Direct dating of the Sete Lagoas cap carbonate (Bambuí Group,Brazil) and implications for the Neoproterozoic glacial events

The end of the Neoproterozoic is punctuated by glacial deposition, but the chronology of these deposits is hindered presently by the paucity of geochronological data. Here, we present new radiometric dating for the basal Sete Lagoas cap carbonate deposits that overlie glacial units in the São Francisco craton. Six samples from aragonite‐pseudomorph crystal‐rich facies, showing pristine textures and constant ⁸⁷Sr/⁸⁶Sr ratios around 0.7075, yielded a Pb–Pb isochron age of 740 ± 22 Ma, which is interpreted as the depositional age for these remarkably preserved rocks. This age can be used to infer a low‐to‐moderate palaeolatitude of 20–30° for carbonate (and glacial) deposition. In addition, as it overlaps the ages obtained for the oldest Neoproterozoic glacial successions, our result reinforces the idea of a long‐standing ‘Sturtian’ interval, suggesting that this event represents either different discrete glaciations or a protracted event encompassing almost 80 Ma.     

Sedimentology,stable‐isotope geochemistry and palaeoenvironment of dolostones capping late Precambrian glacial sequences in Australia

Thin (<10 m), extensive dolostones conformably capping late Precambrian glaciogenic rocks in the Kimberley region and the Adelaide Geosyncline commonly comprise micritic dolomite apparently of ‘primary’ or early diagenetic origin. Their aphanitic texture, the presence of algal and cryptalgal lamination, tepee structures and intraclastic rocks, and the lack of saline evaporites or their pseudomorphs, suggest deposition mainly in supratidal to lagoonal environments under a seasonally evaporitic/humid climate. Less common oolitic or pelletal rocks and a finely laminated shaly dolostone facies also suggest peritidal to marine influences.

The ‘cap’ dolostones (omitting possible large erratics) have a δ¹⁸O mean of —6.4 ± 1.9% vs. PDB and a δ¹³C mean of —1.9 ± 2.1% vs. PDB. The oxygen mean is comparable to that determined for other late Precambrian ‘primary’ or early diagenetic dolostones not associated with glaciogenic rocks and presumably deposited under warm conditions. The carbon mean, in contrast, is lighter than that determined for other late Precambrian dolostones.

The sedimentological and oxygen‐isotope data are consistent with relatively high formation‐temperatures for the cap dolostones. Abrupt climatic warming at the close of late Precambrian glacial epochs is implied.     

Understanding carbonate factories through palaeoecological and sedimentological signals – Tribute to Luis Pomar

The carbonate factories, their controlling factors and their palaeoecological and sedimentological signals recorded in sedimentary successions are key elements for understanding the evolution of carbonate platform systems. Luis Pomar has dedicated most of his academic life to the study of carbonate rocks and carbonate factories. The idea for this special issue to celebrate Pomar’s career arose during a session at the 34^th International Association of Sedimentologists meeting held in Rome, entitled ‘Understanding carbonate factories through palaeoecological and geochemical signals’. The proposal encountered great response among participants, and additional contributions followed an email invitation to other specialists. This issue contains a variety of papers on carbonate sedimentology and carbonate factories. Here, an introduction that contextualizes the papers and key concepts discussed in this thematic issue is provided. It reviews Luis Pomar’s major achievements in carbonate sedimentology and discusses the evolution of the concept of the carbonate factory and the series of palaeoecological and sedimentological signals used to characterize the wide spectrum of carbonate depositional systems found in the natural world.     

Palaeolatitude of glacial deposits and palaeogeography of Neoproterozoic ice ages

At the end of the Neoproterozoic, the Earth may have experienced important environmental changes, with a transition between two supercontinents (from Rodinia to Gondwana), extensive glaciations with ice caps reaching the Equator and the beginning of metazoan diversification. In such a context, the palaeomagnetic record can be used to constrain both the palaeogeography and the palaeoclimate (palaeolatitudinal distribution of glacial deposits). Here we present an up-to-date geochronological and palaeomagnetic database for the Neoproterozoic glacial deposits, including poles recently obtained on ‘cap carbonates’ from China, Oman, and Amazonia. The database comprises ten poles (from eight different cratons), obtained directly on the glacial deposits or on the overlying ‘cap carbonate’, and two other palaeolatitudes derived from reference poles coeval to well-dated glacial units in the same craton. The occurrence of glacial deposition at low latitudes (<30 °) is attested by some good-quality poles, two of them well dated at ∼740 and ∼635 Ma. Based on these poles and on reference poles obtained on igneous rocks, tentative palaeogeographic reconstructions for ∼750, ∼620, and ∼580 Ma (ages for which the database has limited but still sufficient entries) were performed in order to investigate the tectonic context within which the glacial events were produced.     

