Skill target: Why reservoir characterisation is more complicated in carbonates than sandstones
Differences between sandstone and carbonate reservoirs influence the way we study them (Table 1). Sandstone porosity is mainly interparticle; therefore, it is related geometrically to depositional texture and fabric. Because permeability usually correlates rather well with interparticle porosity in sandstones, it can be related to depositional texture and fabric. Assuming that porosity and permeability are closely related, laboratory measurements made on small core plugs of terrigenous sandstones may be assumed to be representative of large rock volumes. That is, small samples are representative of large populations as the sampled populations are relatively homogeneous.
Porosity distributions in many carbonates do not reflect primary interparticle porosity; instead, they exhibit a variety of primary and secondary pore sizes, shapes, and origins, and measured porosity values do not always correspond closely with permeability. In short, carbonate pore systems are not typically homogeneous. While a 1 - inch perm - plug will provide reliable data on sandstone porosity and permeability, entire whole core segments may be required for reliable measurements on carbonates.
Relatively simple porosity classification schemes are useful for siliciclastics, but a compound scheme of genetic classification augmented by measurements of pore geometry is needed for carbonate reservoirs.
able 1. Significant differences in reservoir porosity are defined at the time of deposition and continue into the diagenetic (subsurface) realm
Porosity and permeability controls in a sandstone
The porosities of sandstone reservoirs range from 5% to 30%, but most generally lie between 10%and 20%, while most reservoir permeabilities are in the range 10-200 mD (Tables 2, 3 ). Porosity less than 5% is seldom commercial (some tight gas sands), and any porosity more than 35% is unusual.
Porosity in siliciclastic reservoirs are partially compacted reflections of primary porosity and with magnitudes controlled mainly by :
Magnitude and uniformity of original grain size: Uniformity, or sorting, is the degree of gradation of grain sizes in a sample. Sorting in a siliciclastic depends on at least four major factors: size range of material at deposition, type of depositional setting, current authigenic character, and the geometry of the depositional environment. If small particles of silt or clay are mixed with larger sand grains, the effective (intercommunicating) porosity will be considerably reduced, as shown in Figure 1. A reservoir sand with a significant clay content is referred to as dirty or shaley and such dirty sands are less reliably modelled by the unmodified application of Archies Law when calculating water saturation. The multiple porosity types present in most carbonate reservoirs make a simple application Archies Law even more tenuous.
1. Compaction and reservoir. A) Contrasting compaction processes that explain porosity loss with increasing burial depth sandstone and shale. B) Shale porosity loss with increasing depth largely due to for compactional dewatering.
As a general rule, we assume sandstone reservoir quality is tied to a combination of:
• Degree of cementation or consolidation: Highly cemented sandstones tend to have low porosities, whereas the soft, poorly-indurated rocks tend to have high porosities. Cementation in sandstone takes place both at the time of lithification and during rock alteration by circulating groundwater. The process is essentially that of filling void spaces with mineral material, which reduce porosity. Cementing materials include calcium carbonate, magnesium carbonate, iron carbonate, iron sulphides, limonite, hematite, dolomite calcium sulphate, clays, and many other materials, including any combination of these materials.
• Amount of compaction during and after deposition: Compaction tends to lose voids and squeeze fluid out, bringing the mineral particles close together, especially the finer-grained sedimentary rocks (Figure 1). Compaction in a sandstone reservoir is often the most critical mechanism driving porosity loss with increasing depth of the reservoir. Compaction in a sand is dominated by physical or mechanical breakage at shallower depths, while chemical or pressure solution becomes more prominent at greater depths. Expulsion of fluids by compaction at an increased burial temperature is the underlying mechanism for primary migration of petroleum from the shaley source rocks to reservoir rocks. Whereas compaction is a critical lithifying process in claystones, shales, and fine-grained carbonate rocks, it isn't very important in closely-packed sandstones or conglomerates. Generally, porosity is reduced in deeper, older rocks.
• Degree of packing: With increasing overburden pressure, poorly sorted angular sand grains show a progressive change from random packing to a closer packing. Some crushing and plastic deformation of the sand particles occurs. Likewise, the degree of packing (sheet rotation) in a shale framework comes to dominate at depth (Figure 1).
