by Vaclav Smil
This brief review shows that natural gas is an exceptional source of primary energy for all modern economies—not only because it combines a number of practical advantages and benefits but also because it either fits existing infrastructures or because any further expansion of extraction, transportation, and use can rely on proven and economical technical and managerial arrangements. This book will offer a concise, but systematic, review of key aspects that define and delimit the fuel’s great potential during the twenty-first century: its biogenic origins, crustal concentrations, and widespread distribution; its exploration, extraction, processing, and pipeline transportation; its common uses for heat and electricity generation and as a key feedstock for many chemical syntheses; its pipeline and intercontinental exports and the emergence of global trade; diversification of its commercial sources; and the fuel’s role in energy transitions and environmental consequences of its combustion.
Given the scope of the coverage and the length of the text, it is easy to appreciate that writing this book was a serial exercise in exclusion; at the same time and with every key topic, I have tried to go sufficiently below the shallows that now dominate the (web page-like) writings about technical, environmental, and economic complexities for nonexperts. That is why I hope that those readers who will persevere (and tolerate what I always think of as a necessary leavening by many requisite numbers) will be rewarded by acquiring a good basis for an unbiased and deeper understanding of a critical segment of modern energy use and of its increasing importance.
2
Origins and Distribution of Fossil Gases
Many gases encountered in the biosphere are of organic (biogenic) origin: as already explained in the previous chapter, methane is actually the best example of a gas whose concentration is orders of magnitude higher than it would be on a lifeless Earth. Other gases generated by microbial metabolism include hydrogen (H2), hydrogen sulfide (H2S), carbon monoxide and dioxide (CO and CO2), nitric oxide (NO), nitrous oxide (N2O), and ammonia (NH3), while photosynthesis releases huge volumes of oxygen (O2), and all animal metabolism emits CO2. But many of these gases are also of inorganic origins: CO from incomplete combustion of fuels and CO, H2S, and SO2 from volcanic eruptions—and methane was found in hydrothermal vents at the bottom of the Pacific Ocean and in other crustal fluids.
Given these realities, we have to ask if all fossil methane, or at least an overwhelming majority of its presence in the crust’s topmost layer, is of biogenic origin (arising from transformation of organic matter) or if a significant share of it (if not most of it) is abiogenic, arising from processes that have not required metabolism of living organisms. These obvious questions have been asked and answered: debates about the origin of natural gas have been a part of a wider inquiry about the genesis of fossil hydrocarbons, with the consensus coming on the side of their organic provenance and with dissenters arguing that inorganic process is a better way to explain the genesis of vast gas volumes in the Earth’s upper crust.
I will recount this contest in the first section of this chapter before I get into a fairly systematic review of the known global distribution of natural gas resources and a few closer looks at major resource endowments in key producing countries and at some of the world’s most notable natural gas fields. In the chapter’s final section, I will assess the development of our understanding of the world’s natural gas resources (whose total cannot be quantified with satisfactory accuracy) as well as the progression of natural gas reserves (whose appraisal, while far from perfect, offers a much more revealing understanding of realistic, economic prospects for exploiting the fuel).
2.1 BIOGENIC HYDROCARBONS
A widely accepted conclusion is that natural gas was formed by thermal decomposition (under temperatures of 150–200°C and higher), primarily of crude oil in reservoirs and, to a lesser degree, of organic matter in shales that were the source rock for the liquids (Hunt, 1995; Selley, 1997; Buryakovsky et al., 2005). But organic origins of oils are not as obvious as those of coals which contain fossilized trunks and imprints of leaves of trees that grew during the Carboniferous periods (360–286 million years ago) and in more recent pre-Quaternary eras (many lignites are younger than 10 million years). While coals are fossilized phytomass of the largest photosynthesizers, crude oils originated mostly from dead biomass of the smallest photosynthetic organism, single-cell phytoplankton composed mostly of cyanobacteria (photosynthetic bacteria that used to be known erroneously as blue-green algae) and diatoms (unicellular and colonial algae enclosed in siliceous cell walls), and from dead zooplankton (dominated by foraminifera, amoeboid protists with food-catching pseudopodia). In addition, there were secondary contributions of dead algal phytomass, invertebrate and fish zoomass, and assorted dead organic matter carried to oceans and lakes by rivers.
But there is no simple, direct path to transform microbial biomass into a pool of oil hydrocarbons: they originated mostly from nonhydrocarbon organic compounds (carbohydrates, proteins, and lipids) that were transformed by bacteriogenesis (microbial metabolism) followed by lengthy thermogenesis (heat processing within sediments). The process begins with gradual accumulation of dead biomass in coastal or lake sediments followed by aerobic microbial degradation releasing CO2 and then by anaerobic fermentation releasing CH4 and H2S, and burial and compaction of organic matter in anoxic sediments produced complex mixtures of large organic molecules. These insoluble kerogens arise from different materials—mainly from algae in lacustrine settings, from plankton and some algae in marine formations, and mainly from higher plants in terrestrial environments—and they could be up to 10% (commonly just 1–2%) of the mass in shales and limestones, the usual source rocks of fossil hydrocarbons. Rate of their formation and their eventual concentration is an outcome of competing processes of accumulation, destruction, and dilution of organic matter.