Depositional systems,composition and geochemistry of Triassic rifted‐continental margin redbeds of the Internal Rif Chain,Morocco

The Middle to Upper Triassic redbeds at the base of the Ghomaride and Internal ‘Dorsale Calcaire’ Nappes in the Rifian sector of the Maghrebian Chain have been studied for their sedimentological, petrographic, mineralogical and chemical features. Redbeds lie unconformably on a Variscan low‐grade metamorphic basement in a 300 m thick, upward fining and thinning megasequence. Successions are composed of predominantly fluvial red sandstones, with many intercalations of quartzose conglomerates in the lower part that pass upwards into fine‐grained micaceous siltstones and massive mudstones, with some carbonate and evaporite beds. This suite of sediments suggests that palaeoenvironments evolved from mostly arenaceous alluvial systems (Middle Triassic) to muddy flood and coastal plain deposits. The successions are characterized by local carbonate and evaporite episodes in the Late Triassic. The growth of carbonate platforms is related to the increasing subsidence (Norian‐Rhaetian) during the break‐up of Pangea and the earliest stages of the Western Tethys opening. Carbonate platforms became widespread in the Sinemurian. Sandstones are quartzose to quartzolithic in composition, testifying a recycled orogenic provenance from low‐grade Palaeozoic metasedimentary rocks. Palaeoweathering indices (Chemical Index of Alteration, Chemical Index of Weathering and Plagioclase Index of Alteration) suggest both a K‐enrichment during the burial history and a source area that experienced intense weathering and recycling processes. These processes were favoured by seasonal climatic alternations, characterized by hot, episodically humid conditions with a prolonged dry season. These climatic alternations produced illitization of silicate minerals, iron oxidation and quartz‐rich red sediments in alluvial systems. The estimated burial temperature for the continental redbeds is in the range of 100 to 160 °C with lithostatic/tectonic loading of ca 4 to 6 km. These redbeds can be considered as regional petrofacies that mark the onset of the continental rift valley stage in the Western Pangea (Middle Triassic) before the opening of the western part of Tethys in the Middle Jurassic. The studied redbeds and the coeval redbeds of many Alpine successions (Betic, Tellian and Apenninic orogens) show a quite similar history; they identify a Mesomediterranean continental block originating from the break‐up of Pangea, which then played an important role in the post‐Triassic evolution of the Western Mediterranean region.     

Neoproterozoic cap carbonates: a critical appraisal of existing models and the plumeworld hypothesis

Evidence for glaciation during the mid-late Neoproterozoic is widespread on Earth, reflecting three or more ice ages between 730 Ma and 580 Ma. Of these, the late Neoproterozoic Marinoan glaciation of approximately 635 Ma stands out because of its ubiquitous association with a characteristic, microcrystalline cap dolostone that drapes glacially influenced rock units worldwide. The Marinoan glaciation is also peculiar in that evidence for low altitude glaciation at equatorial latitudes is compelling. Three models have been proposed linking abrupt deglaciation with this global carbonate precipitation event: (i) overturn of an anoxic deep ocean; (ii) catastrophically accelerated rates of chemical weathering because of supergreenhouse conditions following global glaciation (Snowball Earth Hypothesis); and (iii) massive release of carbonate alkalinity from destabilized methane clathrates. All three models invoke extreme alkalinity fluxes into seawater during deglaciation but none explains how such alkalinity excess from point sources could be distributed homogeneously around the globe. In addition, none explains the consistent sequence of precipitation events observed within cap carbonate successions, specifically: (i) the global blanketing of carbonate powder in shallow marine environments during deglaciation; (ii) widespread and disruptive precipitation of dolomite cement; followed by (iii) localized barite precipitation and seafloor cementation by aragonite. The conceptual model presented here proposes that low latitude deglaciation was so massive and abrupt that the resultant meltwater plume could extend worldwide, physically separating the surface and deep ocean reservoirs for ≥10³ years. It is proposed that cap dolostones formed primarily by microbially mediated precipitation of carbonate whitings during algal blooms within this low salinity plumeworld rather than by abiotic precipitation from normal salinity seawater. Many of the disruption features that are characteristic of cap dolostones can be explained by microbially mediated, early diagenetic dolomitization and cementation. The re-initiation of whole ocean circulation degassed CO₂ into the atmosphere in areas of upwelling, triggering localized, abiotic CaCO₃ precipitation in the form of aragonite fans that overlie cap dolostones in NW Canada and Namibia. The highly oxygenated shallow marine environments of the glacial and post-glacial Neoproterozoic world provided consistently favourable conditions for the evolutionary development of animals and other oxygenophiles.     

