SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

We used seven scaled physical models to explore the near-surface structural evolution of shallowly buried, actively rising salt stocks. The models consisted of dry sand, ceramic microspheres and silicone. Previously dormant stocks rose because of lateral squeezing or pumping of salt from below. The pressure of rising salt created a dynamic bulge in the crest of the diapir, which arched the overlying roof sediments. Eventually this dynamic bulge collapsed and its overlying roof broke into rafts along subradial grabens. The rafts were dispersed outwards by shear traction of spreading salt, surmounting an upturned collar of country rock and eventually grounding at the front of the extrusive flow. Flow of salt around these stranded fragments created a lobate extrusion front, common in submarine salt sheets in the Gulf of Mexico and subaerial salt glaciers in Iran. Stock geometry, regional dip and roof density affected extrusion rates and spreading directions. Stocks leaning seaward extruded salt faster and farther than did upright stocks. Dense roofs foundered and plugged the vent, limiting surface extrusion. In tilted models, broad salt sheets spread asymmetrically downslope. Stock contents were inverted within the extruded salt sheet: successively deeper parts of the stock's core rose to the surface and overran salt extruded from the shallower parts of the diapir. As shortening continued, salt from the source layer reached the surface after being driven out by thrusting. A central thrust block, or primary indenter, moved ahead of surrounding thrust blocks, impinging against and squeezing the stock into an elliptical planform. After high shortening, secondary indenters converged obliquely into the salt stock, expelling salt from the periphery of the diapir. The models shed light on (1) the origin and fate of large rafts or carapace blocks atop allochthonous salt, (2) cuspate margins of salt sheets and (3) interaction of thrusting, diapir pinch-off and emplacement of allochthonous salt sheets.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

Active rise of diapirs through thin or negligible roofs is the precursor to salt extrusion. Halokinetic active diapirism is driven solely by buoyancy, so it requires the overburden to have a bulk density greater than that of the underlying salt. But a density inversion alone is insufficient if a mechanically competent roof resists diapir rise. An active diapir can rise only if driving forces exceed resisting forces. Analytical modelling by Schultz-Ela et al. (1993) shows that buoyancy-driven active diapirism is promoted by increasing density contrast, overburden thickness, roof width and fault dip, as well as by decreasing roof thickness and coefficient of friction. The minimum thickness of roof that can be lifted by a halokinetic active diapir under given conditions is its threshold thickness. Roofs thicker than this cannot be arched or pierced by halokinetic rise.

These physical limitations can be largely overcome by regional compression, which can rejuvenate a halokinetically stable diapir having a roof too thick to raise by buoyancy alone (Letouzey et al., 1995; Vendeville & Nilsen, 1995; Letouzey & Sherkati, 2004; Dooley et al., 2009; Callot et al., 2012). Rock salt has negligible yield strength and low-effective viscosity (1017–1019 Pa s; Van Keken et al., 1993), so it deforms more easily than most other rocks. Potash salts are even weaker. Shortening can also rejuvenate a diapir already actively growing by buoyancy. Lateral compression adds a convergent displacement load to the gravitational load. Thus under lateral compression, the sides of the diapir converge, displacing the upper diapiric salt upwards. Under this displacement loading, a pressurized diapir can arch and pierce even a thick roof. This topographical bulge is also prone to erosion, which further unroofs the salt and enhances salt rise. A diapir responds to compression long before any contractional structures appear in the surrounding country rock, although lateral compaction is likely (e.g. Koyi et al., 2004; Dooley et al., 2009). If the roof is too thick to be pierced or is thickened and strengthened by high-level thrusting during lateral compression, diapiric salt can also be displaced downwards (Dooley et al., 2009).

If an active diapir breaks through a thin roof, it extrudes salt. Where salt extrusion outpaces dissolution, it spreads laterally as an allochthonous salt sheet over younger strata. These sheets are known in about 40 basins worldwide (Fig. 1; Hudec & Jackson, 2006). Buried allochthonous salt sheets are most common in the northern Gulf of Mexico (Figs 2 and 3). These sheets are almost entirely buried, some of them so deep that most of the salt has been expelled from them to higher levels, leaving salt welds (e.g. Worrall & Snelson, 1989; Diegel et al., 1995; Peel et al., 1995; McDonnell et al., 2009, 2010). Even though allochthonous salt in the Gulf of Mexico is not now extruding, much has been learnt about salt extrusion from four types of evidence: (1) topography of shallowly buried salt sheets in the Gulf of Mexico (e.g. Schuster, 1995; Rowan et al., 1999; Pilcher et al., 2011; Rowan & Inman, 2011), (2) inferences from seismic data of deeper salt sheets (Fig. 3; e.g. Peel et al., 1995; Rowan et al., 2004; Mount et al., 2006; Radovich et al., 2007; McDonnell et al., 2009, 2010), (3) field geology and surveying of spectacularly exposed subaerial salt glaciers (namakiers) in Iran (e.g. Talbot & Jarvis, 1984; Talbot, 1998; Talbot & Pohjola, 2009; Aftabi et al., 2010; Baikpour et al., 2010; Baikpour & Talbot, 2012) and (4) physical models (e.g. Gilbert & Merle, 1987; Jackson & Talbot, 1989; Cobbold & Szatmari, 1991; Koyi, 1996; Talbot & Aftabi, 2004; Gaullier & Vendeville, 2005; Vendeville, 2005; Rowan & Vendeville, 2006; Aftabi et al., 2010; Dooley et al., 2012).

image

Figure 1. World map of basins containing allochthonous salt. Modified from Hudec & Jackson (2006). AG, Agadir-Tarfaya; AL, Atlas; AQ, Aquitaine; AT - Afghan-Tajik; BG, Benguela-Namibe; CP, Campos; CR, Carpathian; CT, Cantabrian-West Pyrenees; DD, Dnieper-Donets; EN, Eritrean; ER, Essaouira; ES, Espirito Santo; FL, Flinders; GB, Guinea-Bissau; GC, Gulf Coast-Gulf of Mexico; GK, Great Kavir-Garmsar-Qom; GN, Gabon; GQ, Guadalquivir; KT, Katanga; KQ, Kuqa; KZ, Kwanza; LC, Lower Congo; MJ, Majunga; MT, Mauritania; OU, Oriente-Ucayali; PC, Precaspian; RM, Rio Muni; SF, Safi; SI, Sivas; SK, Somali-Kenya; SL, Salina-Sigsbee; SN, Santos; SR, Salt Range; SS, Scotian Slope; SU, Suriname; SV, Sverdrup; YE, Yemeni; ZG, Zagros; ZQ, Zipaquira; ZS, Zechstein.

Download figure to PowerPoint

image

Figure 2. Shaded-relief bathymetric image of part of Sigsbee salt canopy, north-central Gulf of Mexico. Bathymetry from Bryant & Liu (2000). Key to Gulf of Mexico protraction areas: EB, East Breaks; GB, Garden Banks; AC, Alaminos Canyon; KC, Keathley Canyon; GC, Green Canyon; AV, Atwater Valley; WR, Walker Ridge; LU, Lund.

Download figure to PowerPoint

image

Figure 3. Geoseismic section from Northern Gulf of Mexico illustrating squeezed salt stocks and secondary welds below a canopy of allochthonous salt. Salt from squeezed diapirs fed the growing canopy above. Subsequent sedimentary loading remobilized canopy salt, driving it seawards and forming tertiary welds between diapirs. Modified from Mount et al. (2006).

Download figure to PowerPoint

The sequence of events that emerges from these studies begins with a buried diapir that is rising sluggishly or static because its source layer is depleted or because its roof is too strong. Rejuvenated by lateral squeezing imposed by gravity or orogeny, the diapir begins to rise actively. This rise is almost as abrupt as the start of regional compression. The active diapir arches then bursts through its roof. The extruding salt spreads as a salt glacier over the surrounding area to form an allochthonous sheet. Such sheets can approach and coalesce (suture) with neighbouring sheets to form a salt canopy (e.g. Jackson & Talbot, 1989; Jackson et al., 1990; Dooley et al., 2012). This sequence of events prompts questions fundamental to understanding salt diapirism, formation of allochthonous salt, and their interaction with regional shortening.

  • What happens to the roof as the diapir inflates and eventually breaks out and begins to spread? How does this activity affect the spreading salt? What happens to the rafted fragments?
  • What is the source of the extruding salt? How much is from the diapir, and how much is from the deep source layer and how did this balance change over time?
  • How does the diapir close? What is the relation between welds and thrusts? How far from the diapir can its effects be recognized?