Table 2. Porosity evaluation levels for most reservoir rocks.
Table 3. Qualitative evaluation of permeability
2. Schematic illustrating the conceptual basis for a Vclay or a Vshale calculation, using the Total Gamma log.
For siliciclastics with sufficient bed thickness to be resolved in a conventional log suite, the gamma log is usually a reliable indication of the amount of shale present, with shaley sections indicated by elevated gamma) values (Figure 2). Generally, for core plug-determined values of porosity and permeability taken across a sandstone reservoir, there is a readily established linear relationship between core plug porosity and permeability on a semilog plot (Figure 3). For the same reason there a reasonable link to the GR log reflective of shale content. This is the basis for the generally used Vclay calculation tied back to poroperm plug values that are used to establish pay cut-offs in a sand-shale reservoir. In terms of porosity and permeability in a siliciclastic succession, the dominant controls of relative magnitude is usually a combination of grain size variations and sorting, and Archies Law is applicable without significant modification. With carbonates, the use of Archies Law is complicated by the much more complex diagenetic history preserved in the porosity distribution of a reservoir carbonate.
3. Typical well-behaved linear trend exhibited by porosity and permeability measurements in core plugs, as evidenced by most clay-free sandstone reservoirs, in this case from an offshore Permian Rotliegend Sandstone reservoir, North Sea.
Why are carbonate reservoirs more complicated than sandstone?
Greater complexity due to multiple porosity types in a carbonate reservoir makes quantification and prediction more difficult. This complexity reflects both the biological origin of carbonates and the much greater diagenetic susceptibility of carbonates compared to sandstones (Table 1). Historically, many of these differences were ignored so that petrophysical models and predictors, such as Archies Law, that work well in sandstones were applied without modification to carbonate reservoirs. This lead to problems.
In sandstone reservoirs, depositional stratigraphy is the dominant control on reservoir petrophysics and geobody geometry. In carbonates, we must always factor in the additional complexities of diagenetic history and structuration (especially fracturing) (Figure 4). Without a detailed understanding of all aspects of a carbonate reservoir, our subsurface model will be a gross oversimplification rather than the desired approximation to subsurface reality,
Back in 1995, Akbar et al. made the analogy that applying classic petrophysics to a carbonate reservoir is like attacking a Phillips screw with a regular flat-head screwdriver — progress is possible, but not without a struggle. Today, with carbonate reservoirs producing around half the world's oil and the proportion growing, we must hone our tools to understand these more challenging carbonate reservoirs better.
4. Petrophysical properties have no spatial information and in a 3-D seismic framework model must be linked with geologic modelsof depositional, diagenesis and structure before they become reliable representations of a reservoir.
5. Typical bioclast size ranges generated at the site of deposition
At the time of deposition, the essential difference between porosity styles in a sandstone and a carbonate is that almost all carbonate sediment is the result of life processes. Porosity shapes and levels in a recently deposited carbonate sediment reflect the petrophysical properties of the dominant biota contributing sediment into the depositional setting, be it microbial, plant or animal..
Rather than then tendency toward spherical grain shapes we see in a quartz sand, natural bioclast grain shapes and sizes in a carbonate are much more variable ranging from plates to spheres to more elongate and sausage-shaped forms. Grain size and sorting at deposition can range from cobble to mud across biological variations at a metre-scale mosaic of lithofacies (Figures 5, 6 and 7). Many carbonate grains are hollow or internally porous, as the grains preserve the life space of the plant or animal that initially constructed the grain. Over time, some sand-sized grains ( as in the green alga Halimeda) can disaggregate into mud-sized particles.
In contrast, porosity in a siliciclastic sediment is intergranular and at the time of deposition reflects a passive response to varying physical energy levels at the depositional site. This tends to created beds with laterally consistent petrophysical properties at the time of deposition. This contrasts with the typical metre-scale laterally variable petrophysical character on many carbonate platform deposits (e.g. Figures 6, 7). Some mechanically reworked carbonate sand deposits such as ooid shoals can possess similar lateral continuity scales to sandstones at the time of deposition. This reflects the dominance of physical process at the time of deposition of these sands, such as tidal currents or storm waves (e.g. the platform edge sand belt visible in Figure 7).