Stratigraphie isotopique du strontium et datation des formations carbonatées et glaciogéniques néoprotérozoiques du Nord et de l’Ouest du craton du Congo

Numerous occurrences of Neoproterozoic carbonate platforms and glacigenic lithofacies are present around the Congo craton. They are especially well developed on its western and northern borders, i.e. in the forelands of the West Congo and Oubanguides belts. Sr isotopic stratigraphy enables us to characterize the deposition age of some carbonate units from these two domains. The ⁸⁷Sr/⁸⁶Sr isotopic ratios of limestones from the late ‘Haut Shiloango’ (0.7068) and the ‘Schisto-calcaire’ (0.7075) of the West-Congo domain are of post-Sturtian and post-Marinoan ages, respectively. The Lenda carbonates (0.7066) from the Northeast of the Democratic Republic of Congo and the limestones (0.7076) from the Bangui Basin, both in the Oubanguides foreland, are of pre-Sturtian and post-Marinoan ages, respectively. These data associated with lithostratigraphic correlations allow us to ascribe the ‘Bas Congo’ lower mixtite (tillite) and the Akwokwo tillite (Lindian) to the Sturtian ice age. In the same way, the ‘Bas Congo’ upper mixtite (tillite) and the Bondo tillite (Bakouma Basin) are likely Marinoan in age. A new synthetic stratigraphy for these Neoproterozoic domains is developed.     

Petrological,geochemical and geochronological evidence for a Neoproterozoic ocean basin recorded in the Marlborough terrane of the northern New England Fold Belt

Petrological, geochemical and radiogenic isotopic data on ophiolitic‐type rocks from the Marlborough terrane, the largest (～700 km²) ultramafic‐mafic rock association in eastern Australia, argue strongly for a sea‐floor spreading centre origin. Chromium spinel from partially serpentinised mantle harzburgite record average Cr/(Cr + Al) = 0.4 with associated mafic rocks displaying depleted MORB‐like trace‐element characteristics. A Sm/Nd isochron defined by whole‐rock mafic samples yields a crystallisation age of 562 ± 22 Ma (2σ). These rocks are thus amongst the oldest rocks so far identified in the New England Fold Belt and suggest the presence of a late Neoproterozoic ocean basin to the east of the Tasman Line. The next oldest ultramafic rock association dated from the New England Fold Belt is ca530 Ma and is interpreted as backarc in origin. These data suggest that the New England Fold Belt may have developed on oceanic crust, following an oceanward migration of the subduction zone at ca540 Ma as recorded by deformation and metamorphism in the Anakie Inlier. Fragments of late Neoproterozoic oceanic lithosphere were accreted during progressive cratonisation of the east Australian margin.     

On the reliability of stable carbon isotopes for Neoproterozoic chemostratigraphic correlation

The reliability of δ¹³C trends in Neoproterozoic carbonate-dominated successions for regional and global chemostratigraphic correlation is discussed. In the light of recent findings of a predominantly non-marine rare earth element and yttrium signature in most Neoproterozoic carbonates and a comparatively short oceanic residence time of carbon, trends towards enrichment in ¹³C seen in many of these carbonates are considered to reflect facies variations rather than temporal signals of ocean chemistry. Positive δ¹³C_Carb excursions are explained by elevated bioproductivity and/or increased evaporation in shallow marine, near-coastal, temporarily restricted depositional environments. Examples are provided that illustrate that C isotope trends can be highly ambiguous temporal markers and are in the absence of other chemostratigraphic data, such as Sr isotope ratios, and radiometric age control of only limited use for stratigraphic correlation. The overall enrichment in ¹³C recorded by most Neoproterozoic carbonates, except for those in close stratigraphic proximity to glacial deposits, is suggested to reflect a dominance of microbially mediated carbonate formation in the Neoproterozoic. This might explain why C isotope chemostratigraphy in Neoproterozoic successions is less reliable than in Phanerozoic successions in which carbonates are, with only few exceptions, biogenic products of shelly fossils.     