Answering these questions is difficult from studying natural examples. Physical modelling has several advantages: (1) a full evolution, rather than a snapshot in time, (2) three-dimensional geometry, (3) material properties, internal and external stratigraphy and boundary conditions are all known. In the models featured in this paper, rise of diapiric stocks was activated by either pressurized salt pumped from below or by far-field compression imposed by a moving endwall (Fig. 4). This lateral compression simulates either (1) orogenic collision and convergence or (2) gravity acting on downdip parts of passive margins. Lateral shortening helps allochthonous salt sheets extrude but is not necessary everywhere. For example, an emergent diapir fed by a sufficiently thick source layer could extrude merely because of a hiatus in sedimentation. Alternatively differential loading of a salt canopy by uneven sedimentation could expel a younger generation of allochthonous salt. Our modelled stocks were emergent or only thinly covered. We focus on (1) the importance of a dynamic bulge generated by the pressure of moving salt in a rising diapir; (2) the vulnerability of thin roofs of active diapirs, which break up and are dispersed by extrusive flow as the underlying dynamic bulge collapses outwards; (3) relative contributions of salt expelled from the diapir and salt from the source layer; and (4) interactions of the closing diapir with an engulfing thrust belt. Throughout the paper, our model results are applied to natural examples, mostly from two basins in Iran and from the northern Gulf of Mexico.

image

Figure 4. Experimental setups for physical models illustrated in this paper (summarized in Fig. 5). (a) Map view of squeezed-diapir experiments; model rig dimensions were 130 × 60 × 20 cm; source layer of salt pinched out seaward. Endwall moving at 1 cm day−1 shortened the model and squeezed the stock, causing stock to extrude as its sides were pressed together. Outcrop of cylindrical salt stock was elliptical because stock was tilted. (b) Cross section of prebuilt stocks, which were vertical or tilted seaward, as common for downbuilding diapirs. Passive markers in source layer and diapir tracked flow patterns during deformation. (c) Underside view of Model 2 prior to building salt stock and adding of surrounding overburden. Note locations of undeformed passive markers in source layer (plugs 1–8). Blue passive marker layer C forms base of stock. (d) Cross section for setup for Models 1 and 3 (Fig. 5). A pump forced salt analogue up a vertical feeder into prebuilt flaring diapirs that were buried by roofs of varying thickness.

Download figure to PowerPoint

A companion series of models by Dooley et al. (2009) also squeezed diapiric stocks by a moving endwall, but the scope of that study was different. That paper focused on diapirs that were deeply buried and that intruded diapiric salt back into the deep source layer. Because of the thick roof, negligible salt escaped to the surface, even during advanced shortening.

Model Methodology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

Materials

As is common for physical models of salt tectonics, we simulated rock salt using ductile silicone polymer (polydimethylsiloxane, trade named SGM36; Weijermars, 1986) and its siliciclastic overburden using brittle, dry, granular material (Table 1). Like the squeezed-stock models in Dooley et al. (2009), the sedimentary overburdens in our squeezed-stock models had a realistic bulk density 1.1 times that of the silicone simulating salt. This density ratio was achieved by varying the proportions of silica sands and ceramic microspheres in the brittle overburden (see Dooley et al., 2009; for further details). In one of the experiments (Model 7, Fig. 5), we increased density of the prekinematic roof to examine how this increase affected surface extrusion.

Table 1. Dynamic scaling properties of the models
QuantityModelNature (prototype)Model ratio
Length, stock minor axism = 21 cm = 0.21 m0 = 5.25 km = 5250 mr = 4 × 10−5
Displacement, regionalm = 30 cm = 0.3 m0 = 7500 mr = 4 × 10−5
Density, saltρm = 970 kg m−3ρ0 = 2200 kg m−3ρr = 0.44
Density, overburdenρm = 1070 kg m−3ρ0 = 2400 kg m−3ρr = 0.44
Acceleration, gravityam = 9.8 m s−2a0 = 9.8 m s−2ar = 1
Strain  εr = 1
Stress or pressure  σr = ρrr a = 1.8 × 10−5
Viscosity dynamic, saltμm = 2.5 × 104 Pa sμ0 = 1018 Pa sμr = 2.5 × 10−14
Timetm = 450 h = 1.6 × 106 st0 = tm/tr = 1.1 × 1015 s = 36 Matr = μr∕/σr = 1.4 × 10−9
image

Figure 5. Chart summarizing parameters of the seven physical models. The models are numbered in the order they are introduced in the main text.

Download figure to PowerPoint

Design

Setups for the two types of active diapirs presented in this paper are illustrated in Fig. 4, and parameters for the seven experiments are summarized in Fig. 5. For pump-driven diapirs (Models 1 and 3, Fig. 5), a motor-driven pump was attached below the vertical feeder and forced the salt analogue up into an upward-flaring diapir 4-cm tall (Fig. 4d). Regional shortening drove all our other models of active diapirism. The primary squeezed stock (Model 2, Fig. 5) simulated the effects of increasing shortening on an inclined stock. This prebuilt stock was 8-cm tall and tilted 45° away from a moving endwall that compressed the model at a constant rate of 0.07 cm h−1 (Fig. 4a). Inclined stocks are common where uneven aggradation rates asymmetrically mould the shape of growing passive diapirs. All the squeezed-stock models were built in a deformation rig 130-cm long and 60-cm wide (Fig. 4a). In Model 2 the base of salt was horizontal to exclude effects of downdip flow, unlike the other squeezed-stock models described here. In all our squeezed-stock models, the source layer had a tabular thickness of 1–2 cm and pinched out at a distal 20° ramp. To ensure reproducibility, we built stocks in steps using moulds 0.5–1.0-cm thick, encased by the overburden. To reduce edge effects, the base of the stock was positioned at least 60 cm from the moving endwall and 25 cm from the sidewalls. For the same reason, source-layer salt was purposely decoupled from the rig sidewalls by strike-slip faults that formed spontaneously between the source layer and lateral sand buffers.

In overhead views of the squeezed-stock models, a north arrow on each map view provides an arbitrary reference direction. The foreland is to the west, and the hinterland is to the east. In a gravity-driven system on a passive margin, foreland and hinterland equivalents are generally seaward and landward, respectively.

Data capture, visualization and interrogation

Computer-controlled cameras photographed the obliquely lit upper surface of the models at set time intervals. Laser scanning mapped changing structural topography on a millimetre scale. Each scan produced more than 8 × 106 points to render high-resolution, interactive, 3-D surfaces. Analysis of archival imagery by digital image-correlation software tracked the surface-strain history and displacement vectors.

After deformation ended, we turned the model over and photographed its underside. In these upward views, passively deformed marker plugs and marker discs tracked salt flow within the source layer (Fig. 4b and c; cf. Dooley et al., 2009). Each marker comprised silicone polymer mixed with minute quantities of coloured pigments. These passively deforming markers were embedded as either vertical cylindrical plugs within the source layer or horizontal discs at the base, middle and top of the salt stock. Digital photographs of closely spaced serial sections (≤5 mm thick) yielded a 3-D voxel model of orthogonal cross sections and depth slices.

Scaling

Dynamic scaling of the models in dimensions of length, mass and time is shown in Table 1 (following Ramberg, 1981). Model ratios are model quantity divided by equivalent quantity in nature. Model ratio for acceleration due to gravity is 1 because both model and nature deform under 1 G. Model strain ratio is also 1 because the model is used to simulate the same natural strain. Most natural stocks are 2–8-km wide, excluding any allochthonous fringe. If we assume an average width of a stock (minor axis if elliptical) of 5.25 km, it yields a convenient length ratio of 4 × 10−5. Diapir heights in our models thus scale to 1–3 km, with roof thicknesses of 30–250 m, source-layer thickness of 0.25–0.5 km and total shortening of up to 7.5 km in our Model 2 (Fig. 5). Likewise, if we assume a natural salt viscosity of 1 × 1017 Pa s (Van Keken et al., 1993), deformation in nature lasted 3.7 Myr. This duration yields a stock rise rate of 0.5 mm year−1 (0.5 m Myr−1) and a regional shortening rate of 2.0 mm year−1 for models driven by a moving endwall. The equivalent natural strain rate for the shortened part of the model is 3.6 × 10−15 s−1, which falls within the range of orogenic strain rates, from 10−13 s−1 to 10−15 s−1 (Pfiffner & Ramsay, 1982). Model shortening rates are also realistic for fold-and-thrust belts driven purely by gravity. For example, shortening rates for the northern Gulf of Mexico are 0.1–0.5 mm year−1 (Rowan et al., 1999, 2004). Translation rates from offshore Congo, West Africa, range from 0.4 to 1.0 mm year−1 (Rouby et al., 2003), whereas those from the Kwanza Basin have much higher translation rates, ranging from 2.3 to 2.7 mm year−1 (Jackson & Hudec, 2005). These values are within the same order of magnitude as those calculated from the models. For a natural viscosity of 1 × 1018 Pa s, natural rates of shortening and diapir rise would scale 10 times slower.

