Abstract The detailed analysis of folding in rocks was in part pioneered by John Ramsay, and resulted in a range of techniques and criteria to define folds. Although folding of unlithified or ‘soft’ sediments is typically assumed to produce similar geometries to those in ‘hard rocks’, there has to date been little detailed analysis of such folds. The aim of this paper is therefore to investigate folds developed during soft-sediment deformation (SSD) by applying techniques established for the analysis of tectonic folds during hard-rock deformation (HRD). We use the Late Pleistocene Lisan Formation exposed around the Dead Sea as our case study, as the laminated lake sediments record intricacies of fold detail generated during seismically triggered slumping of mass transport deposits (MTDs) towards the depocentre of the basin. While it is frequently assumed that folds created during SSD are chaotic and form disharmonic structures, we provide analyses that show harmonic fold trains may form during slumping, although larger upright folds cannot be traced for significant distances and are more typically disharmonic. Our analysis also reveals a range of fold styles, with more competent detrital-rich layers displaying buckles (Class 1B), as well as upright Class 1A folds marked by thickened limbs. Class 1A buckle folds are generally considered to be created by flattening that overprints folds with an original Class 1B geometry. As thickened fold limbs are truncated by overlying erosive surfaces, the vertical flattening is considered to have occurred during the slump event. Different fold shapes may partially reflect variable flattening, depending on the original orientation of upright or recumbent folds, together with continued downslope-directed simple-shear deformation that modifies the fold geometry. Analysis of fold wavelength, amplitude and bed thickness allows us to plot strain contour maps, and indicates that beds defining slump folds display viscosity contrasts in the range of 50–250, which are similar to values estimated from folds created during HRD in metamorphic rocks. A range of refold patterns, similar to those established by John Ramsay in metamorphic rocks, are observed within slumps, and are truncated by the overlying sediments, indicating that they formed during a single progressive slump event rather than distinct ‘episodes’ of superimposed deformation. This study confirms that techniques developed for the analysis of folds created during HRD are equally applicable to those formed during SSD, and that resulting folds are generally indistinguishable from one another. Extreme caution should therefore be exercised when interpreting the origin of folds in the rock record where the palaeogeographical and tectonic contexts become increasingly uncertain, thereby leading to potential misidentification of folds created during SSD.

Abstract Salt is a crystalline aggregate of the mineral halite, which forms in restricted environments where the hydrodynamic balance is dominated by evaporation. The term is used non-descriptively to incorporate all evaporitic deposits that are mobile in the subsurface. It is the mobility of salt that makes it such an interesting and complex material to study. As a rock, salt is almost unique in that it can deform rapidly under geological conditions, reacting on slopes ≤0.5° dip and behaving much like a viscous fluid. Salt has a negligible yield strength and so is easy to deform, principally by differential sedimentary or tectonic loading. Significant differences in rheology and behavioural characteristics exist between the individual evaporitic deposits. Wet salt deforms largely by diffusion creep, especially under low strain rates and when differential stresses are low. Basins that contain salt therefore evolve and deform more complexly than basins where salt is absent. The addition of halokinetic processes to the geodynamic history of a basin can lead to a plethora of architectures and geometries. The rich variety of resultant morphologies have considerable economic as well as academic interest. Historically, salt has played an important role in petroleum exploration since the Spindletop Dome discovery in Beaumont, Texas in 1906. Today, much of the prime interest in salt tectonics still derives from the petroleum industry because many of the world's largest hydrocarbon provinces reside in salt-related sedimentary basins (e.g. Gulf of Mexico, North Sea, Campos Basin, Lower Congo Basin, Santos Basin and Zagros). An understanding of

Abstract Halokinetic sequences are unconformity-bound packages of thinned and folded strata adjacent to passive diapirs. Hook halokinetic sequences have narrow zones of deformation (50–200 m), >70° angular discordance, common mass-wasting deposits and abrupt facies changes. Wedge halokinetic sequences have broad zones of folding (300–1000 m), low-angle truncation and gradual facies changes. Halokinetic sequences have thicknesses and timescales equivalent to parasequence sets and stack into composite halokinetic sequences (CHS) scale-equivalent to third-order depositional cycles. Hook sequences stack into tabular CHS with sub-parallel boundaries, thin roofs and local deformation. Wedge sequences stack into tapered CHS with folded, convergent boundaries, thicker roofs and broad zones of deformation. The style is determined by the ratio of sediment-accumulation rate to diapir-rise rate: low ratios lead to tabular CHS and high ratios result in tapered CHS. Diapir-rise rate is controlled by the net differential load on deep salt and by shortening or extension. Similar styles of CHS are found in different depositional environments but the depositional response varies. CHS boundaries (unconformities) develop after prolonged periods of slow sediment accumulation and so typically fall within transgressive systems tracts in shelf settings and within highstand systems tracts in deepwater settings. Sub-aerial settings may lead to erosional unroofing of diapirs and consequent upward narrowing of halokinetic deformation zones.