The Witputs diamictite in southern Namibia and associated rocks: constraints for a global glaciation?

The Witputs section of the Gariep Belt (S Namibia) comprises a sequence of clastic and chemical sediments, which have been interpreted as representative of a Late Neoproterozoic global or near-global ice age event, and recent biostratigraphic work in the upper rocks of the Witputs suggest a Late Ediacaran age. To further characterise this sequence and provide additional age constraints, a detailed sedimentological and detrital zircon study has been carried out. The petrographic, sedimentological and geochemical characteristics of the Witputs diamictite determined in this study are homogenous and indicative of debris flow or palaeo-valley infill sediments, deposited in an oxic environment with no glaciogenic evidence. This homogeneity is also reflected in the detrital zircon age spectra with most ages falling between 1.0 and 1.3?Ga, representing the local geology, with the youngest grain at 1030.2?±?10.9 (2??) (n?=?92?<10% discordance), despite the fact that mid and Late Neoproterozoic volcanic activity is known in the local region. The overlying carbonate rocks, often considered to be ??cap carbonates??, show high Mn (up to 60% MnO), with base metal precipitation (Zn, V, Co), and are recrystallised. Their ??¹³C_VPDB isotope ratios are homogeneous at around-3. Major and trace element ratios reach values which indicate that C?CO isotopes may be disturbed and might not reflect primary global seawater composition, thus questioning their use for global correlation and comparison with composite chemostratigraphic curves. The contact to the overlying Late Ediacaran Sanddrif Member is not exposed, and the rocks dip in a different direction than the underlying carbonate rocks. The c. 40-m-thick section is characterised by rapid lithology changes including shales, calcareous sandstones and wackes, fine-grained conglomerates and rare clean quartz-rich sandstones, all of which have strikingly similar detrital zircon populations, and the youngest zircon is dated at 1082.8?±?10?Ma (2?? errors, from 72 grains with <10% discordance). Acritarchs earlier found in the Sanddrif Member, however, indicate a post-570?Ma depositional age. If the diamictites are glacio-marine deposits, then an interesting conclusion is that the clastic sediments can display a very immature geochemical signature, indicating a localised provenance, with derivation purely from the local basement rocks, which is also reflected in the detrital zircon populations. However, we would hesitate to assign a glacial origin to the deposits as no glacial indicators, other than a diamictitic texture, were observed. Clearly, far more work on the detailed mapping and sedimentology of the Neoproterozoic Gariep Belt deposits is required, particularly as many are currently used for global correlation. Age constraints derived from extensive detrital zircon work can only constrain the deposits as being post 1.03?Ma with the detritus being purely locally derived.     

Ediacaran in Uruguay: palaeoclimatic and palaeobiological implications

The Ediacaran to lowermost Cambrian successions of south‐eastern Uruguay preserve an unusual and significant record of deposits generated during the Gondwana assembly (ca 590 to 535 Ma). This study presents a review of data obtained through extensive field‐based mapping coupled with detailed sedimentology and stratigraphy of key formations. The geological units within the study area consist of the Maldonado Group (Playa Hermosa, Las Ventanas and San Carlos formations), the Arroyo del Soldado Group (Yerbal, Polanco Limestones, Barriga Negra and Cerro Espuelitas formations) and the Arroyo de la Pedrera Group (Piedras de Afilar and Cerro Victoria formations). The Maldonado Group is characterized by a glacially influenced volcanogenic‐sedimentary sequence with ice‐rafted debris and dropstones in the Playa Hermosa and Las Ventanas formations. The Arroyo del Soldado Group is a mixed siliciclastic‐carbonate succession, mainly represented by an intercalation of basal pink dolostones, banded siltstones, rhythmites of dolostone‐limestone, iron formations, cherts and conglomerates. Carbonates in the Polanco Limestones Formation are characterized by a negative δ¹³C excursion up to ?3·26‰ PeeDeeBelemnite. The Arroyo de la Pedrera Group consists of quartz arenites and stromatolitic/oolitic dolostones. Preliminary data indicate that the Precambrian–Cambrian could be contained within or at the base of this group. The entire succession is almost 6000 m thick, and contains a rich fossil assemblage composed of organic‐walled microfossils and small shelly fauna, including the index fossil Cloudina riemkeae. The stratigraphic and chemostratigraphic features are suggestive of a Gaskier age (ca 580 Ma) for the basal glacial‐related units. In this scenario, the results show the importance of lithostratigraphic, biostratigraphic and chemostratigraphic data of these Ediacaran units in the global correlation of terminal Proterozoic sedimentary rocks.     