Model description

This paper summarizes the results of seven models (Fig. 5). To avoid tedious repetition in fully describing each model in turn, results are grouped into the following themes: (1) roof arching and breakup above an active diapir; (2) transport and fate of rafts at the margins of the extruding salt; (3) source of extrusive salt; and (4) thrusting and welding. A column headed ‘Model design’ in Fig. 5 describes the rationale for each experiment. Where they are known, natural analogues are also grouped into these themes rather than presenting the examples out of context at the end of the paper. This format helps to highlight the relevance of each feature of the experimental models.

Roof Arching and Breakup above an Active Diapir

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

Initial doming

Before the roof of a rising salt diapir breaks up, a topographical dome forms. Particularly good examples are in the Zagros fold belt. Here salt diapirs are pressurized by their thick overburdens and by orogenic collision of the Arabian and Iranian microplates. The Zagros fold belt is shortening at a rate of 20–25 mm year−1 (Hessami et al., 2001; Aftabi et al., 2010). As a result, pressurized salt is forced to the surface. Where salt extrudes faster than it dissolves, emergent salt swells to form a bun-shaped topographical dome. Under static conditions, such domes could not exist. However, the dynamic pressure exerted by rising salt builds these summit domes up to about 600 m in relief in the Zagros examples (Talbot, 1998; Talbot & Aftabi, 2004; Aftabi et al., 2010). At the same time, gravity causes these domes to sag and spread, so that the dynamic bulge is surrounded by an apron of extruding allochthonous salt. This combination of summit dome and extrusive apron defines a salt fountain (Bailey, 1931; Talbot, 1998; Aftabi et al., 2010). Eventually, the supply of diapiric salt wanes as the source layer depletes, the diapir stem pinches off or shortening ends. As a result, the summit dome sags and merges with its apron of extrusive salt and assumes the profile of a viscous droplet (Talbot, 1998; Aftabi et al., 2010).

The fluid mechanics inferred for these emergent diapirs is illustrated by a simple physical model (Figs 5 and 6, Model 1). Salt was pumped up a vertical, upward-flaring stock then emerged and spread under gravity. The surrounding overburden was horizontal, so the salt spread radially (Fig. 6a). Maximum radial extension was in the centre of the diapir and decreased outwards. Radial shortening occurred within the periphery of the spreading salt and adjoining sediments (Fig. 6a). The topographical profile was that of a viscous fountain (Fig. 6b). Turning off the diapiric salt supply and leaving the model to rest for 14 h dissipated the dynamic pressures. Hence, the dynamic bulge collapsed, and continued radial spreading flattened and widened the salt sheet to form a viscous droplet (Fig. 6c).

image

Figure 6. (a) Digital image-correlation analysis and (b and c) laser scan-generated overhead and profile views of Model 1, where a vertical feeder actively pumped salt up into flaring salt diapir. The roof was mechanically insignificant and consisted of a dusting of ceramic beads to reflect laser light. (a) Above active feeder, diapir inflated and spread radially, advancing across peripheral plain. Fastest spreading was above centre of diapir. (b) While salt was still being pumped up feeder, a dynamic bulge formed a viscous fountain. (c) Once salt supply switched off, this viscous fountain gradually collapsed outwards under its own weight to form viscous droplet.

Download figure to PowerPoint

Talbot (1998) profiled five Iranian examples of salt fountains. Two other notable examples of viscous salt fountains are noteworthy (Fig. 7). Hoja Mumin diapir near Kulob city in the intermontane Afghan-Tajik basin (Tajikistan) is a spectacular mountain of Jurassic salt towering over 800 m above the surrounding landscape (Fig. 7a–c). The prominent dynamic bulge of Hoja Mumin is surrounded by a salt apron that has flowed subradially away from the east, where spreading is hampered by an upturned roof flap. According to Leith & Simpson (1986), the extrusion rate of Hoja Mumin is comparable to that of Kuh-e-Namak (Bushehr), which is about 1 km per 6000 year (Talbot & Jarvis, 1984). Hoja Mumim and other exposed diapirs in the Kulob area are responding to Himalayan shortening (Leith et al., 1981; Bourgeois et al., 1997; Jordan et al., 2009). In the Zagros foreland, Syahoo diapir is another example of a viscous salt fountain exposed at the southeast end of Finu anticline (Fig. 7d; Aftabi et al., 2010). Flow of the salt glacier is impeded slightly by a cuesta on the limb of Finu anticline (Fig. 7d; Aftabi et al., 2010). An example of a viscous droplet is Moghuieh diapir from the Zagros foreland in Iran (Fig. 7e; compare to Fig. 6c).

image

Figure 7. Natural examples of salt fountains and droplets. (a) Overhead view, (b) oblique view and (c) E–W topographical profile of Hoja Mumim salt diapir near Kulob, Tajikistan (37°43′ N, 69°38′ E). Dynamic bulge of this diapir of Jurassic salt rises some 900 m above surrounding plains. Viscous fountain profile indicates rapidly rising salt. More-vegetated area on east side of dome is likely to be a tilted roof flap that blocks eastward spreading of the salt sheet. Imagery from Google Earth. (d) Oblique view of Syahoo salt diapir and salt glacier in the Zagros foreland of southern Iran (27°49′ N, 56°15′ E). Hormoz salt flowing outwards from the summit dome is diverted around a buttress of upturned country rock along frontal limb of Finu pericline (Aftabi et al., 2010). Although not as active as Hoja Mumin, the diapir profile indicates that salt is still rising (compare with Fig. 6). Imagery from Google Earth. (e) View looking west at Moghuieh diapir, Zagros foreland, Iran (26°37′ N, 54°26′ E). Profile is that of viscous droplet, after the dynamic bulge collapsed into the surrounding salt apron owing to cessation of salt supply from depth. The smooth profile of droplet is now degrading by erosion and dissolution.

Download figure to PowerPoint

Mild extension of roof

Model 1 (Fig. 6) had a mechanically insignificant roof, consisting of merely a dusting of ceramic microspheres in order to reflect laser light. All our other models had thin, but mechanically significant, roofs. These models illustrate the fate of diapir roofs disrupted by outward collapse of the underlying dynamic bulge and breakout of diapiric salt (Fig. 5). As the thin roof of a stock began to respond to the underlying dynamic bulge of a squeezed stock, the roof initially extended mildly. Initial inflation, breakup and early dispersion of the roof of an inclined diapir (Model 2, Fig. 5) are shown by oblique views (Fig. 8); digital elevation models created by laser scanning (Fig. 9) show evolving structural topography; vector maps (Fig. 10) show the directions and velocities of salt and raft-block movements as the diapir broke out.

image

Figure 8. Oblique views of Model 2 showing evolutionary stages of roof breakup and salt breakout: (a) rising salt arched roof, and radial grabens formed in response to widening of diapir crest; (b) salt lobes broke out where radial grabens intersected the periphery; (c) rafts (R1–R5) separated along grabens and began to move off diapir or became grounded on upturned collar (R1); (d) rafts dispersed to periphery of sheet; some (R5) were overrun from behind by the spreading salt sheet; (e) grounded raft R1 split the salt sheet into two lobes that encircled this roof fragment and eventually would meet along an autosuture. Shear traction by diverging flow began to split this raft along V-shaped graben.

Download figure to PowerPoint

image

Figure 9. Digital elevation models of Model 2 generated by a laser scanner. (a) Pressurized salt pumped up stock and lifted roof. Fold scarp rimmed stock as salt flow was diverted towards periphery by roof. (b) Continued inflation of the diapir raised entire roof above surrounding plain. As this inflated diapir spread outwards, roof began to fragment. Rafts separated and moved radially outwards above salt sheet, overflowing upturned collar of diapir through tears in peripheral fold scarp. (c) Raft R1 grounded above upturned collar, forming a hill around which lobes of salt flowed. Upturned collar visible beneath transparent salt in the south. (d) R1 raft remained stranded on upturned collar, deflecting flow of extrusive salt around it.

Download figure to PowerPoint

image

Figure 10. Vector-field views of Model 2 derived from digital image correlation, illustrating surface displacements during initial roof breakup. Vectors decrease towards the foreland because of lateral compaction and also decrease along strike away from yielding diapir.