6. A "sea of atolls' on the Maldive carbonate platform, Indian Ocean. At the scale of reefal buildups, these are all bioconstructed units that define a mosaic combination of concentric reefal accumulations, with circular reef flats. Overall, the atolls are separated by somewhat deeper water lagoonal muds. Each atoll construct also has a central core of lagoonal mud. Some regions of the atoll flats have built up above sealevel to become vegetation-covered islands. By definition, a vegetation-covered island indicates an underlying freshwater lens and the propensity for syndepositional meteoric diagenesis.
7. Great Blue Hole, Belize carbonate platform; note the contrasting lateral continuity between a metre-scale reefal facies mosaic in the foreground. This mosaic is made up of a combination of coralgal boundstones, grainstones, packstones and wackestones, plus an overprint by karstification that formed a blue hole during the penultimate sealevel low. This contrasts with the km-scale lateral continuity in the reef-flat grainstones and rudstones in the midground (light-blue sand belt landward of the main barrier reef crest).
Early and ongoing diagenesis is pervasive in most carbonate reservoirs
Aside from contrasting petrophysical character at the time of deposition, carbonate reservoirs, unlike most sandstones, tend to possess porosities that have been pervasively altered by the ongoing processes of diagenesis. Diagenesis is much less critical in sandstone petrophysics where reservoir quality tends to be a partially reduced set of poroperm properties mostly reflective of the original depositional sand geometry (although with lesser magnitude mainly due to compaction).
The diagenetic realm is divided into three broad subsurface settings - eogenetic, mesogenetic and telogenetic (Figures 8, 9). The effects of ongoing alteration are evident in the textures of all carbonates. In fact, for many calcite and dolomite reservoirs little evidence remains, either mineralogically or texturally, of the original metastable aragonite or Mg-calcite.
The eogenetic realm is defined as the subsurface interval where diagenetic fluid crossflows are related to processes active in the depositional setting (e.g. tidal pumping, capillary evaporation, brine reflux). It extends from the surface down to depths where surface-related processes are no longer active (Figure 8; Choquette and Pray, 1970).
It transitions with depth into the mesogenetic realm. The mesogenetic domain is defined by subsurface fluid crossflows of the burial realm. Fluid flow rates are slower than in the eogenetic realm, and driving processes tend to be varying gradients, related to temperature, pressure and salinity contrasts. Rock fluid interactions in the mesogenetic realm may be diffuse, as in carbonates where some residual matrix permeability remains, of more focused as in the vicinity of active fractures and faults and zones of pressure solution.
When sediment is uplifted and its overburden eroded, its diagenetic character moves into the telogenetic realm (Figure 8). This is the region where pore fluid flow is once again driven by surface-related processes such as deep-meteoric alteration and dissolution (karstification). This set of overprints can sometimes be tied back to its formative unconformity.
The three styles of diagenetic overprint can be tied to a variety of porosity and permeability altering processes textures and isotopic signatures (Figures 8, 9). Each of these porosity and permeability modifying process sets has a particular set of geometries, along with indicative mineralogical/ geochemical/ isotope and sequence stratigraphic signatures that separate this type of reservoir from the much simpler largely depositional models utilised in siliciclastic models (Table 4; the various diagenetic processes, geometries and indicators are discussed in detail in other training modules).
8. While some form of subsurface permeability is present (e.g. interconnected macro-, micro-, intercrystalline and fracture porosity), then carbonate sediment mineralogies, textures and petrophysical properties evolve in the diagenetic realm
9. Zones and processes of diagenesis
Table 4. Significant differences between petrophysical character of sandstone and carbonate reservoir continue to evolve in the subsurface realm
This short overview is a limited sampling of some of reservoir skill sets offered across various training modules in our in-house and workshop sessions
At Saltworks, the aim of all our training modules and workshops is two-fold. 1) give an understanding of the relevant process, 2) train participants in the application of the skill sets tied to the concept and prioritise the skill sets needed to apply this understanding. Below we illustrate the skills and knowledge necessary to recognise subsurface evaporite salts using a conventional suite of well logs.
If you want to know more, please download the relevant saline geosystems or carbonate geosystems catalogue and choose a combination of the various training modules that best suite your company needs.