Basin Research - 1, 2014

Deep‐water margins have been the focus of considerable research during the past decade. They comprise vast, underexplored regions, in which only recently have improvements in seismic imaging and drilling technology allowed the discovery of significant hydrocarbon accumulations. This volume comprises of a series of manuscripts based on studies from continental margins bordering India, East Africa, Australia, China, Norway, the United Kingdom, Iberia, Newfoundland, the southern US, West Africa and Brazil, thus offering a global perspective on the evolution and economic significance of deep‐water margins. The articles in this volume examine: (i) the quantification of extension and hyperextension in distal parts of continental margins, and their relationship with regional subsidence, (ii) the importance of magmatism in the structural and thermal evolution of rifted continental margins, (iii) the processes driving and the significance of regional exhumation during and after syn‐rift stretching, (iv) the tectonic setting of salt basins and (v) depositional patterns along deep‐water margins. To complement this work, we present a personal view of some of the specific questions that need to be addressed in the next few years of deep‐water continental margin research.

There is growing recognition that pulses of compressive tectonic structuring punctuate the post‐breakup subsidence histories of many ‘passive’ rifted continental margins. To obtain new insights into the nature and origin of compression at passive margins, we have conducted a comprehensive analysis of the post‐breakup (<43 Ma) deformation history of the offshore Otway Basin, southern Australian margin, using a regional seismic database tied to multiple wells. Through mapping of a number of regional intra‐Cenozoic unconformities we have determined growth chronologies for a number of major anticlinal structures, most of which are ˜NE–SW‐trending folds that developed during mild inversion of syn‐rift normal faults or through buckling of the post‐rift succession. These chronologies are supplemented by onshore structural evidence and by thermochronological data from key wells. Whilst our analysis confirms the occurrence of a well‐documented pulse of late Miocene–early Pliocene compression, post‐breakup deformation is not restricted to this time interval. We highlight the growth of a number of structures during the mid‐late Eocene and the Oligocene‐early Miocene, with evidence for considerable temporal and spatial migration of strain within the basin. Our results indicate a long‐lived ˜NW–SE maximum horizontal stress orientation since the mid‐late Eocene, consistent with contemporary stress observations but at variance with previous suggestions that this stress orientation was initiated in the late Miocene by increased coupling of the Australian‐Pacific plate boundary. We attribute the observed record of deformation to a compressional intraplate stress field, coupled to the progressive evolution of the boundaries of the Indo‐Australian Plate, ensuring that this margin has been subject to ongoing compressional forcing since mid‐Eocene breakup. Our results indicate that compressional deformation at passive margins may be more common than is generally assumed, and that passive margin basins with evidence for protracted post‐breakup deformation histories can provide useful natural laboratories for obtaining improved understanding of the evolution of intraplate stress fields over geological timescales.

The Mesozoic‐Cenozoic basins located between the Faroe, Orkney and Shetland Islands along the NE Atlantic Margin are actively explored oil and gas provinces whose subsidence histories are complicated by multiple tectonic factors, including magmatism, inversion and regional‐scale uplift and tilting, that have resulted in spatially variable exhumation. These basins also exhibit nonburial related, transient Cenozoic heating anomalies that make thermal history interpretation and burial history reconstructions problematic. In this study, we have applied a compaction‐based approach, which is less susceptible to distortions from transient heating, to provide new constraints on Cenozoic burial and exhumation magnitudes in the UK sector of the Faroe‐Shetland region using sonic transit time data from Upper Cretaceous marine shales of the Shetland Group in 37 wells. As estimates of exhumation magnitude depend critically on the form of the normal sonic transit time‐depth trend, a new marine shale baseline trend was firstly constructed from shales presently at maximum burial, consistent with other marine shale baseline trends of different ages from nearby basins. Our results indicate that Upper Cretaceous marine shales are presently at or near (i.e. within ≤100 m net exhumation) maximum burial depths in the Møre and Magnus basins in the northeast of the study area as well as in the deeper water Faroe‐Shetland Basin (i.e. Flett and Foula sub‐basins). However, Upper Cretaceous strata penetrated by wells in the southwest have been more deeply buried, with the difference between maximum burial depth and present‐day values (net exhumation) increasing from ca. 200 to 350 m along the central and northeastern parts of the Rona High to ca. 400–1000 m for wells located in the West Shetland Basin, North Rona Basin and southwestern parts of the Rona High. Although the precise timing of exhumation is difficult to constrain due to the complex syn‐ to post‐rift tectonostratigraphic history of vertical movements within the Faroe–Shetland region, our estimates of missing section, together with available thermal history constraints and seismic‐stratigraphic evidence, implies that maximum burial and subsequent exhumation most likely occurred during an Oligocene to Mid‐Miocene tectonic phase. This was probably in response to major post‐breakup tectonic reshaping of this segment of the NE Atlantic Margin linked to a coeval and significant reorganization of the northern North Atlantic spreading system, suggesting that fluctuations in intraplate stress magnitude and orientation governed by the dynamics of plate‐boundary forces exert a major control on the spatial and temporal variations in differential movements along complexly structured continental margin.