Download figure to PowerPoint

In the first stage of Model 2, far-field shortening pressurized the diapir and displaced salt upwards, doming the diapir's crest (Fig. 9a). A weak dynamic bulge rose above the centre of the stock, where lateral boundary drag was least, and bowed up the overlying roof. Rising salt was diverted to the periphery by the roof, much like a fountain is diverted by a solid object above it (Fig. 9a). This created a peripheral bulge surrounded by a fold scarp formed by a local upturned collar of roof sediments (Figs 8a and 9a). Once the raised roof was high enough to surmount the upturned collar, it began to collapse outwards. The salt began to spread radially, stretching its roof (Fig. 10a and b). Outer-arc extension of the peripheral bulge formed a peripheral graben (Figs 8a and 9a, b). The diapir roof also began to extend concentrically, forming radial grabens striking almost orthogonal to the diapir margin (Figs 8a, b and 9b).

An axisymmetric model stock having a thicker and more-cohesive roof is illustrated in Fig. 11 (Model 3, Fig. 5). This experiment investigated the detailed roof deformation above an active diapir. A vertical diapiric stem flared upwards into a buried diapir (Fig. 4d). As silicone was pumped upwards, the diapir inflated a central dynamic bulge. The overlying roof began to spread outward as it initially broke into three discrete rafts, forming a Y-shaped extensional system (Fig. 11a and b). With further extension and spreading, five discrete rafts were separated by radially striking grabens (Fig. 11a and b). The extruding diapir spread by radial extension in its centre and by radial shortening on its decelerating margin (although, in nature, dissolution could trim this salt rim). Throughout the diapir's crest, concentric extension took place as the salt body's circumference expanded. At the margin of the diapir's roof and in an aureole of country rock, there was minor radial shortening. This early stage of radial shortening intensified to form a discrete ring thrust (Fig. 11b and d). As in Model 2, lobes formed where radial grabens intersected the peripheral graben of the diapir (Figs 8a–c and 11b, c).

image

Figure 11. (a) Strain and vector-field map of Model 3 during initial active salt rise. Maximum extensional strain (positive) was above the centre of the diapir. Belt of concentric shortening (negative) formed around rising diapir and evolved to a ring thrust (see d). (b) Overhead view of Model 3 illustrating radial and peripheral grabens. Periphery of diapir roof expanded by means of ring thrust, by this stage covered by landslides. (c) Depth slice through reconstructed 3D volume of Model 3 at end of experiment, showing breakout salt lobes between translated roof rafts. (d and e) In section X-X′, raft R2 slid outward along the ring thrust and then on a basal thrust, lubricated by a thrust weld. Slump blocks continually shed from leading edge of raft blocks. In Section Y-Y′ raft R4 was partially overrun by neighbouring salt lobes.

Download figure to PowerPoint

Examples of spreading salt sheets having mechanically significant roofs in the northern Gulf of Mexico are shown in Fig. 12a and b. In the first example, radial grabens are forming at right angles to the margins of the lobe, as in Models 2 and 3 (Fig. 12a; cf. Figs 8a, 9b and 11a, b). More-advanced roof breakup is shown in Fig. 12b, where radial grabens again strike orthogonally to the margin. However, the overall graben pattern is complicated because the planform of the diapir is irregular and because a peripheral graben on the north rim of the diapir is intensified by southward-skewed flow down the continental slope (Fig. 12b).

image

Figure 12. Roof breakup above spreading salt diapirs in the northern Gulf of Mexico. (a) Radial grabens in early stages of roof breakup above inflating salt lobes near the Green Knoll area at the leading edge of the Sigsbee salt canopy. (b) Enhanced roof breakup above a rising diapir near the Garden Banks/Green Canyon border. Major radial grabens have thinned the roof as this lobate diapir spreads preferentially southward down the regional slope. Peripheral graben and tilted flap formed along northern edge of the diapir. (c) Advanced roof breakup of multidirectional spreading diapir near Green Canyon/Walker Ridge border. Roof is thinned and dispersed by shear traction of underlying salt flow. Rafts of roof strata undock from thicker, more intact roof and flow to the north and west. Bathymetry from GeoMapApp.

Download figure to PowerPoint

Intense extension of roof

Continued extension of the radial and peripheral grabens broke the diapir roof into discrete rafts (carapace blocks, in this case). Shear traction by salt spreading outward from the active diapir carried these rafts down onto the peripheral plain (Figs 8a, b and 9b, c). Flow from any point source tends to be axisymmetric and radially outward. Therefore, models run without shortening or a regional dip flowed purely radially (Models 1 and 3, Fig. 5; Figs 6 and 11), as in Zagros examples that extrude onto a mostly flat plain (such as Syahoo diapir, Fig. 7d). However, the following boundary effects can skew a flow asymmetrically: (1) topographical slope, (2) a stock inclined towards the foreland and (3) flow in the source layer driven from the hinterland by an advancing thrust belt (cf. Dooley et al., 2009). For example, in Model 2, flow was skewed slightly towards the foreland or seaward side of the salt stock by the seaward lean of the diapir, even without a regional dip (Figs 10a, b and 13a). On a dipping surface, salt extrusions are highly asymmetric, even from the early stages of breakout, dispersing roof rafts primarily towards the downdip side of the salt glaciers (Fig. 13b and c), as in Gach and Burkh diapirs of the Zagros Foreland (Fig. 13d). Diapir inclination also affects the rate of salt extrusion from a squeezed stock. Models 4 and 5 both had a 5° regional dip, but the salt sheet was much larger from the inclined diapir (Model 4, Fig. 5; Fig. 13b) than from the vertical diapir (Model 5, Fig. 5; Fig. 13c).

image

Figure 13. Overhead views (a–c) of Models 2, 4 and 5 after 4.5 cm of endwall displacement; displacement rate was uniform in each model. All models shortened from right to left by moving endwall. (a) In Model 2, during initial breakup, spreading was mostly radial above inclined stock without regional dip. (b) In Model 4, spreading and breakup of inclined stock was asymmetric as large, extrusive sheet preferentially flowed down the regional 5° dip. (c) With a vertical stock and regional dip of 5° in Model 5, spreading and breakup also asymmetric, but salt extrusion volume was far less. (d) Oblique view, looking east at Gach diapir (foreground) and Burkh diapir in Zagros fold belt, Iran. Each diapir flowed as single tongue down the limbs of adjacent anticlines. Combined length of Gach diapir and its salt glacier ca. 8 km. Images from Google Earth. (e) Model 7 had same setup as Model 5 except for thicker and denser roof, hindering extrusion as it gradually foundered in throat of diapir, as shown in cross section (f). Shortening of this diapir was accommodated by limited surface extrusion and deep expulsion of diapiric salt (yellow markers) down into source layer (cf. Dooley et al., 2009).

Download figure to PowerPoint

As roof rafts were dispersed, they had to surmount the upturned collar of overburden around the original diapir (Figs 8b–e, 9b–d and 10c, d). The barrier created by this collar partly pinned these rafts, which formed an incomplete ring of topographical highs rimming the diapir (Fig. 9c and d). Partial grounding of the rafts caused some of them to rotate in the current of flowing salt. For example, in Model 2, raft R1 rotated anticlockwise before being torn asunder by shear traction exerted by adjacent salt lobes (Figs 8c, d and 9c, d). These lobes converged on the peripheral plain in front of the impeding rafts (Figs 8d, e and 9c, d). Raft 4 tore into three smaller rafts as its downstream part pulled away from its upstream part blocked by the collar (Rafts 4a-c, of Fig. 8c). Raft 2 partly grounded on the subsalt collar and rotated clockwise before being partly dismembered (Fig. 8c–e). Daughter rafts undocked from the parent raft and moved rapidly down onto the peripheral plain. A depth slice through Model 3 after advanced spreading intersects the upturned collar around the diapir and six breakout lobes separating intervening rafts (Fig. 11c).

Figure 12c illustrates a diapir in the northern Gulf of Mexico having a highly extended roof. Rafts undocked from the intact roof along an escarpment (equivalent to a glacial calving line). From here, rafts were extensionally shredded and carried to the rim of the shallowly buried salt sheet. Another example of multidirectional roof breakup is the extrusive Garmsar salt nappe in the Alborz foothills of northern Iran (Fig. 14a). Here, Miocene roof rafts have undocked and rotated in the spreading salt sheet. An overhead view from Model 6 shows the same pattern of undocking, variable rotation, and extensional shredding (Fig. 14b).

image

Figure 14. (a) Landsat TM image of the extrusive Garmsar salt nappe, Alborz Range, northern Iran. (b) Overhead view of Model 6. Small, slower roof rafts at periphery cause salt sheet to form lobate front. In both Garmsar salt nappe and Model 6 roof, rafts undocked, fragmented, rotated and were transported by shear traction of underlying salt. Compare with natural example in Fig. 12c. (c) Oblique satellite image of part of the salt nappe, showing recumbent isoclinal fold within nappe. (d) Recumbent sheath folding in quarried face near front of Garmsar evaporitic nappe. In this panorama, two markers define the sheath fold exposed by quarrying. Inferred tectonic transport direction of the nappe is away from viewer. (e) Cross section through mildly squeezed diapir in Model 6. Section illustrates steep fabrics in diapir neck, subhorizontal fabrics and sheath folds in salt sheet, and transported roof raft close to the sheet periphery.