Shallow-burial dolomite cement: a major component of many ancient sucrosic dolomites

Dolomite cement is a significant and widespread component of Phanerozoic sucrosic dolomites. Cements in dolomites that were never deeply buried are limpid, have planar faces (non‐saddle forms), often distinct zonation in cathodoluminescence and form syntaxial overgrowths on crystals facing pores. Five samples of sucrosic dolomites, interpreted as having had mostly lime‐mudstone or wackestone precursors in four carbonate aquifers, provide insights into the abundance of planar cements in sucrosic dolomites. Such cement comprises 11% to 45% (32% mean) of peritidal to sub‐tidal dolomites on an outcrop in the Edwards aquifer (Early Cretaceous) of central Texas; 19% to 33% (25% mean) of ramp dolomites in the Hawthorn Group (Oligo‐Miocene) and 50% to 70% in shelf dolomites of the Avon Park Formation (Eocene) in the Upper Floridan aquifer of sub‐surface peninsular Florida; 18% to 45% (32+% mean) of sub‐tidal shelf dolomites in quarry sections of the Burlington‐Keokuk Formation (Early Mississippian) in south‐eastern Iowa; and 18% to 76% (50% mean) in shallow cores and outcrops of outer‐shelf dolomites from the Gambier Limestone (Oligo‐Miocene) of South Australia. Backstripping the cement phases revealed by cathodoluminescence colour photomicrographs documents the effects of cements on textural coarsening, pore‐space reduction, induration and general ‘maturation’ of these dolomites. Most pre‐Holocene dolomites are multiphase crystalline rocks composed of: (i) seed crystals or ‘cores’; (ii) crystal cortices that concentrically enlarged the cores; and (iii) free‐space, syntaxial precipitates of limpid cement around the crystals. Remaining CaCO₃ grains and micrite can be replaced by dolomite, but typically they are dissolved between stages (ii) and (iii), creating systems of intercrystal and mouldic pores typical of sucrosic dolomites. Networks of cement overgrowths, aided by water‐filled pore systems under hydrostatic to lithostatic pressure, are judged to slow or prevent compaction in sucrosic dolomites. It can be argued that cortex growth involves both replacement of CaCO₃ particles and microcementation of their interparticle pores. This interpretation, and the abundance of cements in so many dolomites, would obviate the controversy over the volumetrics of ‘replacement dolomitization’. Limpid, planar and syntaxial dolomite cements of early diagenetic origin are interpreted to have precipitated from clear pore waters, at low temperatures (<30 to 35 °C) and shallow burial depths (<100 m), in water‐saturated networks of dolomite ‘silt’ and ‘sand’. Cements in many dolomites in island and continental–aquifer systems appear to result from event‐driven processes related to sea‐level highstands. Cementation events can follow ‘replacement dolomitization’ events by time intervals ranging from geologically ‘instantaneous’ to tens of million years.     

Corrigendum: Strange bedfellows: glacial diamictite and cap carbonate from the Marinoan (635 Ma) glaciation in Namibia

Tectonic fold tests conducted in Namibia demonstrate that the inclination with respect to bedding of geoplumb (palaeovertical) tubular structures in the Marinoan (635 Ma) syndeglacial cap dolostone is mainly the result of tectonic strain. Therefore, tubestone inclination data cannot be used to estimate the gradient of the sea floor on the foreslope of the Otavi carbonate platform during the Marinoan glaciation. A gradient steeper than 0·1 (slope angle ca 5·7°), implying a glacial base‐level fall ≥0·5 km, is nevertheless supported by boulder‐size intraclast debrite in the falling‐stand wedge directly beneath the glacigenic sequence. Cryogenian oceans lacked skeletal carbonate production, raising the carbonate saturation state and persistent deep water anoxia excluded acid‐producing aerobic respiration, facilitating early diagenetic carbonate precipitation, lithification and steep submarine slopes.     

