing of the stratigraphic units, each formation has only one magma- type. As mentioned above, a magma-type may be present in different formations. The vertical sequence of stratigraphic formations can vary considerably in different sections of the volcanic pile. The following stratigraphic units are present in the Paraná large igneous province (Table 1); the formations are not listed in vertical sequence of occurrence.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 195 References Bellieni, G., Comin-Chiaramonti, P.; Marques, L.S.; Melfi, A.J.; Piccirillo, E.M.; Nardy, A.J.R.; Roisenberg, A. 1984. High- and low-Ti flood basalts from the Paraná plateau (Brazil): petrology and geochemical aspects bearing on their mantle origin, Neues Jahrbuch für Mineralogie - Abhandlung 150, 272-306.

Cordani, U.G. and Valdoros, P., 1967. Basaltic rocks of the Paraná Basin. In: Problems in Brazilian Gondwana Geology, Eds. J.J. Bigarella, R.D. Becker and I.D. Pinto, Curitiba, 207-231.

Comin-Chiaramonti, P., Ernesto, M., Velázquez, V.F. and Gomes, C.B. (2005) Plumes beneath the Paraná Basin, eastern Paraguayfact or fiction. www.mantleplumes.org Discussing the Origin of “Hotspot” volcanism.

Ernest, R. and Buchan, K.L., 2001, Large mafic magmatic events through time and links to mantle-plume heads, in Ernest, R. and Buchan, K.L., eds., Mantle Plumes: their identification through time. Geological Soviety of America, special paper 352,

Gladzenko, T.P., Hinz,K., Eldhoim, O., Meyer, H., Neben, S., Skogseid, J. 1997, South Atlantic volcanic margins. Journal of the Geological Society of London, v. 154, p.465-470, 1997.

Marques, L.S., Dupré, B. And Piccirillo, E.M. 1999. Mantle source compositions of the Paraná Magmatic Province (southern Brazil): evidence from trace element and Sr-Nd-Pb isotope geochemistry. Journal of Geodynamics, 28, p.439-458.

Morgan, W.J. 1971. Convection plumes in the lower mantle. Nature, v.230, p.41-43.

Peate, D.W.; Hawkesworth, C.J.; Mantovani, M.S.M. 1992. Chemical stratigraphy of the Paraná lavas (South America):classification of magma types and their spatial distribution. Bulletin of Volcanology, v. 55, p. 119- 139.

Peate, D.W. 1997. The Paraná-Etendeka province, in Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism edited by J.J. Mahoney, and M.F. Coffin, pp. 217-245, American Geophysical Union Monograph, Washington DC.

Piccirillo, E.M. & Melfi, A.J. 1988. The Mesozoic flood volcanism of the Paraná basin: petrogenetic and geophysical aspects IAG-USP press, São Paulo, 1988.

Richards, M.A., Duncan, R.A., and Courtillot, V.E. 1989.Flood basalts and hot spot tracks: Plume heads and tails: Science, v.246, p. 103-107.

Scherer, C.M. 2002. Preservation of Aeolian genetic units by lava flows in the Lower Cretaceous of the Paraná Basin, southern Brazil. Sedimentology 49:97-116

Stewart, k., Turner, S., Kelley, S., Hawkesworth, C., Kirstein, L. and Mantovani, M., 1996. 3-d 40Ar/39Ar Geochronology in the Paraná continental flood basalt province: Earth and Planetary Science Letters, v. 143, p.95-109.

Turner, S., Regelous, M., Kelley, S., Hawkesworth, K. and Mantovani, M.S.M.. (1994). Magmatism and continental break-up in the South Atlantic: high precision 40Ar-39Ar geochronology. Earth and Planetary Science Letters 121, 333-348.

White, R.S. and McKenzie, D., 1989, Magmatism at rift zones: the generation of volcanic continental margins and flood basalts: Journal of Geophysical Research, v.94, p. 7685-7729.

White, I.C. 1908. Relatório Final da Comissão de Estudos das Minas de Carvão de Pedra do Brasil. Rio de Janeiro, DNPM, 1988. Parte I; Parte II, p. 301-617.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 196

Figure 3 - Schematic N-S cross section through the center of the volcanic pile, the geochemical-geophysical correlations and the formation names proposed to the Sera Geral Group.

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts 197 Table 1 - Stratigraphic nomenclature of the Serra Geral Group, and chemical connection with the eight magma type classification from Peate, (1997).

Formation Magma type Characteristics Esmeralda Esmeralda Basalt - low Ti/Y Várzea do Cedro Palmas Rhyolite – low Ti Palmas Palmas Rhyolite – low Ti Ourinhos Chapecó Rhyolite – high Ti-Zr Chapecó Chapecó Rhyolite – high Ti-Zr Paranapanema Paranapanema Basalt - intermediate Ti/Y Campo Erê Paranapanema Basalt - intermediate Ti/Y Cordilheira Alta Paranapanema Basalt - intermediate Ti/Y Capanema Paranapanema Basalt - intermediate Ti/Y Campos Novos Esmeralda Basalt - low Ti/Y Alegrete Gramado Basalt - low Ti/Y Gramado Gramado Basalt - low Ti/Y Nova Laranjeiras Pitanga Basalt - high Ti/Y Pitanga Pitanga Basalt - high Ti/Y Urubici Urubici Basalt - high Ti/Y Ribeira Ribeira Basalt - high Ti/Y Morungava Complex Morungava Picritic mafic-ultramafic sills

Problems in Western Gondwana Geology I, Gramado 2007 Extended Abstracts