Download figure to PowerPoint

Transport and Fate of Rafts at Extrusion Margin

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

As roof rafts approach the margin of an expanding salt sheet, radial extension is increasingly replaced by radial shortening (Figs 8c–e, 9d, and 11b, c). Extrusion slowed in the thinner periphery of the salt sheet (vector maps in Fig. 10b–d) because rafted pieces partly grounded and because the flow as a whole slowed as it widened. Around the stranded rafts, lobes of salt escaped and flowed onto the peripheral plain (Figs 8b and 9b), especially where the original radial grabens intersected the diapir's margin (Fig. 8a, b and 9b). In this decelerating periphery, roof pieces collided and were assembled into a mosaic of raft fragments (Figs 8c–e and 9d). Rafts that were relatively small, thin and flexible overturned at the sheet front. In contrast, big, coherent raft blocks were thrust outwards above thin salt by the expanding salt sheet (Figs 8d, e and 11d). In cross section X-X′ through Model 3 (Fig. 11c and d), raft R2 moved to the seaward edge of the advancing salt sheet above a smear of salt and a carpet of debris shed from the raft. Slower peripheral rafts were eventually overrun or encircled by rapidly spreading salt lobes (R5 in Fig. 8d and e; R4 in Fig. 11b, c and e). Some advancing lobes rejoined to form encircling autosutures, such as around R4b in Model 3 (Fig. 11b and c; for suture terminology, see Dooley et al., 2012).

Spectacular examples of rafts stranded at the periphery of a salt sheet as a kind of terminal moraine are preserved in Qom Kuh salt glacier in north-central Iran (Gansser, 1960; Jackson et al., 1990; Talbot & Aftabi, 2004; Fig. 15a and c). These were not originally part of a diapir roof. Instead, they consist of hornblende porphyry that was originally interbedded within the salt as stratiform extrusions or intrusions that were carried to the surface. The even distribution of these rafts attests to a uniform flow pattern, even though this particular extrusion is asymmetric. The rafts survive because they are insoluble, whereas the salt glacier dissolves in a raggedly thinned margin around them. Model 6 shows a similar dispersal of rafts to the margin of a salt sheet (Fig. 5; Figs 14e and 15b). Examples of rafts near the periphery of a salt sheet in the northern Gulf of Mexico are shown in Fig. 16. Figure 16a illustrates a Miocene-aged roof raft buried by continuous sedimentation before it could reach the leading edge of the salt canopy. A section of Model 6 illustrates a similar stranded roof raft that did not reach the periphery of the salt sheet (Fig. 14e). The raft in the second seismic example reached the periphery of the salt sheet before it was pinned by overlying sediments (Fig. 16b). Similarly in Model 3, the roof raft was stranded at the periphery of the salt sheet and advanced along a basal thrust lubricated by a thin smear of salt (Fig. 11d).

image

Figure 15. (a) Geologic map of Qom Kuh diapir, Qom Basin, central Iran. Modified from Talbot & Aftabi (2004). Diapir consists of lower Oligocene and lower Miocene salts with interbedded igneous sills and flows. Igneous rocks dismembered radial flow to form line of rafts that accumulated as terminal moraine in the south. (b) Oblique view of Model 6 showing roof rafts dispersed to periphery of the salt extrusion. (c) Oblique view of Qom Kuh showing dynamic bulge at summit of salt fountain and ring of igneous inclusions stranded at its dissolving terminus. (d) Ground view of 30-m-long igneous inclusion within the extrusive salt sheet. Location shown in c. (e) Topographical profile across Qom Kuh. Asymmetry of the Qom Kuh extrusion is attributed to tectonic control.

Download figure to PowerPoint

image

Figure 16. Line drawings of raft blocks from seismic data in the Mad Dog area, Green Canyon protraction block, northern Gulf of Mexico. (a) Raft block of Miocene age buried by sediments before reaching leading edge of this composite salt sheet. Compare with Fig. 14e. (b) Raft block that reached the leading edge of the salt sheet where it was pinned by continued sedimentation. Compare with stranded raft in Fig. 11d

Download figure to PowerPoint

The hornblende porphyries at Qom Kuh were carried up the diapir to the surface, even though they are denser than salt; thus salt rose faster than the porphyry blocks sank (as modelled by Weinberg, 1993; Koyi, 2001). For sluggish diapirs, dense roof rafts or intrasalt blocks may sink back down the throat of the diapir and obstruct further salt rise. Model 7 consisted of a vertical stock below an 8-mm-thick roof of sediments significantly denser than the model salt (Fig. 5). As it shortened, the diapir's roof was breached at the peripheral graben, and small sheets began to extrude from the diapir (Fig. 13e). Once detached from surrounding country rocks by this peripheral extension, the dense roof sank down the vertical feeder, partly clogging the vent and preventing it from pinching off during shortening (Fig. 13f). With extrusion hindered by the foundering roof, a deep outward plume of salt was ejected from the diapir into the source layer (Fig. 13f). Compressional uplift of the overburden in the foreland created space for this outward plume of salt (Fig. 13f; see Dooley et al., 2009, for further details).

Source of Extrusive Salt

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

To our knowledge, no published attempt has been made to infer the relative contribution of salt from a source layer and from a diapir in supplying an allochthonous salt sheet. Different colours of passive markers in our models (see 'Model Methodology'), allow us to assess these relative contributions. A key to the colours of passive markers is shown in Fig. 4b, c and in Fig. 17. The undeformed stock in Model 2 contained three horizontal marker layers: orange at the top, green in the middle and blue at the base of the diapir. Likewise, the undeformed source layer contained eight cylindrical marker plugs, numbered and positioned as follows: 1–5 in the hinterland, 6–7 adjacent to the stock and 8 in the foreland.

image

Figure 17. Overhead views of Model 2 illustrating extrusion of progressively older salt. (a) Extrusion of salt originating in diapir. (b) Source-layer salt (within dashed white line) reached the surface and flowed out over salt from stock as diapir was engulfed by advancing thrust belt. Another four marker plugs from the source layer also emerged at surface. (c) Salt originating in source layer continued to reach surface, as evidenced by three marker plugs. (d) View of base of the model showing original location and color of source-layer marker plugs and blue marker layer at base of diapir.

Download figure to PowerPoint

Figure 17 shows snapshots of the sequence of salt extrusion in Model 2. Older salt layers overrode younger salt layers. In the first snapshot (Fig. 17a), the top-diapir (orange) marker layer, which was the first to extrude, had been largely overrun by the originally deeper, mid-diapir marker (green). The orange top-diapir marker was still visible at the periphery of the scalloped extrusion (Fig. 17a). The extrusion consisted solely of salt expelled from the diapir itself. As shortening continued, salt from the source-layer reached the surface and spread over the diapiric salt (Fig. 17b). That the salt originated from the source layer is evident from its stratigraphic position below the blue layer marking the base of the diapir and because it contained marker plug 3, which was embedded in the source layer (Fig. 17d). After this marker plug emerged from the diapir, all other plugs in the source layer also surfaced (Figs 17b, c and 19b). The area of source-layer salt expanded as it overrode the early-expelled diapiric salt. In Model 2, the allochthonous sheet had a volume of 7089 cm3. The initial volume of the inclined diapir was 2597 cm3, so the source layer contributed 64% of the salt in the allochthonous sheet.

As in the models, the internal layers of natural diapirs tend to be expelled from a diapir in reverse stratigraphic order. For example at Kuh-e-Jahani diapir (Zagros fold belt), Cambrian salt extruded first and was overridden by Neoproterozoic salt (Talbot et al., 2000). Likewise at Qom Kuh diapir (central Iran), lower Oligocene salt extruded over lower Miocene salt (Fig. 15a; Talbot & Aftabi, 2004). As physically modelled by the latter authors, older salt layers are pumped out above younger layers in the first-order recumbent fold created by a spreading extrusion. The upper limb of this large, recumbent fold can be thinned by intense shear because it spreads faster than the rest of the extrusion. Therefore, any overhead view is dominated by the less-sheared but overturned lower limb. Kilometre-scale recumbent folds within salt sheets are spectacularly exposed within the Garmsar salt nappe in the Great Kavir of Iran (Fig. 14c); smaller sheath folds are exposed in quarries within the extrusion (Fig. 14d). A cross section of Model 6 shows how subvertical fabrics in the neck of the diapir bend over to form subhorizontal fabrics in the extrusive salt sheet, conforming to the bedrock and to the upper free surface (Fig. 14e). A similar curvature has been noted at Jabal al Milh diapir in northern Yemen (Davison et al. (1996) and in Shur Ab diapir in central Iran (Talbot & Aftabi, 2004).