Greater India's northern edge prior to collision with Asia is typically modelled as a rifted passive margin. We argue for a quite different geometry as a consequence of two tectonic episodes that happened sometime before the main impact. Whilst the western segment of India's northern boundary had formed in the Late Triassic as a rifted margin, the central and eastern portions developed between 132 and 110 Ma when the sub‐continent separated from Australia–Antarctica as the inner wall of a dextral ‘scything’ transform fault along the Wallaby–Zenith Fracture Zone off western Australia. Key features would have been (i) the very narrow (20–30 km wide) ocean–continent transition zone marking the sub‐continent's eastern northern boundary, and (ii) similar to the region offshore South Africa's Garden Route coast, Greater India's NE corner may have developed a series of ‘perched’ half grabens due to shearing related to its motion along the Wallaby–Zenith Fracture Zone, from initial break‐up until it passed the Zenith Plateau (ca. 110 Ma). Differences in the development of NW Greater India may be reflected in restriction of ultra‐high pressure metamorphic rocks to the western Himalaya where late Paleocene subduction of the rifted passive margin occurred at sub‐equatorial latitudes beneath the intra‐Tethyan arc. Further east, where the margin developed along the scything transform, the continent–ocean boundary would have been more abrupt and probably less strongly welded. Ophiolite emplacement appears to have been penecontemporaneous along the margin. A subsequent slab break‐off episode then eliminated the original plate boundary. Thereafter, remaining oceanic lithosphere north of the arc sutured to the sub‐continent, albeit rather weakly, was consumed beneath Eurasia, culminating in India–Asia collision.

The movement of magma in sedimentary basins often occurs through an extensive and interconnected complex of sills. Field‐, modelling‐, and seismic reflection‐based studies indicate that the emplacement of shallow‐level sills is commonly accommodated by the formation of forced folds, which may be expressed at the free surface and onlapped by younger strata. If the age of these onlapping strata can be constrained, important insights can be gained into the timing of magma emplacement and associated regional, tectono‐magmatic events. Previous studies have focused on isolated intrusions that are overlain by an individual forced fold formed during a single tectono‐magmatic event. However, the structure and evolution of ‘compound’ folds developed above stacked, interconnected sills, and what they may reveal about polyphase intrusive events has not been investigated. In this study, we use 3D seismic reflection data from the Irish Rockall Basin, offshore western Ireland, to constrain the structural style and emplacement history of a sill complex that contains 82 seismically resolved intrusions. Individual forced folds, <41 km² in plan view and with mean fold amplitudes of 111 m, are developed above single intrusions. However, where sills are stacked, broader (100–244 km²), larger amplitude (mean of 296 m) compound folds occur following the coalescence of individual folds. Stratigraphic onlap and truncation observed within the folds throughout the Palaeocene‐to‐Middle Eocene succession, indicates that emplacement and forced folding initiated at the end Maastrichtian and lasted for ca. 15 Ma, before ceasing near the end of the Ypresian. We demonstrate (i) intrusion‐induced forced folds evolve dynamically and can form broad areas of sustained local uplift, and (ii) that the formation of sill complexes within the upper crust may occur over prolonged time periods. This study also highlights the importance of seismic reflection data to understanding the structural style and age relationships between igneous systems.

