Evaporites as a seal

In terms of seal capacity and resistance to fracturing, the most effective sedimentary seals are gas hydrates, followed by evaporites, then shales. To seal an economic accumulation of hydrocarbons, any sealing surface or lithology needs to be laterally continuous, maintain constancy of properties over large areas, be relatively ductile, and be widespread over a sedimentary basin. A fundamental requirement for any effective seal is that the minimum displacement pressure of the lithologic unit comprising the sealing surface be greater than the buoyancy pressure of the hydrocarbon column in the subjacent accumulation. Hence, the size of the pore throats and the density of entrained hydrocarbons and water are of great importance in determining whether or not a particular lithology can act as a seal. Density/buoyancy differences between oil and gas mean there are fields, such as some evaporite-cemented San Andres Formation mudstones in the Permian Basin of Texas, where dolomitic mudstones act as a seal to liquid hydrocarbons, yet flow gas.

Differing ability to form a seal is related to a rock’s inherent ductility in the subsurface (in part after Downey, 1984).

Any lithology can be a seal, but other than clathrates, evaporites are the most effective regardless of hydrocarbon type and structural setting. Unlike the low temperatures requirements for a clathrate seal, evaporite seals, with their extremely high entry pressures, ductility, very low permeabilities and large lateral extents, can maintain seal integrity over broad areas, even when exposed to a wide range of subsurface temperature and pressure conditions. A typical shale seal has a permeability ≈ 10-1 to 10-5 md, with rare values as low as 10-8 md (Figure A). Quantitative measurement of evaporite permeability is beyond the capacity of standard instruments used in the oil industry and is mostly a topic of study for engineers working with waste storage caverns. Their work shows the permeability of halite is a nanodarcy or less, that is, undamaged subsurface salt has measured permeabilities that are less than 10-21 m2 (10-6md) with some of the tighter halite permeabilities ≈10-7 to 10-9 md and typical massive anhydrites ≈10-5 md (Beauheim and Roberts, 2002).

Pore pressures in such halites typically approach lithostatic (Ehgartner et al., 1998). Measures of salt permeability or brine permeation generally are described through Darcy’s law (Bérest et al., 2011). However, some authors believe that undisturbed rock salt is both liquid-tight and gas-tight in its primary state, and that the origin of the non-zero permeability measured during laboratory tests are likely due to damage that occurred during test sampling and preparation, or to undetected tiny leaks from the testing equipment. For these authors, the permeability observed in the vicinity of a wellbore or a salt storage cavern is a secondary permeability, i.e., induced by the stress changes generated via drilling and cavern creation see Warren, 2016; Chapter 13). According to Bérest et al., (2011), these differing views may also be explained, in part, by differences in salt compositions and crystal character. Bedded salt formations can contain a significant amount of insolubles (clay, anhydrite), which might be more permeable than the salt itself (Warren, 2017).

Seals. A) Relative ranking of permeability of various lithologies (in part after Beauheim and Roberts, 2002) B) Hg-injection curves from representative lithologies in a sequence undergoing reflux dolomitisation and overdolomitisation (from Warren, 1992). C). Seal classification (A-E) based on air-mercury pressure curves for typical seal lithologies, denoted 1-9 (after Sneider et al, 1997 and Sneider, 1995)

As a general rule, even as a halite bed fractures, its inherent lack of strength and the consequent ability to flow means any microscale intercrystalline fractures quickly reanneal by a combination of flow and pressure-solution induced recrystallisation. Current consensus is that some thin impurity-rich salt beds, interlayered with carrier beds, do leak small amounts of volatiles, but much less efficiently than thicker organic-rich mudstones and shales; whereas organics encased in thicker salt beds probably cannot leak from the unit until the enclosing salt dissolves or natural hyrofracturing occurs (as in the Ara Salt of Oman). Evaporite beds and allochthons constitute some of the most robust longterm subsurface barriers to the vertical migration of hydrocarbons in a sedimentary basin both as a seal to hydrocarbons and in CO2 sequestration.