Abstract La Popa Weld in La Popa Basin, Mexico, is a 24 km long near-vertical structure with a prominent bend approximately halfway along its length. Halokinetic folding, local unconformities and diapir-derived detritus in flanking strata document a precursor salt wall. Shortening during the latest Cretaceous to Eocene Hidalgoan Orogeny squeezed the salt wall to form the weld. Deformation varies significantly along the weld. The northwestern third has remnant gypsum (including a diapir at the northwestern end), little large-scale folding of flanking strata and only background fracture intensity. Directly NW of the bend are pods of gypsum linked by complete welds, a large-scale cuspate anticlinal geometry and significant fracturing within 5–10 m of the weld. The southeastern half is completely welded with no remnant gypsum, a prominent cuspate anticlinal geometry and a 50 m wide damage zone. The variable deformation was controlled by the original width of the salt wall and the amount and direction of shortening. Where orthogonal to the wall, shortening locally closed the diapir but little further deformation took place. Where oblique, shortening caused post-weld dextral strike-slip movement and significant fracturing and shearing of the wall rock. The resulting deformation variability likely impacted the sealing capability of the weld.

Abstract The Eocene Carroza Formation in La Popa Basin, Mexico, represents fluvial sedimentation in a shortening-influenced salt-withdrawal minibasin, termed the Carroza Syncline. The Carroza Syncline lies adjacent to the La Popa salt weld, which was formerly a passively-rising salt wall that was shortened during the Hidalgoan Orogeny in Late Cretaceous and Palaeogene time. The Carroza Formation displays distinct upsection changes in fluvial facies distribution and geometry of halokinetic drape folding. Fluvial channel distribution changes upwards from widespread thin, broad channels with variable palaeocurrents in the lower part of the formation to thick, stacked channels concentrated in the hinge of the Carroza Syncline with weld-parallel palaeocurrent directions in the upper part. The upper and middle members of the Carroza contain debris-flow facies derived from diapir roof strata and the diapir itself. The style of halokinetic drape fold upturn and thinning towards the weld changes upsection from a broad (800–1500 m) to a narrow (50–200 m) zone, where upper Carroza strata are overturned and in direct contact with remnant gypsum along the weld. The upsection changes in fluvial facies distribution and geometry reflect an overall decrease in local sediment-accumulation rates relative to salt-rise rates controlled by both Hidalgoan shortening and passive diapirism.

Abstract Parts of two third-order Neoproterozoic (Marinoan) depositional sequences are documented in the Wilpena Group (Wonoka Formation and Bonney Sandstone) at Patawarta diapir, located in the central Flinders Ranges, South Australia. These sequences represent an overall regressive succession transitioning upwards from outer to middle wave-dominated shelf deposits to a tidally dominated barrier bar to coastal plain. The lower, middle, upper limestone and green mudstone informal members of the Wonoka Formation comprise the Highstand Systems Tract of the lower sequence. The Sequence Boundary is at the top of the Wonoka green mudstone member and is overlain by the Lowstand Systems Tract of the upper sequence, which includes the lower dolomite, sandstone and upper dolomite beds of the Patsy Hill Member of the Bonney Sandstone. The upper sequence Transgressive Systems Tract comprises the Bonney Sandstone. These units comprise one complete tapered composite halokinetic sequence (CHS). The lower halokinetic-sequence boundary is associated with the Maximum Flooding Surface of the lower depositional sequence and the upper halokinetic-sequence boundary is interpreted as the Transgressive Surface of the overlying depositional sequence where an angular truncation of up to 90° is documented.

Abstract Isotopic and fluid inclusion analyses of veins and host rocks constrain the compositions, temperatures and sources of palaeofluids along the La Popa salt weld. Most veins formed after the salt was evacuated from the precursor salt wall; veins are generally more abundant on the downthrown side of the weld and near a significant bend in the trace of the weld. The spatial distribution of fluid types and temperatures suggests the weld served as a vertical fluid conduit and a horizontal baffle. Stable isotopes indicate there was significant fluid–rock interaction and little vertical fluid communication between rock units in areas away from the weld. Fluid temperatures along the weld ranged from 84 to 207 °C, salinities ranged from 4 to 25 wt% NaCl equiv. and methane was abundant in the weld zone and on the downthrown side of the weld. Strontium isotopes suggest that some of the vein-forming fluids were derived from the evaporites that once occupied the weld. Our results suggest the sealing potential of similar welds may be related to the presence of abrupt changes in weld geometry such as cusps or bends, the amount of shortening across the weld and the amount of vertical displacement across the weld.

