Burial Salts

The term burial salt is a general descriptor that encompasses a set of mineral salts created by diagenetic and metamorphic processes that can precipitate mineral salts in the vicinity of a dissolving and altering subsurface salt mass (salt beds, diapirs and allochthons). This is especially so in zones where two brines mix, which premixing had different salinities and chemical constituent proportions. In the subsurface, such burial salts tend to form pore or fracture fills, located wherever the most permeable subsurface conduits were at the time of brine mixing and precipitation.

A burial salt is not a true evaporite in that it is not formed by solar evaporation but, the fact that its ionic constituents are typically derived by the dissolution/alteration of a nearby evaporite, means most geologists consider burial salts in sedimentary basins to be evaporites. The precipitative processes driving the formation of burial salts are typically heating, cooling or mixing of basinal waters of two salinities, typically derived by the dissolution of subsurface salt masses or relict hypersaline pore brines.

The most common burial-salt forming within or near a mass of buried evaporites is sparry poikilotopic anhydrite, as typifies porosity-occluding cements in many reservoir sands in the North Sea (Sullivan et al., 1994). Halite cements can also be burial salts, and both minerals are derived by the subsurface flushing of the nearby evaporites, as in carbonate-sliver reservoirs encased in Infracambrian Ara Salt in the South Oman Salt basin (Schoenherr et al., 2009). Timing of the onset of such subsurface dissolution and reprecipitation as burial salts can be late or early (e.g. Schoenherr et al., 2009; McNeil et al., 1998; Kendall, 2000).

Late-stage sparry anhydrite (burial salt) in Leman Field, North Sea. A) Location of the Leman gas field and regional geological cross-section. B) Subsidence history of Rotliegende reservoir sandstone showing time and temperature of anhydrite precipitation postdates other diagenetic events and precedes secondary porosity formation. C) Strontium isotopic ratios and composition of various anhydrite burial salts compared with formation waters and Zechstein salts. There is a general trend of increasing isotopic ratio with increasing Sr content. D) Schematic section showing major sources of sulphur, strontium and carbon along with their migration from above and below the Rotliegende and subsequent mixing that precipitated anhydrite cement and fracture fill (modified form Sullivan et al., 1994).

Cement haloes, convection and saltout

Over time, the ongoing dissolution of the edges of a salt bed, allochthon or salt wall sets up a dense subsurface brine halo. As it builds, the halo becomes so saline and dense that it starts to drive free convection. Saline pore waters slowly sink beneath the underbelly of the dissolving salt bed or allochthon, or spread laterally atop the upper side of a salt bed or edge of a diapir. Pore water from the salt underbelly sink to depths where it reheats, flows laterally and then rises to return once more to the underside of the evaporite bed to complete the convection cell.

Buried evaporite beds in contact with adjacent aquifers become sites of phreatic salt flushing as soon as the solar-driven active phreatic brine reflux hydrology starts to dissipate. Hence, burial dissolution of salt from its edges inward begins as soon as the bed enters the zone of compactional water flow. This explains why the upper and lower contacts of many thick ancient evaporite beds are not depositional contacts. Rather, they are zones characterised by dissolution residues, burial salt cements, recrystallisation and pressure solution textures.

For the same reason, well preserved primary and syndepositional textures in bedded evaporites are rarely found at the contacts with adjacent nonevaporite lithologies. Depositional fabrics are best preserved internally within an evaporite bed, away from the edges where basinal fluids are eating into and altering the original evaporite textures. Diagenetic waters continue to dissolve or “melt” evaporite contacts until the evaporite dissolves or all effective permeability is lost in the adjacent beds.

Modelled temporal evolution of brine plumes associated with ongoing dissolution of a salt sheet in a shale matrix at times of 2, 6 and 20 My. Contours of plumes are given in terms of salinity difference: calculated as pore water salinity minus seawater salinity (3.5%). Numbers are in wt %. The leftmost plume, best developed in earlier stages (2 and 6 mya)  is mostly fed by a brine cascade focused by ongoing dissolution of the salt edge. Without halokinesis this would be a retreating bed feather-edge. Vertical exaggeration is 2x.