The SW South China Sea preserves a propagating oceanic spreading centre and associated continent–ocean transition (COT) that characterizes the continental margin offshore SE Vietnam. We investigated the nature of strain accommodation in the region immediately in front of the propagating rift using a combination of 1‐ and 2‐D backstripping subsidence reconstructions, coupled with forward modelling based on the measured upper crustal extensional faulting applied to a flexural cantilever model. Major normal faulting ceases after an inversion event at ca. 16 Ma, although moderate extension was also noted at 22–23 Ma, representing the end of an extensional phase that initiated around 28 Ma. 1‐D subsidence models indicate rapid ‘syn‐rift’ subsidence, possibly lasting until 10 Ma, despite the lack of observed extensional faulting. We show that depth‐dependent extension is required to explain the great depth of the basins despite the modest observed upper crustal faulting. Brittle faulting could not have extended much deeper than 10 km, suggestive of weak crust in the presence of high heatflow. The regional topographic slope on the basement suggests very low mid‐crustal viscosities of 10¹⁹–10²⁰ Pa.s., consistent with the idea that flow in the ductile mid and lower crust was responsible for much of the subsidence prior to, and possibly after, seafloor spreading, which extended ca. 300 km from the tip of the mid ocean ridge. Flow is inferred to be dominant towards the spreading centre prior to 16 Ma. Extension in the COT postdates seafloor spreading and further supports the idea of this crust being very weak, albeit with more coherent, less extended crustal fragments, now forming banks offshore the Sunda continental shelf and surrounded by hyperextended crust of the COT.

A common observation at rifted margins is that the amount of extension measurable from faulting is too little to explain the observed crustal thinning and subsidence. This is the ‘extension discrepancy’. Several causes have been proposed, including depth‐dependent stretching (DDS) or thinning (DDT), sequential faulting, subseismic faulting and polyphase faulting. In this contribution, we explore the different possibilities, with specific reference to the west Galicia margin. If we take the observations at face value, then it seems unavoidable that the upper crust must be stretched and thinned less than the middle and lower crust, i.e. that DDS or DDT has occurred. However it is unclear where the displaced lower crust has gone as there is no inverse discrepancy. Furthermore, independent evidence against large‐scale DDT is provided by seismic velocities and by the occurrence of upper crustal, lower crustal and mantle rocks in close proximity at the deep margin. We thus reject DDT as a sole cause of the extension discrepancy, although recognize that small‐scale local DDT associated with asymmetric faulting is expected. Such small‐scale DDT is an integral part of models of sequential faulting, but these do not predict an overall extension discrepancy, so alone cannot explain one. Subseismic faulting also alone seems inadequate as it cannot explain the magnitude of the extension discrepancy observed at the deep margin. However, as subseismic faulting is a requirement of the fractal distribution of fault sizes and the limited resolution of the seismic method, it must contribute to the extension discrepancy, something that is commonly ignored. Polyphase faulting, in which the thinnest crust has been affected by more than one phase of faulting, resulting in complex and poorly imaged structural architecture, is both predicted at deep margins, and in combination with subseismic faulting, capable of explaining the extension discrepancy. We demonstrate that the west Galicia margin has undergone more than one phase of faulting, with later faults offsetting earlier ones to create complex geometries and, in combination with the expected amount of subseismic faulting, a substantial underestimate of the amount of extension.

Passive‐margin salt basins are classified as prerift, syn‐stretching, syn‐thinning, and syn‐exhumation. Prerift salt, such as the Triassic Keuper in the Western Pyrenees, undergoes thick‐skinned extension, first decoupled and then coupled, along with its substrate and cover. The base salt develops significant relief, is attenuated on the largest faults, and ends up distributed across the entire margin. Syn‐stretching salt, such as along the Iberian and Newfoundland margins, is deposited during early rifting and is thus concentrated in proximal areas with variable thickness and extent, with decoupled and coupled thick‐skinned deformation dominant. Syn‐thinning salt, such as in the northern Red Sea, is also deposited during extension, with the base salt unconformably above proximal stretching faults but offset by distal thinning faults. Both thick‐skinned and gravity‐driven thin‐skinned deformation occur, with the latter strongly influenced by the ramp‐flat geometry of the base salt. Syn‐exhumation salt, such as in the Gulf of Mexico and South Atlantic salt basins, is deposited as part of the sag basin with broad distribution and a generally unfaulted base. Conjugate syn‐exhumation salt basins are originally contiguous, form partly over exhumed mantle on magma‐poor margin segments, and are commonly flanked by magma‐rich segments with volcanic highs (seaward‐dipping reflectors) that isolate the salt basin from marine water. Salt tectonics is characterized by gravitational failure of the salt and overburden, with proximal extension and distal contraction, and the development of allochthonous salt that includes frontal nappes that advance over newly formed oceanic crust.