Abstract This work addresses the geological and geophysical interpretation of salt structures in selected Brazilian sedimentary basins, from intracratonic Palaeozoic evaporites in the Amazon and Solimões basins to divergent margin evaporite basins formed during the Mesozoic break-up of Gondwana. There is an intriguing correlation between evaporite basins and hydrocarbon accumulations in all the selected basins discussed. The Solimões and Amazonas basins developed evaporite depositing environments as the Middle Carboniferous sea was closing during a plate convergence phase. The salt basin along the eastern Brazilian and western African margins developed along the Mesozoic rifts of the South Atlantic. Regional seismic interpretation and potential field (gravity and magnetic) data over the eastern Brazilian and west African margins suggest a very thick autochthonous salt layer deposited over rifted continental crust and particularly above the thick sag basin sediments over the hyperextended crust that marks the transition from continental to oceanic crust. Most of the hydrocarbon discoveries in the eastern Brazilian and western African margins are in post-salt turbidite and carbonate reservoirs, but recent discoveries in the deepwater salt basins along the southeastern Brazilian margin indicate that pre-salt plays will represent an important contribution to hydrocarbon production in the near future.

Abstract The southern Brazilian salt basin, comprising the three sub-basins Santos, Campos and Espirito Santo, was deposited over a pre-existing rifted basin with c. 1–2 km of relief bordered by an outer basin high that separated the basin from the conjugate African margin. The evaporites are interpreted to have been deposited very rapidly (<1 Ma) during the waning of extension. Deposition of salt caused rapid loading of the basin, so that further basin subsidence occurred and mobile salt drained from structurally higher zones into the subsiding basins. Seismic evidence indicates that downslope salt drainage occurred before any sediment overburden accumulated. Withdrawal synclines within salt units developed adjacent to diapirs, which have intruded the evaporite sequence, and salt extrusions are observed which were buried by later salts. The early movement of the salt probably contributed to significant fault reactivation and redistribution of salt load, so that the final half-graben salt fill reached up to 4.5 km thick where only 1–2 km of salt was originally deposited.

Abstract Salt–sediment interplay in the Santos Basin is investigated integrating seismic interpretation, kinematic restoration and analogue modelling. Deformation within the post-salt sequence results from thin-skinned gravitational gliding and spreading, driven primarily by halokinesis, greatly affected by massive sediment inflows. The impressive landward-dipping listric Cabo Frio Fault controls the major depocentres updip, whereas salt-cored folds accommodate downdip shortening. Sediment supply from confluent directions creates a complex interference pattern of superposed folds with intervening polygonal minibasins. A new structure is identified (termed the ‘Ilha Grande Gravitational Cell’), a linked system of updip extension and downdip contraction detached on salt, comprising the Cabo Frio Fault and minibasins. It moves to the SE, with eastern and western borders suggesting lateral gradients of slippage. This thin-skinned feature results from the differential load imposed by a thick prograding wedge over the ductile salt and is independent of pre-salt structures. The post-salt sequence moves basinwards due to halokinesis, thereby changing position relative to the pre-salt sequence, which implies that any present-day correspondence between pre- and post-salt structures may not attest to linkage in the past. Application of kinematic restoration techniques allows the true position and geometry of the key elements through time, improving petroleum systems assessment.

Abstract Salt flows downslope, irrespective of overburden. In salt basins on passive margins, the salt will tilt and flow towards the ocean immediately after continental rifting has ended due to thermal subsidence. Using real examples, as well as physical and numerical models, tilting is shown to be relatively rapid, enhanced by isostatic rebound updip and loading downdip where salt pools and inflates behind an outer high. In the Santos, Campos and Kwanza basins, this outer high is represented by an embryonic mid-Atlantic ridge, amplified in height by the differential weight of the inflating salt. Widespread extension and translation of overburden, utilizing both seaward- and landward-dipping normal faults, characterizes the early evolution of the inboard region. Inflation and contraction occur outboard, the effects of which tend to expand in a landward direction over time. Rapid accumulation of salt implies wholesale dewatering of pre-salt sediments, the water possibly permeating the salt once it has reached a burial depth of c. 3 km. The process of thermal subsidence, salt drainage and isostatic amplification is an efficient mechanism for moving sediment on passive margins tens of kilometres seaward during a relatively short period and helps explain why great thicknesses of salt can accumulate there in the first place.

