by Vaclav Smil
Figure 2.2 Supergiant gas fields in Western Siberia.
Nonconventional gas is present in four major formations. Enormous quantities of gas are locked in tiny bubbles in shales, a resource whose recent commercial exploitation in the United States has been so extensive and so rapid that many commentators have called it a revolution. Natural gas is also locked in tight formations in exceptionally hard, impermeable rocks. Presence of methane in coal beds has been an unwelcome reality due to an increased risk faced by miners as a result of sudden underground explosions—but in many areas, that gas offers a much cleaner alternative to highly polluting coal. And natural gas is also concentrated in massive underground and undersea deposits of methane hydrates (or clathrates), with the gas trapped inside a cage-like lattice of ice.
As yet there is no commercial recovery of methane hydrates, and most gas-producing countries tap only one or two of the remaining gaseous resources. The United States extracts all of them, and output statistics show the relative importance of these four principal sources: in 2011, 43% of gross gas withdrawals came from gas wells, 21% from oil wells, 6% from coal bed wells, and 30% from shale gas wells and tight gas deposits, the category whose output was lower than coal bed gas as recently as 2007 (USEIA [US Energy Information Administration], 2014a). I will take a closer look at all nonconventional sources of natural gas in Chapter 6, and in this section, I will explain where we search for conventional gas.
Finding fossil hydrocarbons became easier once we came to understand their biogenic origins beginning with large accumulations of kerogens. Looking for hydrocarbons means focusing on sedimentary basins where layers rich in organic matter became source rocks, where permeable, porous sediments became reservoir rocks and where, due to structural or stratigraphic reasons, these porous rocks became sealed at the right time by impermeable cap rocks. Magoon and Dow (1994), building on at least half a dozen previous syntheses, related all of these elements and processes in terms of the petroleum system, later renamed the total petroleum system (TPS).
This naturally occurring hydrocarbon system entails all the elements and processes required for large-scale accumulation of crude oil and natural gas: thermally mature source rock; a complex sequence of hydrocarbon generation, primary migration, and storage in a porous reservoir rock; and correct timing of forming a trapping configuration with an impermeable seal and overburden covering the system that help to preserve hydrocarbons from bacterial biodegradation, evaporation, and leakage (Magoon and Schmoker, 2000; Peters, Schenk, and Wygrala, 2009). This all takes time, and hence, there are no massive gas reservoirs only 103–105 years old. More than half of all kerogens come from the middle and late Mesozoic era: nearly a third of these oil and gas precursors are about 100 million years old (originating in the mid-Cretaceous era), a quarter goes back about 150 million years (late Jurassic era), and most of the rest originated between the late Devonian and early Cambrian, 360 to more than 550 million years ago. This means that the combustion of natural gas oxidizes carbon that has been commonly sequestered for 108 years.
Conventional oil and gas are not usually found in close proximity to kerogens because almost invariably the release of hydrocarbons from their source rocks (the primary migration) was followed by the secondary migration through permeable rocks, inevitably a very slow progression on the order a few km over a million years; the largest reservoirs of oil and gas are found in place where these rock are capped by impermeable traps that allow gradual, and often immense, accumulations of hydrocarbons that may be quite far (more than 200 km) from the source rocks. In an overwhelming majority of cases, oil and gas reservoirs (much like water reservoirs) are not large caverns filled with hydrocarbons; they are rocks whose porosity and permeability allowed first the ingress and then a lasting storage of large volumes of liquids and gases that could eventually, after drilling, flow upward through wellbores.
As a result, hydrocarbon reservoirs range from large, thick, and contiguous formations to small, shallow, and broken-up storages. More than half of all reservoir rocks are of Mesozoic age, almost two-fifths are Cenozoic, and a small remainder are the oldest Paleozoic sediments, with carbonates and sandstones dominant (Slatt, 2006). Carbonate rocks harbor some of the world’s most notable hydrocarbon reservoirs. They were deposited in shallow seas either through the precipitation of calcium and carbonate ions or the biomineralization by marine organisms (mainly by coccolithophorids and foraminifera). Limestones are composed of calcite and aragonite, both being CaCO3, calcium carbonate, only with different crystal structure. Dolostones are composed of more porous dolomite, CaMg(CO3)2. As for their age, many of the world’s largest hydrocarbon reservoirs, including the enormous accumulation in the Persian Gulf, Arabian Peninsula, and West Texas, are in Cretaceous, Jurassic, or Permian carbonates deposited 66–286 million years ago.
