Locating Sweet Spots: Shale Petroleum Systems
In the last decade US oil production increased by 60%, to over 13.4 MMbopd by early 2020. This increased production, which has been partly responsible for depressing oil prices, has been made possible essentially because of the shale revolution, since US conventional crude production has been in decline for decades.
The shale revolution in the US is currently limited to a few basins and formations: the Permian, Barnett and Eagle Ford in Texas, the Bakken in North Dakota, and the Marcellus in the Appalachian east. However, there are numerous other shale plays in North America and elsewhere in the world, and many other countries regard their shale resources producible if economic and technological conditions became optimal. Meanwhile, there are some fundamental questions about shale science that we do not understand. Given that shale formations are self-sourced and self-sealed reservoirs, how can we apply petroleum system analysis, a methodology developed for conventional reservoirs and prospects, to these formations? Can petroleum system analysis reduce risk and increase productivity of shale petroleum? This article aims to address these questions.
Petroleum System Analysis
When the oil industry began in the 1860s, all oilmen wanted to find were seeps and anticlines to drill. In the 1920s, they began to develop concepts and technologies to quantitatively study reservoir rocks. For the next 50 years, subsurface mapping and characterization of reservoirs and traps by geological and geophysical methods was the focus in exploration. It was not until the 1970s that source rock studies drew serious attention from the industry and geochemists developed techniques to identify kerogen types and estimate thermal maturity of source rocks, based mainly on vitrinite reflectance microscopy. As computational and digitization techniques were developed in the 1980s, it became possible to perform basin-scale modeling of oil generation, migration, and accumulation, as geochemists developed ideas and methods to characterize petroleum source rocks.
A petroleum system consists of physical elements and associated processes and here we briefly describe them for shale formations.
Source Rock and Generation
Petroleum source rock contains kerogen – highly complex organic compounds that thermally crack to hydrocarbon molecules as the rock is progressively buried and heated by overlying sediments. The source rock is the petroleum basin’s ‘kitchen.’ Black claystone (mudstone, shale and marl) is the best source rock because it retains plenty of kerogen within its pore space. Limestone can also be a good source rock. Coal essentially generates natural gas, although deltaic coal formations can produce some oil.
Not all clay-rich rocks generate hydrocarbons; it depends on the amount of organic carbon in the rock, which is measured as percentage of total organic carbon (TOC). Less than 0.5% suggests poor source rock, while TOC of 2–4% indicates very good source rocks. Aside from high TOC, the source rock should also possess organic hydrogen (measured as Hydrogen Index or S2/TOC from pyrolysis); otherwise, pure organic carbon would produce graphite.
Moreover, the rock should be sufficiently buried and heated to generate hydrocarbons. There are several methods to estimate the paleo-temperatures of source rocks, including burial history diagrams, Tmax readings from S2 peak in pyrolysis experiments, and vitrinite reflectance (R0) measurements. Hydrocarbons are generated in a temperature-dependent, step-wise sequence of kerogen (insoluble in ordinary organic solvents) to bitumen (soluble in organic solvents) to heavy crude, light crude, wet gas, and dry gas. In this fractionation, hydrocarbons become simpler and lighter molecules.
Many Faces of Migration
Once oil and gas have been generated in the source rock, some will be expelled and will enter the basin, but some will remain within the formation. Hydrocarbon expulsion from the source rock is called primary migration. The expelled oil and gas, being buoyant, will then flow (secondary migration) through porous carrier beds or open fractures and will eventually accumulate in enclosed reservoirs (pools). Most of the expelled hydrocarbons, however, diffuse in the basin.
The ratio of how much oil or gas is expelled from the source rock and how much is retained in the rock formation differs for various source rocks depending on their conditions. Studies of Barnett Shale, for instance, suggest that 60% of generated oil was expelled and 40% was retained in the shale. However, some researchers do not consider expulsion mechanisms to be very effective, and believe that perhaps as much as 80% of hydrocarbons are retained in the source rock, possibly because oil saturation in shale plays has been found to be over 75% of the rock’s porosity. How much hydrocarbon is generated in shale and how much is expelled or retained remain interesting questions to be investigated using various cases and simulations.
The retained oil in shale can both stay in place and migrate within the formation. Intra-formational migration seems to be counter-intuitive because shale as a source rock is a very tight (low permeability) oil-wet rock. But two important mechanisms can migrate oil or gas within the shale formation. First, shale formations are not homogenous; they contain relatively high-permeability layers (sand, silt or limestone) and fractures which can facilitate oil and gas migrating updip the formation. Second, during the compositional fractionation of hydrocarbons (from heavy oil to dry gas), the buoyancy of lighter hydrocarbons (with high APIs, high solution gas-to-oil ratios and lower viscosity) will provide a migratory force. Compositional fractionation of hydrocarbons also happens during production when the produced hydrocarbons have different gas-to-oil ratios than their initial values in the rock.
Intra-formational migration observed in the US Barnett, Niobrara and Bakken Formations indicates that polar hydrocarbon compounds (resins and asphaltenes) remain in situ because of their high sorption (adsorption and absorption) in kerogen porosity, while saturates and aromatics move into organic-lean porous intervals.