Abstract This paper describes a common type of salt wall found in extensional regimes which possess the characteristics: cover strata truncated against both flanks; an asymmetric appearance in cross-section caused by normal fault-related growth patterns; and at least one unconformity and onlap surface separating strata which are tilted in opposite directions. This type of structure evolves by a process known as flip-flop salt tectonics starting with a roller where a normal fault detaches down one flank of the embryonic salt body. The structure grows as salt flows towards the low-stress zone below the crest of the footwall causing it to swell and tilt backwards until it becomes gravitationally unstable, until the cover strata on one or both sides welds out or until the salt emerges at surface. Further growth is then accommodated by switching to a new counter-dipping fault that detaches on the opposite flank of the salt body leading to a flip in hanging-wall/footwall polarity marked by an unconformity and onlap surface. The salt body continues to grow beneath the new footwall, causing partial inversion of the old hanging wall. Additional switches may occur, leading to tall flip-flop structures until the source of salt is depleted.

Abstract The kinematics of regional-scale salt flow in the northern Gulf of Mexico is analysed using: (i) a map of shelf-break contours at the termination of successive depositional episodes; (ii) the location and geometry of large-scale structures of the slope domain as imaged by seismics; and (iii) digital slope bathymetry. In the north margin, salt has flowed towards the SW since the Cretaceous with three main stages of development prior to, during and after a massive salt extrusion in the Early Miocene time. The corresponding sequence of structural development is discussed using a laboratory model. Contrary to all previous interpretations that invoked sedimentary loading as the main driving force, the analysis of regional-scale salt flow implies that the salt tectonics of the northern Gulf of Mexico is predominantly controlled by gliding above the margin dip. The SW-directed salt flow indicates that the north margin of the northern Gulf of Mexico trends NW–SE, in agreement with plate kinematic models in which the Yucatan continental block has undergone a 45–60° dextral rotation relative to its present orientation.

Abstract In Magnolia Field, deepwater sediments were affected during deposition by allochthonous salt. Pleistocene channel systems developed on a salt flank and were initially deeply incised close to the salt but progressively avulsed down the lateral slope, each time with decreasing depth of incision. Following this degradational stage, a lobe developed on top of the channel fills and a large-scale aggradational system developed. A conceptual model of submarine channel development adjacent to active topography has been developed from this dataset. Channels may become deeply entrenched during stages of salt growth, but only where flow frequency and magnitude are sufficient to outpace topographic growth. Where flows are less frequent topographic growth may present a barrier to successive flows, causing avulsion. The large-scale cycles of salt growth and withdrawal commonly recognized in subsurface systems, combined with eustatic sea-level changes, may result in a cyclic style of evolution whereby channels initially become entrenched and/or step away from the growing topography, switching to backfilling as salt growth slows or pauses, followed by a distributive-style as the entire system backsteps. During salt withdrawal the equilibrium profile may become relatively raised and channels may develop an aggradational style. In these settings, significant cross-channel facies asymmetry may result.

Abstract Two-dimensional plane-strain numerical experiments illustrate the effects of variable evaporite viscosity and embedded frictional-plastic sediment layers on the style of salt flow and associated deformation of the sedimentary overburden. Evaporite viscosity exerts a first-order control on the salt flow rate and the style of overburden deformation. Nearly complete evacuation of low-viscosity salt occurs beneath expulsion basins, whereas significant salt is trapped when viscosity is high. Embedded frictional-plastic sediment layers with yield strength partition salt flow and develop transient contractional structures (folds, thrust faults and folded faults) in a seaward salt-squeeze flow regime. Multiple internal sediment layers reduce the seaward salt flow during sediment aggradation, leaving more salt behind to be remobilized during subsequent progradation. This produces more seaward extensive allochthonous salt sheets. If there is a density difference between the embedded layers and the surrounding salt, then the embedded layers fractionate during deformation and either float to the surface or sink to the bottom, creating a thick zone of pure halite. Such a process of ‘buoyancy fractionation’ may partially explain the apparent paradox of layered salt in autochthonous salt basins and pure halite in allochthonous salt sheets. Supplementary material Animated gif files of the model results are available at http://www.geolsoc.org.uk/SUP18500 .