Clastic sediments (including sandstones, siltstones, and shales) arise after transport and redisposition of fragmented rocks. This can take place in meandering rivers and in braided streams, with Triassic age sandstones of the Prudhoe Bay on the northern coast of Alaska being an excellent example of the latter process. In China, there are many sediments of lacustrine origin, deposited from ancient large lakes in Qaidam and Junggar Basins in China’s western interior, and the country’s largest gas field, Sulige in Ordos (discovered in 2000), contains almost 1.7 Tm3 of tight gas (of which nearly 900 Gm3 are recoverable) in Carboniferous and Permian deltaic sandstones and in Upper Permian lacustrine mudstones (Yang et al., 2008). Reservoirs in deltaic deposits and submarine fans include fields along the northern shores of the Gulf of Mexico, in the Niger’s delta in the Gulf of Guinea, Trinidadian fields in the Orinoco’s delta, and the fields in the southern part of the Caspian Sea.
High porosity of reservoir rocks can be the result of initial rock formation or subsequent fracturing and reformation, and effective porosity (total volume of interconnected pores) is usually much higher in sedimentary formations (primary porosity) than in highly fractured igneous or metamorphic rocks (secondary porosity). Porosity of sedimentary rocks (pore volume/total volume of reservoir rock) is commonly between 10 and 20%; in some smaller volumes, it can be higher than 50%; and 8% is about the lowest level allowing conventional hydrocarbon extraction. In the United States, it ranges between 1 and 35% in carbonate reservoirs and averages 12% in limestones and 10% in dolomites (Schmoker, Krystinik, and Halley, 1985). Examination of nearly 37,000 producing reservoirs shows expected correlation with depth and age: older and deeper reservoirs are less porous, with the youngest siliciclastic reservoirs averaging more than 20% and pre-Cambrian carbonate reservoirs staying mostly below 10% (Ehrenberg, Nadeau, and Steen, 2009).
Permeability, the capacity to transmit fluids and gases through porous materials, is the other key attribute, and it is primarily determined by the size of grains and by the properties of the medium. Its effective rate is usually measured in darcy units (D, after Henry Darcy). Impermeable rocks that form reservoir caps (NaCl, CaSO4) have permeability of just 10−6 to 10−8 D. In contrast, highly permeable reservoir rocks have permeabilities in excess of 1,000 mD (1 D), typical permeabilities of hydrocarbon reservoirs are 10–100 mD, but in some reservoirs, permeability may differ by orders of magnitude between different layers and areas.
Traps that contain gases and liquids in place are classified either as structural (arising, often on a vast scale, through deformation of the crust) or stratigraphic (usually smaller, formed either by gradual accumulation of impermeable rocks or by sudden shifts) and must be topped by impermeable layers. Anticlines, large convex (arched, domed) folds, are by far the most important structural traps confining roughly 80% of the world’s largest hydrocarbon reservoirs (Figure 2.3). Among the best traps are anticlines created by rising salt domes and, in addition, they also appear often together with impermeable anhydrite (CaSO4) or gypsum (CaSO4·2H2O), making perfect seals.
Figure 2.3 Anticlines.
The largest anticlines arise by
compression along the convergent zones of tectonic plates: the Zagros fold belt of Iran is their foremost example. Among the supergiant gas fields, sandstone anticlines include the West Siberian Urengoy, Kuwaiti al-Burqān, Indonesian Minas on Sumatra, Shatlyk in Turkmenistan, Alaska’s Prudhoe Bay, as well as several layers of the Qatari/Iranian North Dome/South Pars field. Among the major North Sea fields, those producing both oil and gas include British Brent (discovered in 1971) and Norwegian Ekofisk (1969), Statfjord (1974) and, above all, Troll, a sandstone anticline discovered in 1979 at a depth of 1,400 m and holding nearly 130 Gm3 of recoverable gas.