The modelling of plumes beneath salt beds and subhorizontal allochthonous salt sheets, such as occur in the Gulf of Mexico, shows free convection has the potential to be a significant mechanism in both salt dissolution and mass transport, even if the sediments beneath the salt have permeabilities as low as 0.01 md In such low permeability sediment it will take some 10 My for the sub-salt convection to develop to full strength. Once active, an average of 3-5 metres of salt is dissolved off the underbelly every million years, with a maximum darcy flux of 3 mm yr-1. Measured matrix permeabilities of overpressured shales immediately beneath allochthonous salt (the gumbo zone) are much higher than 0.01 md, and so the actual flow and dissolution rates are much higher.

Chloride-rich density-derived convective flow cells are an inherent part of the hydrology beneath all dissolving evaporite seals. They continue to repeatedly cycle subsalt pore water through the underlying sediment column until subsalt permeability is shut down by compaction and thermobaric cementation. It can move volatile-entraining chloride-rich brines through base metal and organic-containing shales and cupriferous redbeds beneath buried salt beds and sheets. Wherever such waters break out from beneath the salt seal into suitably prepared ground, the associated cooling and brine mixing can precipitate metal sulphides or facilitate the focused escape of liquid and gaseous hydrocarbons (Warren, 2000b). Breaches typically occur via growth faults, salt welds or via seismic pumping along thrusts and thrust welds; all these features centrifugally drain basinal pore waters from beneath a salt bed or allochthon sheet. Simultaneously, the escape can drive other diagenetic reactions such as secondary porosity formation (via brine mixing and changes in pCO2) or the precipitation of authigenic burial cement.

Where halite dissolution is occurring along the flank of a salt stock or diapir (shale sheath), the resulting dense pore fluids move into adjacent upturned permeable beds. Any cavity created by the local dissolution of the salt stem is quickly filled, either by salt flow from the stem, or, if dissolution is rapid enough, by the collapse of sedimentary beds against the stem to form “downthrown” slivers and blocks. As the dissolution of the salt stem proceeds, the resulting high salinity (high density) pore brines can spall off a sinking brine halo that is maintained by ongoing dissolution. Such high salinity pore waters tend to sink away from the stem down the more porous extensional portions of adjacent steeply dipping beds. Unlike the self-limiting rate of salt loss associated with the undersides of subhorizontal salt geometries, steeply dipping beds about a salt stem create a plumbing system with a tendency to self drain, and so facilitate further dissolution of the stem and the set up of conditions suitable for the “salt-out” of rising hydrocarbons. 

The evolution of such a suprasalt-self draining, self-convecting fluid system has been modelled by Wilson and Ruppel (2007). They conclude that thermohaline suprasalt convection associated with salt domes can drive significant fluid flow, mass and heat transport in suprasalt positions in halokinetic continental margins, especially if halite dissolution is considered part of the suprasalt modelling. Likewise, they found suprasalt faults cause significant, but highly localized, increases in flow rates, typically tied to seafloor seeps and discharge zones. Transient models that include halite dissolution show an evolution of flow during brine formation from early salt-driven convection to later geothermal convection, with characteristics of the latter controlled by an interplay between seafloor relief and evolving salt geometry. Their modelling shows suprasalt dissolution atop a salt structure creates hypersaline haloes more than a few hundred metres to a kilometre thick in time frames of 100,000 to 250,000 years after the creation of the structure. Predicted fluid flow rates are around a few millimetres per year or less for homogeneous sediments with a permeability of a millidarcy, comparable to compaction-driven flow rates.