Chert spheroids of the Monterey Formation,California (USA): early‐diagenetic structures of bedded siliceous deposits

Chert spheroids are distinctive, early‐diagenetic features that occur in bedded siliceous deposits spanning the Phanerozoic. These features are distinct in structure and genesis from similar, concentrically banded ‘wood‐grain’ or ‘onion‐skin’ chert nodules from carbonate successions. In the Miocene Monterey Formation of California (USA), chert spheroids are irregular, concentrically banded nodules, which formed by a unique version of brittle differential compaction that results from the contrasting physical properties of chert and diatomite. During shortening, there is brittle fracture of diatomite around, and horizontally away from, the convex surface of strain‐resistant chert nodules. Unlike most older siliceous deposits, the Monterey Formation still preserves all stages of silica diagenesis, thus retaining textural, mineralogical and geochemical features key to unravelling the origin of chert spheroids and other enigmatic chert structures. Chert spheroids found in opal‐A diatomite form individual nodules composed of alternating bands of impure opal‐CT chert and pure opal‐CT or chalcedony. With increased burial diagenesis, surrounding diatomite transforms to bedded porcelanite or chert, and spheroids no longer form discrete nodules, yet still display characteristic concentric bands of pure and impure microcrystalline quartz and chalcedony. Petrographic observations show that the purer silica bands are composed of void‐filling cement that precipitated in curved dilational fractures, and do not reflect geochemical growth‐banding in the manner of Liesegang phenomena invoked to explain concentrically banded chert nodules in limestone. Chertification of bedded siliceous sediment can occur more shallowly (< 100 m) and rapidly (< 1 Myr) than the bulk silica phase transitions forming porcelanite or siliceous shale in the Monterey Formation and other hemipelagic/pelagic siliceous deposits. Early diagenesis is indicated by physical properties, deformational style and oxygen‐isotopic composition of chert spheroids. Early‐formed cherts formed by pore‐filling impregnation of the purest primary diatomaceous beds, along permeable fractures and in calcareous–siliceous strata.     

Sedimentology and chemostratigraphy of the Bwipe Neoproterozoic cap dolostones (Ghana,Volta Basin): A record of microbial activity in a peritidal environment

Neoproterozoic glacial and post-glacial sediments from the Volta Basin (West Africa) form a stratigraphic ‘Triad’ of tillites, carbonates and cherts. The carbonates that cap the tillites were studied in detail at Bwipe (Ghana), in the western part of the basin. They are made of finely-laminated dolostones with well-preserved sedimentary features, suggesting that dolomite formation was penecontemporaneous of deposition in a warm, arid peritidal environment, with microbial activity. Rare-earth element distribution display seawater-like patterns. Redox-sensitive trace elements indicate suboxic conditions. High Ba/Al ratios can be related to high organic productivity. δ¹³C values are nearly constant at −3.7‰ and δ¹⁸O_PDB are about −6‰. ⁸⁷Sr/⁸⁶Sr ratios range between 0.7061 and 0.7073. The δ¹³C signatures are nearly similar to those of coeval cap dolostones from the northern part of the Basin. The dolostones change upwards to limestones with secondary textures, as well as more negative δ¹³C and higher ⁸⁷Sr/⁸⁶Sr ratio. Therefore, only the dolostones witness the post-glacial conditions in seawater. It is proposed that, due to a bloom of microbial productivity following ice thaw, organic matter likely accumulated at the water–sediment interface and was consumed by sulphate-reducing bacteria, leading to the precipitation of δ¹³C-negative dolomite. This microbially mediated model is supported by present-day field evidence from hypersaline lagoons in Brazil and by previously published culture experiments. It is consistent with the sedimentological and geochemical data from the Ghana cap dolostones and can be applied to other Neoproterozoic cap dolostones worldwide.     