The transport of sediment from the mouth of the Indus River on to the deep‐water submarine fan is complicated by temporary storage within large clinoforms on the shelf on either side of the submarine canyon, where most of the sedimentation since the start of the Holocene has occurred. In contrast, shelf edge clinoform deltas represent the products of forced regression and not the progradation of highstand clinoforms as far as the shelf edge. Clinoform sediments have a mixed provenance that involves significant reworking of older sediment deposited during or before the last glacial maximum. Recent sedimentation in the canyon head has been very rapid in the last few centuries (ca. 10 cm year⁻¹), but has been starved of sand probably because of 20th century damming. Sandy layers appear to represent annual monsoonal floods with a particularly large flood every 50–70 years. This canyon head sediment is also reworked by currents flowing along the canyon axis before being deposited deeper into the canyon. The last sandy sediment to reach the mid‐canyon (ca. 1300 m depth) was transported around 7000 year BP at a time of rising sea‐levels, and might reflect reworking by the transgression, or local slumping from the walls of the canyon. Dating of the uppermost in a series of terraces in the mid‐canyon area suggests that the canyon may have been partly filled and emptied of sediment at least three times since ca. 50 ka. We conclude from the Holocene record that sediment flux to the deep‐water fan experiences major buffering, reworking and recycling both on the shelf and within the submarine canyon prior to its deposition, so that turbidite sands in the deep Arabian Sea cannot be used to correlate with climatic or tectonic events onshore over timescales of 10³–10⁵ years.

Due to the effects of sediment compaction, thermal subsidence and ‘post‐rift’ fault reactivation, the present‐day geometry of buried, ancient rift basins may not accurately reflect the geometry of the basin at any stage of its syn‐rift evolution. An understanding of the geometry of a rift basin through time is crucial for resolving the dynamics of continental rifting and in assessing the hydrocarbon prospectivity of such basins. In this study, we have restored the Late Jurassic–Early Cretaceous geometry of the southern Halten Terrace, offshore mid Norway, using a combination of well log‐ and core‐derived, sedimentological and stratigraphic data, seismic‐stratigraphic observations and reverse subsidence modelling. This integrated geological and geophysical approach has allowed the large number of input parameters involved in flexural backstripping and post‐rift thermal subsidence modelling to be constrained. We have thus been able to determine the regional structure of the basin at the end of the Late Jurassic–Early Cretaceous rift phase and the associated amount of crustal stretching. Our basin geometry reconstructions reveal that, during the latest syn‐rift period in the Late Jurassic–Early Cretaceous, the Halten Terrace was characterized by a series of isolated depocentres, located between footwall islands, which were not connected into a single depocentre until the Late Cretaceous (Coniacian). We show that two major unconformities, which are now vertically offset by ca. 2 km and located ca. 60 km apart, formed at similar subaerial elevations in the Late Jurassic–Early Cretaceous and were subsequently vertically offset by thermally induced tilting of the basin margin. Cretaceous sediments were deposited in a single, relatively unconfined basin in water depths of 1–1.5 km. The β profile that best restores palaeobathymetry to match our geological constraints is the same as that derived from summing visible post‐Late Triassic heave on faults plus 25–60% additional extension to account for sub‐seismic deformation. This indicates that, at least in the southern part of the Halten Terrace, the amount of upper‐crustal stretching during the Late Jurassic–Early Cretaceous rift phase is comparable to the total amount of lithospheric stretching, supporting a uniform pure‐shear stretching model.