Models provide clues to what happens in the source layer of extruding salt. Pathways for the salt in the source layer are readily visible in Fig. 18a, which shows the underside of Model 2 at the end of the experiment. The ribbon-like streaks are deformed marker plugs that were originally embedded in the colourless silicone in the source layer. The original locations of the undeformed marker plugs are shown in Fig. 18b. Each ribbon originated from a numbered spot, and became deformed as a loop anchored at the top and base of source-layer salt. Examples of less-deformed marker plugs from a different experiment illustrate the loop of a deformed marker (Fig. 18c). The noses of these loops were carried into the base of the diapir by salt expelled from the source layer by displacement loading. The hinterland markers (1–5) flowed farthest, but even the foreland marker (8) flowed towards the advancing thrust front and up into the diapir, eventually joining the allochthonous salt sheet at the surface (Figs 18a and 19b). This flow involved intense strain, considering that each deformed loop is a doubled length: for example, marker plug 3 in the hinterland had cumulative stretches of 12 (a 12-fold elongation) to reach the base of the diapir, 18 to reach the surface (Fig. 17c) and 38 to reach the position in the extrusion are shown in the final overhead image in Fig. 19b. These rising plugs of coloured polymer are spectacularly illustrated in the laser-scan image in Fig. 18d.

image

Figure 18. (a) Underside view of Model 2 at end of deformation. Initial locations of the marker plugs are illustrated in (b). Images of less-strained marker plugs are shown in (c) to clarify view in (a). All source-layer markers, even from the seaward side, converged and flowed up diapir. (d) Overhead laser-scan image illustrating rising and spreading plugs of coloured polymer from the source layer.

Download figure to PowerPoint

image

Figure 19. (a) Isometric view of Model 2 illustrating along-strike variability in structural style. In centre of model the highly squeezed diapir absorbed shortening by expelling salt. Along strike, squeezed diapir passed into regional thrusts and backthrusts. (b) Final overhead view of Model 2 showing full extent of the allochthonous salt sheet and locations of sections in (a). (c) Cross section from welded flank of squeezed diapir in Model 2. Secondary thrust weld overlies remnant diapiric pedestal. See Fig. 17d for section locations.

Download figure to PowerPoint

Diapiric salt is not necessarily forced upward during regional shortening. Some can be forced down. A squeezed diapir can eject a plume of diapiric salt back into the source layer under either of two conditions. First, if surface extrusion is inhibited by a strong, thick roof (Dooley et al., 2009). Second, if surface extrusion is inhibited by a foundering roof that settles and blocks the neck of the diapir (as in our Model 7, Fig. 13e and f).

Thrusting and Welding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

A squeezed diapir interacts with a surrounding thrust belt. As noted above, large strains are needed to transport salt from source layer to extrusion. In addition, enormous strains affect salt as a diapiric stock almost pinches shut. Both the squeezed stock and its surrounding country rock shorten in different ways while maintaining kinematic compatibility. Figure 19a shows how a stock—or its welded equivalent—having vertical or steeply inclined sides of a curved strike must merge on its flanks with regional thrusts that dip gently and have a straight strike. A section through the lateral margins of the closing diapir illustrates a steep thrust weld above the remnant pedestal (Fig. 19c). This weld abruptly passes laterally into the more shallowly dipping, main forethrust seen in Section A of Fig. 19a. Assuming that the country rock is layered but otherwise homogeneous, thrusts or folds tend to form transversely, perpendicular to the maximum horizontal compression. This is clearly illustrated by a depth slice through the reconstructed 3D volume of Model 2 (Fig. 20a). In contrast, the planform of a stock shrinks as a series of narrowing ductile ellipses before pinching off as a transverse salt weld, best seen in depth slices of Models 2 and 5 (Fig. 20a and c; see Dooley et al., 2009, for other variably shortened salt stocks).

image

Figure 20. (a) Depth slice through reconstructed volume of Model 2. Main thrusts strike at high angles to regional shortening direction, which was to the west. Primary indenter at this structural level was bounded by oblique-slip tear faults, which curved increasingly parallel to the shortening direction nearer diapir. Squeezed salt stock formed salient in thrust front. Level of depth slice shown in Fig. 19. Section lines A-C are illustrated in Fig. 19. (b) Strike section Z-Z′ of Model 2 illustrating subvertical fault strands bounding primary indenter, main thrust below secondary indenters and deep primary welds. The primary indenter formed a structural low in hanging wall of thrust block. Location of section is shown in (a). (c) Virtual depth slice of Model 5 illustrating advanced closure of originally cylindrical salt stock to a highly elliptical planform. The stock welded at its lateral margins.

Download figure to PowerPoint

For kinematic compatibility, the country rock breaks into blocks that indent a shortened stock and fill space created by expulsion of its salt. The key piece is a primary indenter (Dooley et al., 2009). This indenting fault block tapers towards the stock, ending in a blunt projection that fits into and displaces diapiric salt (Fig. 21a and b). In this model (Model 4, Fig. 5), the indenter moved westward, obliquely underthrusting transverse thrust blocks. Because the bounding faults of the primary indenter strike obliquely to the transport direction, they are transpressional. Natural examples of primary indenters were documented in Dooley et al. (2009) (their Fig. 15).

image

Figure 21. (a) Overhead view of Model 4 showing major asymmetric salt extrusion to west and minor backflow to east. (b) Map of vector field and westward displacement in part of area in (a). Primary indenter is fault-bounded block that tapers towards back of diapir. This indenter has much greater westward displacements than deformed strata along strike to north and south. (c) Colour fill shows N-S displacements: secondary indenters (red arrows) converged towards back of diapir, with one moving SE and the other NW.

Download figure to PowerPoint

In addition to the primary indenter, our models showed secondary indenters, some barely perceptible and some prominent. Like closing pincers, secondary indenters converged obliquely into the sides of a squeezed stock. The north–south-converging component of oblique slip intensified nearer the diapir (Fig. 21c). It also strengthened over time and typically peaked as the diapir finally closed. We interpret secondary indenters as escape structures that follow paths of least resistance. The primary indenter is a structural low, so it takes less work to obliquely overthrust the primary indenter into a diapir than to keep building a mountain range of thrusts (Fig. 20b). A strike section of Model 2 shows that these secondary indenters are positive flower structures that sole out at shallow depths before merging downward into the steep sides of the deformed diapir (Fig. 20b). Our previous models of squeezed diapirs having thick sedimentary roofs documented similar geometries adjacent to the closed diapir (see Fig. 13 of Dooley et al., 2009).

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

A dynamic bulge, which is elevated by the dynamic pressure of salt rising inside a diapir, builds a topographical summit dome above a vigorously venting, emergent diapir. This salt fountain spreads under its own weight, creating an extrusive apron of salt. After diapiric rise ebbs, the dynamic bulge sags to form a flat-topped droplet (Fig. 6). This droplet also spreads outward, but much more slowly than a salt fountain. Flow from a point source tends to be axisymmetric and radially outward (Figs 6 and 11). However, the following boundary effects can skew a flow asymmetrically: (1) local topographical slope, (2) a stock inclined towards the foreland and (3) flow in the source layer driven from the hinterland by an advancing thrust belt (Fig. 13).

An extruding diapir extends radially in its centre and shortens radially in its decelerating but still advancing margin (unless dissolution trims this rim). The distal salt extends circumferentially as the extrusion widens. If the diapir has a roof, this circumferential extension creates radial grabens orthogonal to a stock's margin (Fig. 11). The graben pattern is complicated if the planform of a diapir is irregular or if flow is skewed down a slope (Figs 8 and 12b). Radial grabens can end outward in a ring graben, like a wheel rim. The peripheral ring graben can be surrounded by a ring thrust or by a monoclinal fold scarp rimming the raised roof platform. This fold scarp steepens to form an upturned collar as the roof inflates. Eventually this collar either overturns or collapses as an outward landslide (Fig. 11). Where present, a peripheral graben at the top of the fold scarp is thinned by extension, allowing an irregular annulus of salt to break out. Salt can escape outward from this annular vent across the upturned collar of country rock, especially where radial and peripheral grabens intersect (Figs 8 and 11).