When assessing any hydrocarbon accumulation, the top, lateral and bottom seals must be identified and evaluated. Theoretically, the thickness of a seal does not contribute to sealing capacity: a clay shale (4-10 µm particle size) with an entry pressure of 600 psi is theoretically capable of sealing a 900 m oil column. In reality, it is unlikely a shale bed only a few centimetres thick is a laterally continuous, unbroken, unbreached unit, capable of maintaining stable lithic character over a sizeable area. A thin evaporite perhaps could be, but a thicker seal provides many layers of contingent sealing beds, and so gives a more significant probability of a sealing surface being continuous over an entire prospect. Evaporite seals more than 30 m thick are considered excellent, while shale seals more than 50 m thick and evaporite seals more than 10 m thick are typically considered adequate.

Seal continuity rather than measured seal capacity is likely the most crucial factor in assessing seal quality. In the laboratory, one can measure the displacement or entry pressure necessary to force a given hydrocarbon mixture through a rock sample under given conditions of temperature and pressure. However, there are many difficulties in extrapolating quantitative data from a capillary pressure test on a 4-8 inch (10-20 cm) diameter core into a reliable prediction of the seal character across the entire sealing surface. A simple domal closure of 2,950 ha (6,400 acres) provides a ratio between a typical seal sample (4-inch diameter whole core) and the area of a top seal of 1 to 3.5 billion. Average values of seal properties measured from core without a reliable geological model for the reservoir are next to useless in quantifying seal integrity. What is needed is knowledge of the likeliest weakest points in the seal across the structure of interest. It’s the difference between guarantees that you will on average survive if you swim in these shark- and stinger-infested waters, versus you are advised not to swim in this lagoon. Most of us prefer the second option; it’s the same when investing tens to hundreds of millions of dollars in developing an oil or gas field.

Evaporites as seal and a control of charge distribution. A) Hith Anhydrite isopach showing thickening from the UAE west into Saudi Arabia and some of the main fields producing from Jurassic reservoirs in this region (in part after Alsharhan and Kendall, 1994). Note the lack of Jurassic reservoirs to the east of the Hith erosional edge showing the effectiveness of the saltern evaporite in holding back substantial oil and gas columns sourced in Hanifa mudstones (see Figure 9.4). B) Distribution and timing of charge in the oil and gas fields along the highly productive eastern margin of the South Oman Salt Basin is controlled by the retreating dissolution edge of the Neoproterozoic Ara Salt and the underlying Huqf Fm. source rock, which became mature in the Cretaceous. Oil fields shaded black, stars indicate gas fields (after Terken et al., 2001).

Saltern and deepwater evaporite beds make excellent seals because the density-stratified hydrology of their depositional setting guarantees there was little lateral variation in depositional rock properties over large areas of the salt-covered basin floor. Evaporite-plugged mudflat units also make effective seals, as do platform carbonates plugged by evaporite cement precipitated from dense reflux brines. But high levels of entrained impurities, less predictable depositional topography and lateral variations in early diagenetic fluid flux rates mean that, compared to saltern and basinwide-style seals, these dirtier seal types are more subject to problems with lateral continuity, as well as showing a higher propensity for stress fracture and failure.

Deposition of saltern anhydrite atop the various Arab cycle carbonate reservoir sands of Saudi Arabia and the Emirates not only formed the highly effective Hith seal (Figure A), but aided in the preservation of underlying reservoir porosity and permeability by preventing later entry from above of meteoric waters capable of precipitating eogenetic intra-reservoir carbonate cement. Once dissolution starts to thin a previously effective seal, it begins to leak. Even the Hith Anhydrite can leak once it is less than 10 m thick, as can be seen by the migration of Jurassic-sourced oil in Cretaceous reservoirs along the edge of the Hith Anhydrite (Figure; Hawas and Takezaki, 1995). In my opinion, this leakage in the thinned Hith can be tied to increasing proportions of dolomite residues near the dissolutional feather edge of the Hith and a consequent propensity for brittle responses near this edge, with local faulting linked to collapse fracturing. There would be little loss of seal capacity in an anhydrite bed as thin as 10 metres if it remained as relatively pure nodular anhydrite.