Major shoreline retreat and sediment starvation following Snowball Earth

While cap dolostones are integral to the Snowball Earth hypothesis, the current depositional model does not account for multiple geological observations. Here we propose a model that rationalises palaeomagnetic, sequence‐stratigraphic and sedimentological data and supports rapid deglaciation with protracted cap dolostone precipitation. The Snowball Earth hypothesis posits that a runaway ice‐albedo can explain the climate paradox of Neoproterozoic glacial deposits occurring at low palaeolatitudes. This scenario invokes volcanic degassing to increase atmospheric greenhouse gases to a critical threshold that overcomes the albedo effect and brings the planet back from the ice‐covered state. Once this occurs, Earth should shift rapidly from a snowball to an extreme greenhouse. However, cap dolostone units overlying glacial sediments, typically interpreted as transgressive deposits, exhibit multiple magnetic reversals indicating they accumulated in >10⁵ years. By reviewing modern post‐glacial systems, sequence stratigraphic concepts and principles of sedimentology, we suggest that cap dolostones are not restricted to the transgression but rather represent sediment starvation following a major landward shoreline migration associated with the demise of Snowball Earth. Thus, the duration in which cap dolostone accumulated is not directly coupled to the timescale of the Snowball Earth deglaciation.     

Sediment dispersal and quantitative stratigraphic architecture across an ancient shelf

Two large (200 to 300 km), near‐continuous outcrop transects and extensive well‐log data (ca 2800 wells) allow analysis of sedimentological characteristics and stratigraphic architecture across a large area (ca 60 000 km²) of the latest Santonian to middle Campanian shelf along the western margin of the Western Interior Seaway in eastern Utah and western Colorado, USA. Genetically linked depositional systems are mapped at high chronostratigraphic resolution (ca 0·1 to 0·5 Ma) within their sequence stratigraphic context. In the lower part of the studied interval, sediment was dispersed via wave‐dominated deltaic systems with a ‘compound clinoform’ geomorphology in which an inner, wave‐dominated shoreface clinoform was separated by a muddy subaqueous topset from an outer clinoform containing sand‐poor, gravity‐flow deposits. These strata are characterized by relatively steep, net‐regressive shoreline trajectories (>0·1°) with concave‐landward geometries, narrow nearshore belts of storm‐reworked sandstones (2 to 22 km), wide offshore mudstone belts (>250 km) and relatively high sediment accumulation rates (ca 0·27 mm year^?1). The middle and upper parts of the studied interval also contain wave‐dominated shorefaces, but coeval offshore mudstones enclose abundant ‘isolated’ tide‐influenced sandstones that were transported sub‐parallel to the regional palaeoshoreline by basinal hydrodynamic (tidal?) circulation. These strata are characterized by relatively shallow, net‐regressive shoreline trajectories (<0·1°) with straight to concave‐seaward geometries, wide nearshore belts of storm‐reworked sandstones (19 to 70 km), offshore mudstone belts of variable width (130 to >190 km) and relatively low sediment accumulation rates (ca ≤0·11 mm year^?1). The change in shelfal sediment dispersal and stratigraphic architecture, from: (i) ‘compound clinoform’ deltas characterized by across‐shelf sediment transport; to (ii) wave‐dominated shorelines with ‘isolated’ tide‐influenced sandbodies characterized by along‐shelf sediment transport, is interpreted as reflecting increased interaction with the hydrodynamic regime in the seaway as successive shelfal depositional systems advanced out of a sheltered embayment (‘Utah Bight’). This advance was driven by a decreasing tectonic subsidence rate, which also suppressed autogenic controls on stratigraphic architecture.     

Microbial mats implicated in the generation of intrastratal shrinkage (‘synaeresis’) cracks