As a diapir's roof breaks up, salt is expelled by regional contraction from different parts of the diapir and its source layer. Figure 22 summarizes how roof fragments and extruding salt interact in our models. Very thin roofs are extensionally shredded into countless fragments, which are carried off in extrusive salt escaping from the stock's crest. Slightly thicker roofs break into larger, discrete pieces, which successively undock as rafts (Figs 8 and 11). Some rafts are temporarily blocked by the upturned collar, but most are carried away by shear traction from extruding salt. Most of these rafts rotate (Fig. 14). Some founder or overturn within the spreading salt. Some remain blocky and coherent, whereas others are shredded and crumpled. During extrusion, older layers are pumped out above younger layers (Fig. 22). The internal layers of a diapir are expelled in reverse stratigraphic order, followed by salt from the source layer originating farther and farther away from the diapir (Fig. 22b and c). The volume of source-layer salt in an allochthonous salt sheet can be far greater than that contained within the original diapir, providing: (1) there is sufficient salt supply at depth; (2) the vent does not weld shut, and; (3) the driving force for extrusion does not end.

image

Figure 22. Summary diagram illustrating the gradual extrusion of successively older parts of a diapir and its source layer into an allochthonous sheet during shortening.

Download figure to PowerPoint

Extrusion and raft transport slow near the periphery of the salt sheet as it widens and spreads. Unimpeded salt sheets thin towards their margins. At these margins, the rafted pieces of roof can no longer be supported, so they settle like partly grounded boats in a falling tide. Because radial shortening is typical at the decelerating margin, rafts collide and clump into a jumbled mosaic. Rafts that are small, thin and flexible can overturn and crumple at the sheet front. In contrast, big, coherent raft blocks can be thrust outwards above thin salt. Rafts locally impede flow, but if overall flow outpaces dissolution, small lobes of salt escape around these stranded obstacles (Fig. 22a). Lobes of salt tend to escape where the original radial grabens intersected a stock's margin and disrupted roof strata. Eventually salt dissolution or waning flow strand rafts at the margin of a salt sheet like a terminal moraine (Fig. 22a and b).

In a contractional setting, a diapir begins to shorten early, forming a salient in the deformation front, but is eventually enveloped by the advancing thrust belt. To maintain kinematic compatibility, the country rock breaks into pieces that are driven into the stock, filling the space left by expelled salt and gradually closing the diapir. The key piece is a primary indenter. This fault block tapers into a blunt projection that is driven into the stock and displaces its salt. The primary indenter has transpressional margins, which obliquely underthrust regional transverse thrusts adjoining the diapir. In addition, secondary indenters can form. Like closing pincers, secondary indenters converge obliquely into the sides of a squeezed stock as escape structures, following paths of least resistance.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information