Putting all of these attributes together, the standardized descriptions of four of the world’s largest natural gas fields read as follows. The world’s largest accumulation of natural gas, the North Dome/South Pars in the Persian Gulf between Qatar and Iran (discovered in 1971), is a highly stratified Permo-Triassic carbonate reservoir with anticlinal trap about 2,900 m below the sea bottom, 200 m thick with porosity of 9.5% and permeability of 300 mD whose recoverable reserves are at least 22–24 Tm3; deep Silurian (Qusaiba) shale is the most likely source rock, with gas coming from two zones (Upper Dalan and Kangan), about 200 m thick and 2.9 km below the sea bottom, with porosity of about 10% and permeability of 300 mD (Esrafili-Dizaji et al., 2013). The gas is abnormally pressured but it contains high levels of H2S as well as of CO2.
The world’s second largest natural gas field, Galkynysh in southern Turkmenistan in the Amu Darya Basin (discovered in 2006), is actually a cluster of fields (South Yolotan, Osman, Minara, and Yashlar) with anticlinal traps. The former number two, Urengoy in the Western Siberian Basin (discovered in 1966), is a Cretaceous sandstone reservoir with anticlinal trap, about 70 m thick and 1–3.1 km below the ground, with porosity between 14 and 20% and permeability of 7–170 mD (Milkov, 2010; Li, 2011). Its source rock is the late Jurassic Bazhenov shale formation (which also contains tight gas), and as in virtually all reservoirs in the basin, the gas is nearly pure methane with no H2S.
Europe’s largest continental field, Groningen near Slochteren in northern Netherlands (discovered in 1959), is a Permian sandstone reservoir with uplifted and faulted trap, about 160 m thick and 2.7–3.6 km below the ground, with porosity of 15–20% and with permeability ranging over four orders of magnitude from less than 1 mD to as much as 3000 mD (Li, 2011). Its source rocks are about 300 million years old carbonaceous shales. America’s largest natural gas field, Hugoton–Panhandle in Kansas, Texas, and Oklahoma (discovered in 1918), is a sprawling (nearly 24,600 km2) formation whose reservoir rocks include carbonates, dolomites, and limestones, with a complex, giant stratigraphic trap. Its most prominent source rock is late Devonian shale (older than 350 million years), the reservoir thickness is about 170 m, the average porosity is 9.2% (but modal value is just 7%), and the permeability has a huge range from 0.0001 to 2,690 mD; the reservoir has subnormal pressure and contains gas of variable composition (Sorenson, 2005; Dubois et al., 2006).
When the USGS assessed the world’s oil and gas resources (Magoon and Schmoker, 2000), the five largest TPS ranked by their known gas volume were the Northern West Siberian Mesozoic Composite TPS (including Urengoy, Yamburg, and Bovanenkovo) with more than 33 Tm3 of known natural gas; Zagros–Mesopotamian Cretaceous–Tertiary TPS (about 14 Tm3, including the North Dome/South Pars); Silurian Qusaiba (nearly 13 Tm3); Arabian Sub-Basin Tuwaiq/Hanifa-Arab (nearly 8 Tm3); and Amu Darya Jurassic–Cretaceous TPS with more than 6.5 Tm3, including Shatlyk (Ulmishek, 2004). Subsequent discovery of Galkynysh group of fields changed that order, most likely moving up the Amu Darya system to the second place: estimates for Galkynysh have been repeatedly revised upward, the latest claimed proved and probable reserve total is as high as 26.2 Tm3, the latest USGS assessment of undiscovered oil and gas resources of the Amu Darya Basin puts the total gas volume at 52 Tm3 (Klett et al., 2011), and the field’s total output is expected to reach 40 Gm3/year in 2015.
In some instances, the tertiary migration brings some oils and gases all the way to the Earth’s surface: crude oil was found oozing through rocks (in western Pennsylvania) and forming pools in many locales in the Middle East, while seepages of gas were known in antiquity by hissing sound of the escaping gas and by strange phenomena of “burning springs” or “eternal fires,” most notably in what is today’s northern Iraq (Kurdistan). A notable North American description of this phenomenon (a burning spring in the Kanawha River Valley of West Virginia) is worth noting as it is included in the Schedule of Property appended to George Washington’s will: “The tract, of which the 125 acres is a moiety, was taken up by General Andrew Lewis and myself, for and on account of a bituminous spring which it contains, of so inflammable a nature as to burn as freely as spirits, and is nearly as difficult to extinguish” (reprinted in full in Upham, 1851, 385).