Leakage is evident in Al Rayyan Field, Qatar, where the upper Arab anhydrite, the intra “C” anhydrite, and the intra “D” anhydrite of the Arab Formation do not form effective seals; only the Hith acts as a regional seal for the field (Brown and Loucks, 2001; Clark et al., 2003). The intra “C” and intra “D” beds are interpreted as depositionally discontinuous with relatively high dolomite contents. This reflects original shoalwater carbonate deposition in the cycle capstones, which were deposited as thinned, and perhaps pinched-out, evaporitic mudflats onlapping the carbonate shoal reservoir. Likewise, the intra “C” sabkha unit, with its higher fraction of intercalated porous dolomite intrabeds, promotes seal failure via fracturing and faulting. The discontinuous salina deposits in the intra “D” anhydrite also lack sufficient lateral continuity to form an effective seal over the structure. Brown and Loucks (op. cit.) found that the primary cause of seal failure in the thin, upper anhydrite is fault offset, whereby fault throw exceeds seal thickness. The high fraction of brittle dolomite in these anhydrite intrabeds results in more extensive seal damage zones, allowing greater potential for along-fault leakage. In contrast, where fault throw in the field does not exceed the thickness of the Hith, nor the thickness of the middle, and lower anhydrites in the Arab cycles in the field, the evaporites capping the various intra-Arab cycles remain as intact seals. Faults intersecting these beds also cause less structural (brittle fracture) damage due to the higher anhydrite fractions of the beds, so that along-fault leakage from connected intrasalt beds is minimal.

Hydrocarbon migration and trapping in the south Oman salt basin, with its subsalt Neoproterozoic Huqf source rock, is strongly controlled by the continuity of the Neoproterozoic Ara Salt. In the centre of the basin, where thick Ara salt is still present, there is an additional substantial volume of bituminous oil still trapped in intrasalt organic-rich Ara carbonate and siliceous stringers of the salt sealed Athel silicilyte (Schoenherr et al., 2007b; Ramseyer et al., 2013). In the updip peripheral syncline along the east flank, salt dissolution controls structural style and trap formation (Heward, 1990), with the retreat of the salt edge the main factor in charge timing (Figure B; Terken et al., 2001). Traps in Palaeozoic clastics in the eastern basin margin were initially formed by halokinetic structuring in the Paleozoic, and subsequently filled by salt dissolution breaches driving dissolution drape and oil emplacement.

The subsalt Neoproterozoic Hufq source rock began to generate oil in the middle of the Cretaceous, and the retreat of the salt’s dissolution edge can be mapped in the successive fillings of the various fields that define the productive eastern margin of the basin. As a result of the westward retreat of the dissolving salt edge through time, traps get progressively younger toward the basin centre (Figure B). Salt dissolution has today removed most of the salt along the east flank of the South Oman salt basin, and its timely removal is mostly responsible for the present-day hydrocarbon-filled structures. Oil accumulations mimic a negative image of the salt structures first formed during the earlier halokinetic stage as early salt withdrawal basins then became inverted into oil-bearing turtle structures. Anticlines resulted from the drape of strata over Haima and Al Khalata turtle structures and drape over deeper laterally discontinuous siliceous and carbonate stringers floating in the Ara Salt. Where salt dissolution occurred, formerly salt-encased stringers leaked and became the grounded remnants that now form chaotic solution residue horizons both in outcrop and in the subsurface (Terken et al., 2001; Peters et al., 2003).

© 2016-2024 Saltwork Consultants Pty Ltd