TD thanks James Donnelly, Nathan Ivicic, Josh Lambert, Kenneth Edwards and the late Robert Sanchez for logistical support in the modelling laboratories. The manuscript was edited by Lana Dieterich. We thank Chris Talbot, Frank Peel and Chris Jackson for their constructive reviews that helped us to greatly improve the article. The project was funded by the Applied Geodynamics Laboratory consortium, comprising the following companies: Anadarko, BHP Billiton, BP, CGGVeritas, Chevron, Cobalt, ConocoPhillips, Ente Nazionale Idrocarburi (Eni), Ecopetrol, ExxonMobil, Fugro, Global Geophysical, Hess, Instituto Mexicano del Petróleo (IMP), INEXS, ION Geophysical, Maersk, Marathon, Mariner, McMoRan, Murphy, Nexen, Noble, Petróleos Mexicanos (PEMEX), Petrobras, Petroleum Geo-Services (PGS), Repsol, Samson, Saudi Aramco, Shell, Statoil, TGS-NOPEC, Total, WesternGeco and Woodside. The authors received additional support from the Jackson School of Geosciences, The University of Texas at Austin. Publication authorized by the Director, Bureau of Economic Geology.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information
  • Aftabi, P., Roustaie, M., Alsop, G. & Talbot, C. (2010) InSAR mapping and modelling of an active Iranian salt extrusion. J. Geol. Soc., 167, 155. doi:10.1144/0016-76492008-165
  • Baikpour, S. & Talbot, C.J. (2012) The Garmsar salt nappe and seasonal inversions of surrounding faults imaged by SAR interferometry, Northern Iran. In: Salt Tectonics, Sediments and Prospectivity, (Ed. by G.I. Alsop, S.G. Archer, A.J. Hartley, N.T. Grant & R. Hodgkinson), pp. 563578. Geol. Soc. London Spec. Pub., 363, 563578.
  • Baikpour, S., Zulauf, G., Sebti, A., Kheirolahi, H. & Dietl, C. (2010) Analogue and geophysical modelling of the Garmsar Salt Nappe, Iran: constraints on the evolution of the Alborz Mountains. Geophys. J. Int., 182, 599612. doi:10.1111/j.1365-246X.2010.04656.x
  • Bailey, E.B. (1931) Salt plugs. Geol. Mag., 68, 335336.
  • Bourgeois, O., Cobbold, P., Rouby, D., Thomas, J. & Shein, V. (1997) Least squares restoration of Tertiary thrust sheets in map view, Tajik depression, central Asia. J. Geophys. Res., 102, 27553.
  • Bryant, W.R. & Liu, J.Y. (2000) Bathymetry of the Gulf of Mexico: College Station, Texas, Texas Sea Grant, TAMU- SG-00-606, CD-ROM.
  • Callot, J.P., Trocme, V., Letouzey, J., Albouy, E., Jahani, S. & Sherkati, S. (2012) Pre-existing salt structures and the folding of the Zagros Mountains. Geol. Soc. London Spec. Pub., 363, 545561. doi:10.1144/SP363.27
  • Cobbold, P.R. & Szatmari, P. (1991) Radial gravitational gliding on passive margins: Tectonophysics, 188, 249289. doi:10.1016/0040-1951(91)90459-6
  • Davison, I., Bosence, D., Alsop, I. & Al-Aawah, M. (1996) Deformation and sedimentation around active Miocene salt diapirs on the Tihama Plain, northwest Yemen. In: Salt Tectonics (Ed. by G.I. Alsop, D.J. Blundell & I. Davison) Geol. Soc. Spec. Pub., 100, 2339.
  • Diegel, F.A., Karlo, J.F., Schuster, D.C., Shoup, R.C. & Tauvers, P.R. (1995) Cenozoic structural evolution and tectono-stratigraphic framework of the northern Gulf coast continental margin. In: Salt Tectonics: A Global Perspective (Ed. by M.P.A. Jackson, D.G. Roberts & S. Snelson) AAPG Memoir, 65, 109151.
  • Dooley, T.P., Jackson, M.P.A. & Hudec, M.R. (2009) Inflation and deflation of deeply buried salt stocks during lateral shortening. J. Struc. Geol., 31, 582600. doi:10.1016/j.jsg.2009.03.013
  • Dooley, T.P., Hudec, M.R. & Jackson, M.P.A. (2012) The structure and evolution of sutures in allochthonous salt. AAPG Bull., 96, 10451070.
  • Gansser, A. (1960) Über Schlammvulkane und Salzdome: Naturforschende Gesellschaft. Zürich Viertejjahrsschrift, 106, 146.
  • Gaullier, V. & Vendeville, B.C. (2005) Salt tectonics driven by sediment progradation: part II—Radial spreading of sedimentary lobes prograding above salt. AAPG Bull., 89, 10811089.
  • Gilbert, E. & Merle, O. (1987) Extrusion and radial spreading beyond a closing channel. J. Struc. Geol., 9, 481490. doi:10.1016/0191-8141(87)90123-4
  • Hessami, K., Koyi, H.A. & Talbot, C.J. (2001) The significance of strike-slip faulting in the basement of the Zagros fold and thrust belt. J. Petrol. Geol., 24, 528.
  • Hudec, M.R. & Jackson, M. (2006) Advance of allochthonous salt sheets in passive margins and orogens. AAPG Bull., 90, 15351564.
  • Hudec, M.R. & Jackson, M.P.A. (2007) Terra infirma: understanding salt tectonics. Earth Sci. Rev., 82, 128. doi:10.1016/j.earscirev.2007.01.001
  • Jackson, M.P.A. & Hudec, M.R. (2005) Stratigraphic record of translation down ramps in a passive-margin detachment. J. Struc. Geol., 27, 889911.
  • Jackson, M.P.A. & Talbot, C.J. (1989) Salt canopies. GCSSEPM Foundation 10th Annual Research Conference Program and Abstracts, 7, 7278.
  • Jackson, M.P.A., Cornelius, R.R., Craig, C.H., Gansser, A., Stöcklin, J. & Talbot, C.J. (1990) Salt Diapirs of the Great Kavir, Central Iran. GSA Memoir, 177, 139.
  • Jordan, C., Bateson, L., Bow, J., Newell, A., Napier, B. & Johnson Sabibe, R. (2009) Geovisionary software for 3D visualization and petroleum exploration in southern Tajikistan. RSPSoc 8–11 September.
  • Koyi, H. (1996) Salt flow by aggrading and prograding overburdens. In: Salt Tectonics (Ed. by G.I. Alsop, D.J. Blundell & I. Davison) Geol. Soc. Spec. Pub., 100, 243258.
  • Koyi, H.A. (2001) Modeling the influence of sinking anhydrite blocks on salt diapirs targeted for hazardous waste disposal. Geology, 29, 387390.
  • Koyi, H.A., Sans, M., Teixell, A., Cotton, J. & Zeyen, H. (2004) The significance of penetrative strain in the restoration of shortened layers—insights from sand models and the Spanish Pyrenees. In: Thrust Tectonics and Hydrocarbon Systems (Ed. by K.R. McClay) AAPG Memoir, 82, 207222.
  • Leith, W. & Simpson, D. (1986) Earthquakes related to active salt doming near Kulyab, Tajikistan, Ussr. Geophys. Res. Lett., 13 (10), 10191022.
  • Leith, W., Simpson, D. & Alvarez, W. (1981) Structure and permeability - geologic controls on induced seismicity at Nurek Reservoir, Tadjikistan, Ussr. Geology, 9 (10), 440444.
  • Letouzey, J. & Sherkati, S. (2004) Salt movement, tectonic events, and structural style in the central Zagros fold and thrust belt (Iran). In: Salt-Sediment Interactions and Hydrocarbon Prospectivity: Concepts, Applications, and Case Studies for the 21st Century (Ed. by P.J. Post, D.L. Olson, K.T. Lyons, S.L. Palmes, P.F. Harrison, N.C. Rosen) 24th Annual Research Conference SEPM Foundation, 444463.
  • Letouzey, J., Colletta, B., Vially, R. & Chermette, J.C. (1995) Evolution of salt-related structures in compressional settings. In: Salt Tectonics: A Global Perspective (Ed. by M.P.A. Jackson, D.G. Roberts & S. Snelson) AAPG Memoir, 65, 4160.
  • McDonnell, A., Hudec, M. & Jackson, M. (2009) Distinguishing salt welds from shale detachments on the inner Texas shelf, western Gulf of Mexico. Basin Res., 21, 4759.
  • McDonnell, A., Jackson, M.P.A. & Hudec, M.R. (2010) Origin of transverse folds in an extensional growth-fault setting: evidence from an extensive seismic volume in the western Gulf of Mexico. Mar. Petrol. Geol., 27, 14941507. doi:10.1016/j.marpetgeo.2010.03.006
  • Mount, V.S., Rodriguez, A., Chauoche, A., Crews, S.G., Gamwell, P. & Montoya, P. (2006) Petroleum system observations and interpretation in the vicinity of the K2/K2-North, Genghis Khan, and Marco Polo fields, Green Canyon, Gulf of Mexico. GCAGS Trans., 56, 613625.
  • Peel, F.J., Travis, C.J. & Hossack, J.R. (1995) Genetic structural provinces and salt tectonics of the Cenozoic offshore U.S. Gulf of Mexico: a preliminary analysis. In: Salt Tectonics: A Global Perspective (Ed. by M.P.A. Jackson, D.G. Roberts & S. Snelson) AAPG Memoir, 65, 153175.
  • Pfiffner, O.A. & Ramsay, J.G. (1982) Constraints on geological strain rates: arguments from finite strain rates of naturally deformed rocks. J. Geophys. Res., 87, 311321.
  • Pilcher, R.S., Kilsdonk, B. & Trude, J. (2011) Primary basins and their boundaries in the deep-water northern Gulf of Mexico: origin, trap types, and petroleum system implications. AAPG Bull., 95 (2), 219240. doi:10.1306/06301010004
  • Radovich, B.J., Moon, J., Connors, C.D. & Bird, D. (2007) Insights into structure and stratigraphy of the Gulf of Mexico from 2D pre-stack depth migration imaging of mega-regional onshore to deep water, long-offset seismic data. GCAGS Trans., 57, 633637.
  • Ramberg, H. (1981) Gravity, Deformation and Earth's Crust in Theory, Experiments and Geological Application. 2nd edn. Academic Press, London, 452 pp.
  • Rouby, D., Guillocheau, N.F., Robin, N.C., Bouroullec, W.R., Raillard, S., Castelltort, S. & Nalpas, T. (2003) Rates of deformation of an extensional growth fault/raft system (offshore Congo, West African margin) from combined accommodation measurements and 3-D restoration. Basin Res., 15, 183200.
  • Rowan, M. & Inman, K. (2011) Salt-related deformation recorded by allochthonous salt rather than growth strata. GCAGS Trans., 61, 379390.
  • Rowan, M.G. & Vendeville, B.C. (2006) Foldbelts with early salt withdrawal and diapirism: physical model and examples from the northern Gulf of Mexico and the Flinders Ranges, Australia. Mar. Petrol. Geol., 23, 871891.
  • Rowan, M.G., Jackson, M.P.A. & Trudgill, B.D. (1999) Salt-related fault families and fault welds in the northern Gulf of Mexico. AAPG Bull., 83, 14541484.
  • Rowan, M.G., Peel, F.J. & Vendeville, B.C. (2004) Gravity-driven fold belts on passive margins. In: Thrust Tectonics and Hydrocarbon Systems (Ed. by K.R. McClay) AAPG Memoir, 82, 157182.
  • Schultz-Ela, D., Jackson, M.P.A. & Vendeville, B.C. (1993) Mechanics of active salt diapirism. Tectonophysics, 228, 275312.
  • Schuster, D.C. (1995) Deformation of allochthonous salt and evolution of related salt–structural systems, Eastern Louisiana Gulf Coast. In: Salt Tectonics: A Global Perspective (Ed. by M.P.A. Jackson, D.G. Roberts & S. Snelson) AAPG Memoir, 65, 177198.
  • Talbot, C.J. (1998) Extrusions of Hormuz salt in Iran. In: Lyell: the Past is the Key to the Present (Ed. by D.J. Blundell & A.C. Scott) Geol. Soc. London Spec. Pub., 143, 315334.
  • Talbot, C. & Aftabi, P. (2004) Geology and models of salt extrusion at Qum Kuh, central Iran. Journal of the Geological Society, 161, 321334.
  • Talbot, C.J. & Jarvis, R.J. (1984) Age, budget and dynamics of an active salt extrusion in Iran. J. Struc. Geol., 6, 521533.
  • Talbot, C.J. & Pohjola, V. (2009) Subaerial salt extrusions in Iran as analogues of ice sheets, streams and glaciers. Earth Sci. Rev., 97, 155183.
  • Talbot, C.J., Medvedev, S., Alavi, M., Shahrivar, H. & Heidari, E. (2000) Salt extrusion at Kuh-e-Jahani, Iran, from June 1994 to November 1997. In: Salt, Shale and Igneous Diapirs in and around Europe (Ed. by B. Vendeville, Y. Mart & J.-L. Vigneresse) Geol. Soc. London Spec. Pub., 174, 93110.
  • Van Keken, P.E., Spiers, C.J., van Den Berg, A.P. & Muyzert, E.J. (1993) The effective viscosity of rocksalt: implementation of steady state creep laws in numerical models of salt diapirism. Tectonophysics, 225, 457475.
  • Vendeville, B. (2005) Salt tectonics driven by sediment progradation: part I—Mechanics and kinematics. AAPG Bull., 89, 10711079.
  • Vendeville, B.C. & Nilsen, K.T. (1995) Episodic growth of salt diapirs driven by horizontal shortening. In: Salt, Sediment, and Hydrocarbons (Ed. by C.J. Travis, H. Harrison, M.R. Hudec, B.C. Vendeville, F.J. Peel & B.F. Perkins) SEPM–GCS 16th Annual Research Foundation Conference, 285295.
  • Weijermars, R. (1986) Polydimethylsiloxane flow defined for experiments in fluid dynamics. Appl. Phys. Lett., 48, 109111. doi:10.1063/1.97008
  • Weinberg, R.F. (1993) The upward transport of inclusions in Newtonian and power-law salt diapirs. Tectonophysics, 228, 141150.
  • Worrall, D.M. & Snelson, S. (1989) Evolution of the northern Gulf of Mexico, with emphasis on Cenozoic growth faulting and the role of salt. In: The Geology of North America: An Overview (Ed. by A.W. Bally & A.R. Palmer) GSA Decade of North American Geology, A, 97138.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Model Methodology
  5. Roof Arching and Breakup above an Active Diapir
  6. Transport and Fate of Rafts at Extrusion Margin
  7. Source of Extrusive Salt
  8. Thrusting and Welding
  9. Conclusions
  10. Acknowledgements
  11. References
  12. Supporting Information
FilenameFormatSizeDescription
bre12056-sup-0001-VideoS1.movvideo/mov26485KVideo S1. Time-lapse oblique view showing the evolution of Model 2.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.