Reservoir Storage and Production
A petroleum reservoir has two functions: firstly, it stores oil or gas; and secondly, it acts as a conduit to yield oil or gas into production wells. The storage capacity is measured by porosity (the percentage of pore space in a volume of rock); the yield capacity is measured by permeability (milliDarcy for conventional reservoirs).
Shale formations are tight rocks with clay-sized grains less than 4 μm diameter and small pore throats 0.1–0.005 μm. Shale formations have porosities less than 10% and rock permeabilities in nanoDarcies. Because of these tight properties, shale targets are stimulated by hydraulic fracturing to produce oil or gas.
Porosity structure in a shale is divided into matrix (inorganic) porosity, kerogen (organic) porosity, and natural fracture porosity. Free (mobile) oil resides in the matrix and facture pores; adsorbed and absorbed oil is in kerogen (organic) porosity. Organic matter has 44 times more adsorptive power than quartz; therefore, kerogen porosity is important, although the quality and producibility of sorbed oil in shale should also be considered. Fracture porosity has not been observed to be a major storage feature. Shale porosity can be calculated on samples in a laboratory or from wireline log measurements such as bulk density, neutron porosity, and sonic logs, although these have traditionally been developed and calibrated for sandstone and carbonate reservoirs, so well log petrophysics for shale requires more work.
Fractures provide the main permeability in shale formations. Natural fractures in shale may be divided into four categories: bedding-parallel fractures (if mudstone has been deeply buried to develop fissile platy structures, so typical of shale); bedding-vertical joints with spacing related to the layer thickness: fracture spacing increases with thicker beds; tectonic fractures associated with folding and faulting; and micro-fractures arising from hydrocarbon generation in the rock (conversion of kerogen to hydrocarbon involves volume increase and fluid overpressure in the rock). Natural fractures in the rock may be open or closed (healed by mineral veins or compressional stress) but only open fractures will be conductive.
Sealing and Entrapment
In conventional (migrated hydrocarbon) reservoirs, cap (seal) rock is necessary to prevent the upward migration of oil or gas. The cap rock, usually mudrock or salt, is a fine-grained, water-wet, low porosity and low permeability formation. By contrast, except for the expulsion or tectonic fracturing processes, shale formations are self-sealed reservoirs. Traps are geometrical configurations that provide a closure for the accumulated oil or gas in the reservoir.
Traps in conventional prospects are divided into structural, stratigraphic, hydrodynamic and combination types. This classification is not directly applicable to shales; however, lithological variations (both lateral and vertical) related to depositional processes play an important role in the framework of shale formations.
Classification of Shale Reservoirs
Compared to sandstone and carbonate reservoirs, our knowledge of shale as a petroleum reservoir remains poor. In recent years, some researchers have classified shale reservoirs in North America into three categories:
- tight shale, in which source and reservoir localities are the same;
- hybrid shale, where shale is interbedded with more porous but organic-lean siliceous or calcareous layers;
- and fractured shale, often rich in heavy oil or dry gas, where natural fractures provide considerable permeability and porosity.
Of course, these are end-member categories. It is inconceivable to have a shale formation which does not exhibit lithological heterogeneity or has remained unfractured.
Some researchers divide shale reservoirs into argillaceous, calcareous and siliceous, based on the relative abundance of clay, carbonate, and silica minerals in the rock. Ternary diagrams using these mineralogical data indicate if the shale is more brittle (high silica) or not. These are informative exercises to understand the response of the shale formation to hydraulic fracturing.
A third classification is to divide shale plays into ‘active’ (still in the oil or gas window) or ‘inactive’ (overcooked or uplifted) systems. It is important to know if a shale formation has producible oil or not before drilling. Daniel Jarvie has suggested that if S1/TOC ratio is greater than one (i.e. >100 mg oil/g TOC), the shale has producible oil, although he also notes that measurements of S1 peak (free hydrocarbon in pyrolysis) and TOC need to be reasonably accurate for this exercise.
Back to the Future
Shale gas production dates back to the 19th century and hydraulic fracturing began in the 1940s. Nevertheless, at the turn of this century, no one would have predicted the coming shale revolution. Shales were only viewed as sources or seals; they were not cored, logged or studied for their reservoir properties. Shale petroleum system analysis is, therefore, in its infancy. Nevertheless, several important points require attention.
- Firstly, total petroleum systems include both migrated and self-sourced reservoirs; this is a major addition to the global hydrocarbon budget.
- Secondly, like conventional ones, petroleum systems in shale are dynamic, in which all the elements and processes need to be evaluated and integrated to locate sweet spots for drilling and production. Given the heterogeneities and complexities of shale formations, it is necessary to conduct petroleum system analyses in numerous locations of the same shale play.
- Thirdly, the shale revolution has motivated us to investigate significant uncertainties in our measurements of TOC, pyrolysis (for example, correcting for evaporative loss of hydrocarbons in S1 peak), porosity, permeability, and so forth. Shale production requires higher resolution geoscience.
Finally, production from shale formations are drastically different from conventional reservoirs. Recovery in shale is less than 10% for oil and about 15–20% for gas, compared to about 50% oil recovery and up to 80% gas recovery in conventional reservoirs. Shale wells have sharp decline rates and last only a couple of years, compared to decades-long conventional wells. Moreover, shale production consumes a lot of water for fracking and is associated with induced seismicity, fugitive methane and other environmental issues. All this calls for better science and technology.