Intrastratal shrinkage (often termed ‘synaeresis’) cracks are commonly employed as diagnostic environmental indicators for ancient salinity‐stressed, transitional fluvial‐marine or marginal‐marine depositional environments. Despite their abundance and use in facies interpretations, the mechanism of synaeresis crack formation remains controversial, and widely accepted explanations for their formation have hitherto been lacking. Sedimentological, ichnological, petrographic and geochemical study of shallow marine mudstone beds from the Ordovician Beach Formation of Bell Island, Newfoundland, has revealed that crack development (cf. synaeresis cracks) on the upper surface of mudstone beds is correlated with specific organic, geochemical and sedimentological parameters. Contorted, sinuous, sand‐filled cracks are common at contacts between unbioturbated mudstone and overlying sandstone beds. Cracks are absent in highly bioturbated mudstone, and are considered to pre‐date firmground assemblages of trace fossils that include Planolites and Trichophycus. The tops of cracked mudstone beds contain up to 2·1 wt% total organic carbon, relative to underlying mudstone beds that contain around 0·5 wt% total organic carbon. High‐resolution carbon isotope analyses reveal low δ¹³C_org values (?27·6‰) on bed tops compared with sandy intervals lacking cracks (?24·4 to ?24·9‰). Cracked mudstone facies show evidence for microbial matgrounds, including microbially induced sedimentary structures on bedding planes and carbonaceous laminae and tubular carbonaceous microfossils in thin section. Non‐cracked mudstone lacks evidence for development of microbial mats. Microbial mat development is proposed as an important prerequisite for intrastratal shrinkage crack formation. Both microbial mats and intrastratal shrinkage cracks have broad palaeoenvironmental distributions in the Precambrian and early Phanerozoic. In later Phanerozoic strata, matgrounds are restricted to depositional environments that are inhospitable to burrowing and surface‐grazing macrofauna. Unless evidence of synaeresis (i.e. contraction of clay mineral lattices in response to salinity change) can be independently demonstrated, the general term ‘intrastratal shrinkage crack’ is proposed to describe sinuous and tapering cracks in mudstone beds.     

Upper Cretaceous bottom current deposits,north-east Greenland

Reported ancient bottom current deposits in deep marine settings are scarce and most of them remain contentious. This study describes sedimentological, ichnological and petrographical characteristics of a drill core that covers ca 10 Myr of Upper Cretaceous stratigraphy at Hold with Hope, north-east Greenland. The core is divided into four facies associations, which are interpreted to reflect deposition from bottom currents, turbidity flows and hemipelagic settling in slope and/or near slope environments. The evidence for bottom current influence is three-fold. Firstly, pervasive indications of winnowing such as marine bioclast-rich lags and outsized clasts on ‘mud on mud’ contacts are suggestive of low-sediment concentration flows capable of transporting up to pebble-sized clasts. Common Mn–Fe–Mg rich carbonate matrix cements and various types of hiatal chemogenic lag deposits showing glauconite, apatite and carbonate clasts also point to condensation, prolonged exposure at the sediment–water interface and recurrent phases of sea-floor erosion. Secondly, such deposits can show indicators for tidal processes such as double mud-drapes, tangential bottom sets in dune-scale cross-bedding and cyclic rhythmites. Thirdly, inverse to normal grading at various scales is common in fully marine, commonly seafloor-derived sediments. Ichnological data indicate considerable taxonomic variability in the bottom current deposits, but recurrent fabrics are characteristically dominated by morphologically simple burrows such as Thalassinoides and Planolites, with secondary Phycosiphon, Nereites, Zoophycos and/or Chondrites. In general, opportunistic taxa are common whereas mature composite ichnofabrics are rare. The omission surfaces are locally burrowed with stiffground to firmground trace fossil suites. The results contribute to establishing sedimentological, ichnological and mineralogical criteria for recognition of bottom current deposits as well as to the understanding of the Late Cretaceous palaeoenvironmental evolution of the Arctic region.     

Dolomite Controls on Phanerozoic Seawater Chemistry

We investigate the potential role of dolomite as a long-term buffer on Phanerozoic seawater composition. Using a comprehensive model of Phanerozoic geochemical cycling, we show how variations in the formation rate of sedimentary marine dolomite have buffered seawater saturation state. The total inventory of inorganic carbon reflects the sum of fluxes derived from continental weathering, basalt-seawater exchange, alumino-silicate diagenesis (reverse weathering), and global deposition of calcium carbonate. Although these fluxes are approximately balanced, model results indicate that seawater saturation state is sensitive to the marine dolomite depositional flux. This conclusion is consistent with and constrained by independent proxy data for seawater ion ratios, paleo-atmospheric CO₂ concentrations, and paleo-pH data, and dolomite mass-age distribution through Phanerozoic time. Abundant research indicates that dolomite’s occurrence in marine sediments is sensitive to many factors: temperature, seawater composition, paleogeographic setting, continental organization, etc. Although the complexity of the process of dolomite formation prevents a complete understanding of the relative role of these factors, our model results clearly underscore the importance of this mineral in the chemical history of Phanerozoic seawater.