Stream function and salinity distribution for the fully-coupled transient case, with two permeable suprasalt sand layers (yellow) included in the modelled muddy sediments. Black lines indicate convective flow direction, shading denotes normalized salinity, with white corresponding to salinity close to that of seawater and pink representing salt saturation. Thick dark green line indicates flow divide between clockwise and counterclockwise convective flow cells. (A) 100 kyr. (B) 250 kyr. (C) 500 kyr. (D) 1 Myr. (after Wilson and Ruppel, 2007)

Sediment permeabilities likely fall below 1md at depth in the Gulf of Mexico salt province basin, but such thermohaline convection can still drive pervasive mass transport across the seafloor, affecting sediment diagenesis in shallow sediments). They conclude that in more permeable settings, such flow drives water circulation of fluid cells with differing temperatures and so can influence methane hydrate stability, the positions of seafloor chemosynthetic communities, and the longevity of fluid seeps. Across a time frame of a million years, one of the most significant effects of the dissolution of the upper portion of the salt is the increasingly thicker halo of hypersaline brines in the overlying sediment mass (note there is no significant additional sedimentation that would drive ongoing salt flow via differential loading in their simple model. However, it does show that active suprasalt fluid circulation and burial salt cementation in convective cells is driven by thermohaline processes, tied to salt dissolution in the sediment mass atop the salt. It also illustrates that pore salinities increase in the lower part of the sediment column atop a dissolving salt structure.

Salt-out and burial cements

This set of inherent and related fluid flow processes can enhance hydrocarbon foci so define hydrodynamic suprasalt trap configurations and saltouts. Hydrocarbons rising through suprasalt and subsalt brine haloes are less soluble than if in fresher pore waters. Price (1976, 1981) noted that at standard temperature and pressure (STP) the solubilities of n-pentane and methyl-cyclohexane decreased from 40 ppm in freshwater to 3 ppm in water with a salinity of 300,000 ppm. This means hydrocarbons dissolved in less saline basinal waters tend to “salt out,” or exsolve, in the vicinity of brine plumes adjacent to a salt stock or along the underside of a salt bed or allochthon. “Salting out” is a direct result of the increase in the salinity of the carrier fluid as the salt body halo is approached (Figure 8.20). The accumulation of oil and gas along the underside of a salt seal, or adjacent to a salt stock, drastically slows the rate of salt dissolution. Halite is near insoluble when in direct contact with exsolved hydrocarbons rather than pore waters.

Aqueous solubility of a C4 - C10 petroleum distillation fraction as a function of NaCl concentration at a near constant temperature (≈195°C) and pressure (765 bars). Replotted from data in Table 4 of Price (1981).

Such conditions help explain why hydrocarbon legs adjacent to a salt stem have a propensity to stack one atop the other, with the hypersaline halo hydrology of the stem acting as highly efficient top and lateral seals. Of all the various evaporite traps, stacked salt-stem focused and sealed, steeply-inclined reservoir beds are among the most highly productive regions in the world in terms of barrels of oil equivalent (BOE) per unit area (Halbouty, 1979). Large-scale vertical migration of subsurface waters and crude oils, via shears and fractures against salt structures, was noted in the Louisiana Gulf Coast by Hanor and Sassen (1990). They concluded that the vertical upward flux of brine and hydrocarbons in this system followed extensional fault and fracture permeability in sediments adjacent to the salt and did not otherwise move vertically through shales sandwiched between the various sand reservoirs. In the immediate region about the salt stem, they found that higher salinities were localised in the same areas where hydrocarbon production occurred over and around the salt domes. This implies the dissolution of salt is driven by the same warm fluids rising from geopressured intervals and carrying with them buoyant hydrocarbons into shallower depths. Thus, most of the vertical fluid transport out of the deeper levels in a halokinetic basin is focused into the zone of structural disturbance adjacent to diapir stems.

Put simply, salt stems, beds and allochthons create temperature, salinity and fracture discontinuities that induce local thermohaline fluid flow. As long as sediments adjacent to a subsiding/flowing stratiform salt bed or allochthon can maintain some degree of permeability, much of the more interesting subsurface hydrology, mineralogical change and burial salt cementation occurs in and around brine haloes, centred about intervals or masses of slowly dissolving salt. Ongoing thermohaline-driven dissolution, alteration and flow near the edge of the bed or stem, and extending into adjacent beds, create chemical and pressure gradients that exert significant influence on the style and patterns of subsurface fluid flow. This flow focusing can exert a strong influence on the location and style of various hydrocarbon and metal accumulations (see Warren 2016; Chapters 10 and 14-16)