Finally, a few paragraphs on energy storage densities (J/m2) of natural gas reservoirs. Methane has energy density three orders of magnitude lower than crude oil, but because of the thickness of some gas-bearing layers, storage densities of the fuel (energy content prorated per unit of surface) for the largest gas fields are similar to those of the world’s largest oil fields. South Pars–North Dome field, the world’s largest identified accumulation of hydrocarbons in the Persian Gulf, contains about 35 Tm3 of recoverable natural gas (about 70% of its estimated total volume of 51 Tm3), and the field had originally stored also about 8 Gm3 of natural gas liquids (Esrafili-Dizaji et al., 2013). This adds up to about 2.1 ZJ of fossil energy, and (with the area of 9,700 km2) it translates to storage density of about 215 GJ/m2 of sea surface.
The much smaller Urengoy (original storage of 8.25 Tm3, recoverable volume of 6.3 Tm3, area of 4,700 km2) in Western Siberia is the world’s second largest natural gas field, and its storage density is roughly 60 GJ/m2 of tundra (Grace and Hart, 1991). The third largest storage, Yamburg field north of the Arctic Circle in Siberia’s Tyumen region, has nearly 4 Tm3 of recoverable gas and overall storage density of about 35 GJ/m2 of permafrost. Analogical data for Groningen, Europe’s largest onshore gas field, are 2.8 Tm3 and about 110 GJ/m2 (NAM (Nederlandse Aardolie Maatschappij), 2009) and 2.3 Tm3 and less than 4 GJ/m2 for Hugoton, America’s largest natural gas field. That sprawling formation of almost 22,000 km2 extends from southwestern Kansas through Oklahoma to Texas. For comparison, the world’s largest reservoirs of crude oil have storage densities mostly between 200 and 500 GJ/m2 (Smil, 2015).
2.3 RESOURCES AND THE PROGRESSION OF RESERVES
Unless we accept widespread abiogenic origin of hydrocarbons (which would allow their continuing repletion, albeit at a very slow rate), we must reckon with finiteness of natural gas stored in the uppermost layers of the Earth’s crust. Extraction of all minerals, and hence of all fossil fuels, inexorably reduces the stores that were originally in place—but this process will almost never result in actual physical exhaustion of the exploited resource. In order to appreciate this fundamental reality, it is necessary to clarify the basic definitions involved in the exploitation process: different schemes have been proposed in several countries, but the one that has become the US norm is the classification used by the US Geological Survey (McKelvey, 1973; Figure 2.4).
Figure 2.4 McKelvey box.
In the United States, all publicly listed companies must file annually their natural gas reserve totals with the US Security and Exchange Commission. Those numbers are known to be rather conservative, while the figures for many OPEC nations have been exaggerated: the resources may be in place but proved reserves may be considerably smaller; this uncertainty is particularly important in Iran’s case (Osgouei and Sorgun, 2012). Russia reports its reserves based on a classification adopted during the Soviet years that is not directly comparable with the Western assessments: its reporting is based solely on the analysis of geological parameters, not on an economic appraisal of reserves that can be actually recovered—but natural gas re
serves in the first three categories (A, B, and C1) are considered to be fully extractable (Novatek, 2014a).
I will explain the US classification by relying on an analogy of a large dark room whose content we proceed to examine with increasingly more powerful searchlight prior to removing its contents. The room itself is analogous to resource base: its ultimate size is unknown but (unreliable) estimates can be offered as to its volume or mass. The first entries into the room and the first examinations of its contents usually yield modest results: the room’s volume is vast and our search light is not very powerful (sophisticate geophysical exploration was initially absent, drilling was limited to shallow depths), but we begin to get some feeling for what is in there. This knowledge is analogous to assessing a specific resource, that is, a concentration of materials (be they solid, liquid, or gaseous) in the Earth's crust whose economic extraction we regard as feasible, either right away or sometime in the future. In some instances, our searchlight can illuminate some large and very valuable objects (early discoveries of large hydrocarbon fields) which we promptly proceed to remove.
More specifically, resources get subdivided into economic (ready to be exploited), marginally economic, and subeconomic categories, and based on their size, we can begin to make some estimates about the further magnitude of undiscovered resources, beyond our searchlight but reasonably assumed. Further, detailed examination of viable resources moves them from subeconomic and marginal categories to the economic reserve category, itself subdivided into demonstrated (measured and indicated) reserves and more uncertain class of inferred reserves. Commonly used equivalents of measured and indicated reserves in other resource classifications are the categories of proved and probable reserves.