Abstract This study integrates seismic interpretation and 3D analogue experiments monitored by digital image correlation techniques to investigate the evolution of the salt structures and the related depositional systems in the Laurentian Basin offshore Atlantic Canada. During the late Triassic, a layer of more than 3 km thick salt was deposited locally in a set of interconnected rift half-grabens forming a 50–70 km wide evaporite basin in the northern part of the Scotian Basin salt provinces. High sediment input in the Jurassic and early Cretaceous mobilized the salt into complex salt tectonic features, which suggest four kinematic domains with: (1) salt welds and pillows; (2) extensional diapirs and canopies; (3) contractional diapirs and folds; and (4) allochthonous salt nappe. The landward grabens trapped most of the Early Jurassic sediments by passive downbuilding into the salt with local extension. The expelled salt has been evacuated basinwards into a large contractional salt massif. The rapid advance of the allochthonous nappe was coeval with the Late Jurassic extensional collapse of the inflated salt massif due to seaward sediment progradation. Late Cretaceous and Tertiary progradation over the salt nappe caused extensional deformation with growth faulting and formation of minibasins on the secondary salt detachment level.

Abstract The Late Jurassic–Cretaceous Parentis Basin (Eastern Bay of Biscay) illustrates a complex geological interplay between crustal tectonics and salt tectonics. Salt structures are mainly near the edges of the basin, where Jurassic–Lower Cretaceous overburden is thinner than in the basin centre and allowed salt anticlines and diapirs to form. Salt diapirs and walls began to rise reactively during the Late Jurassic as the North Atlantic Ocean and the Bay of Biscay opened. Some salt-cored drape folds formed above basement faults from the Upper Jurassic to Albian. During Albian–Late Cretaceous times, passive salt diapirs rose in chains of massive salt walls. Many salt diapirs stopped growing in the Mid-Cretaceous when their source layer depleted. During the Pyrenean orogeny (Late Cretaceous–Cenozoic), the basin was mildly shortened. Salt structures absorbed almost all the shortening and were rejuvenated to form squeezed diapirs, salt glaciers and probably subvertical welds, some of which were later reactivated as reverse faults. No new diapirs formed during the Pyrenean compression, and salt tectonics ended with the close of the Pyrenean orogeny in the Middle Miocene. Using reprocessed industrial seismic surveys, we document how salt tectonics affected the structural evolution of this offshore basin largely unknown to the international audience.

Abstract The Permian–Cretaceous Polish Basin belonged to the system of epicontinental depositional basins of Western and central Europe and was filled with several kilometres of siliciclastics, carbonates, and also thick Zechstein (approximately Upper Permian) evaporites. Its axial part (the so-called Mid-Polish Trough) characterized by the thickest Permo-Mesozoic sedimentary cover, developed above the Teisseyre–Tornquist Zone, lithospheric-scale boundary separating the East European Craton and the Palaeozoic Platform. The Polish Basin was inverted in Late Cretaceous–Paleocene times. A synthesis of studies based on seismic reflection data allowed some general rules regarding salt tectonics of the Polish Basin to be formulated. Two general classes of structures genetically related to the presence of the Zechstein evaporites have been described: peripheral structures located within NE and SW flanks of the Polish Basin, outside its axial part and structures located within its axial part. The first class of structures includes grabens bounded by listric faults detached above salt or salt pillows that developed where Zechstein evaporites were of relatively smaller thickness and where sub-Zechstein fault tectonics played a relatively smaller role. The second class of structures includes more mature salt structures such as salt pillows and salt diapirs and is related to the more axial part of the basin, characterized by relatively thicker Zechstein evaporites and by more intense basement tectonics. First salt movements (salt pillowing) took place in the Early Triassic that in certain cases was followed by the Late Triassic salt diapirism and extrusion. In Jurassic–Early Cretaceous times, no significant growth of salt structures took place. Most of the salt diapirs have been finally shaped by the Late Cretaceous inversion tectonics. Some salt diapirs also underwent Cenozoic reactivation, associated with localized Oligocene or Miocene subsidence that in some cases was followed by younger (Pliocene–Quaternary) inversion and uplift.

Abstract Salt diapirs preferably rise above basement faults in extensional basins. A series of analogue and numerical models were developed in order to assess the supply of salt from the footwall and hanging wall to a diapir and to study the influence of basin inversion on the diapir development. The modelling scenario was based on the Kłodawa Salt Structure evolution (central Poland). The experiments show that the ductile material derived from the footwall constitutes the dominant portion of the diapir developed due to model extension, and this material occurs both in the footwall and hanging wall parts of the diapir. Shortening of the analogue models resulted in thinning of the diapir and shifting its stem onto the footwall. Ductile material become redistributed inside the diapir, however footwall material still prevails in the diapir structure. Results from the numerical models show that the magnitude of the basement fault governs the amount of salt supply to a diapir across the fault and that there is a differential salt supply from the hanging wall and footwall with time.