Kerogens subjected to higher temperatures and pressures in buried sediments will be eventually degraded by a process that is very similar to deliberate actions in crude oil refineries: thermal cracking breaks up complex molecules, and kerogens are first transformed into dark (black or brown), near-solid bitumens. This transformative process, resulting in bitumen (dark, very viscous but inflammable organic matter) appears to be almost as old as life itself: residues of asphaltic pyrobitumen were found in Australia’s Pilbara shales and were dated to 3.2 billion years ago (Rasmussen, 2000). Eventual cracking of bitumens produces lighter molecules of liquid hydrocarbons, and increasing temperatures mark the principal stages of the process (McCarthy et al., 2011).
Transformation of sediments rich in organic matter to sedimentary rocks (diagenesis) takes place usually at less than 50°C and at a depth of less than 1 km. Thermal cracking (catagenesis) is most effective between 65 and 150°C and typically at depths of 2–4 km. Heavier liquid molecules are formed at temperatures between 80 and 120°C, and catagenesis at higher temperature results in higher shares of lighter alkanes and in gradually increasing gas/oil ratio, and any prolonged thermal processing above 200°C produces only dry gas. Finally, metagenesis converts much of the remaining kerogen into methane and nonhydrocarbon gases at 150–200°C. Stolper et al. (2014), using an isotopic technique, delimited formation temperatures of thermogenic gases between 157 and 221°C (Figure 2.1).
Figure 2.1 Diagenesis, catagenesis, and metagenesis.
Hydrocarbons in young (Cretaceous, 145–65 million years ago) reservoirs tend to be mostly heavy crude oils, lighter crude oils come from Jurassic or Triassic formations (younger than 250 million years), and the lightest alkanes are often of Permian or Carboniferous age (up to 350 million years old). As a result, the composition of hydrocarbon reservoirs spans a huge range from the extreme of highly viscous bitumen through heavy oils not accompanied by any gas to combinations of liquid and gaseous compounds (ranging from oils with only a small amount of dissolved gas to mixtures of oil, gas, and natural gas liquids) and to the other extreme of nearly pure methane with only a marginal presence of
natural gas liquids.
Efficiency of this long process of transforming ancient biomass carbon into carbon in marketed fossil fuels can be quantified in terms of right orders of magnitude (Dukes, 2003). During coal formation, up to 15% of plant carbon ends up in peat, up to 90% of peat carbon is preserved in coal, and the now dominant opencast extraction can get up to 95% of coal in place from thick and level seams. This means that the overall carbon recovery factor (which is the fraction of carbon’s original content in ancient phytomass that remains in extracted fuel) can be as high as 13%—or, restating this in reverse, that some eight units of ancient carbon (with the most common range of 5–20 units) was transformed into one unit of carbon in marketed coal.
In contrast, carbon preservation rates were much lower in marine and lacustrine sediments, and hydrocarbon recovery rate is only rarely close to 50%. As a result, the overall recovery factor for carbon sequestered in crude oil carbon could be as high as nearly 1% and lower than 0.0001%. Common rate of 0.01% means that 10,000 units of ancient carbon in aquatic biomass were transformed into 1 unit of carbon in marketed crude oil, and subsequent catagenesis and metagenesis, even if operating with 80% efficiency, would result in more than 12,000 units of carbon in ancient aquatic phytomass and zoomass to produce a unit of carbon in methane and natural gas liquids.
But what if hydrocarbons were of inorganic, rather than biogenic, origin? That was assumed by Dmitri Ivanovich Mendeleev, Russia’s leading nineteenth-century chemist, and that has been an alternative to the biogenic explanation offered by the so-called Russian–Ukrainian hypothesis about the abiogenic formation of oil and gas in abyssal environments. The theory was first formulated during the early 1950s, and it had been championed by its proponents in the USSR and by some Western geologists (Kudryavtsev, 1959; Simakov, 1986; Glasby, 2006). According to the theory, formation of highly reduced hydrocarbons with high energy content from highly oxidized organic molecules with low energy content would violate the second law of thermodynamics, and high pressures deeper in the Earth’s mantle are the best explanation for the formation of such reduced molecules.
Porfir’yev (1959, 1974) had also argued that abiogenic formation of giant oil fields is a better explanation of their origins than assuming truly gigantic accumulations of organic material that would be needed to create such structures. Modern interpretation of the abiogenic theory was summarized in a paper published in the Proceedings of the National Academy of Sciences and authored by Jason F. Kenney (a leading American advocate of inorganic origins of hydrocarbons) and his Russian colleagues. They recounted the experiments that produced a range of petroleum fluids in an apparatus replicating pressure (50 MPa) and temperature (up to 1500°C) 100 km below the surface (Kenney et al., 2002).
A kindred alternative was advocated by Thomas Gold, an American astrophysicist (Gold, 1985, 1993). Gold noted the presence of abiogenic hydrocarbons on planetary bodies devoid of life and maintained that methane can form by combining hydrogen and carbon under high temperatures and pressures in the outer mantle, and after this mantle-derived methane migrates it is then converted to heavier hydrocarbons in the upper layers of the Earth’s crust. If true, this would have two profound consequences: hydrocarbons created by degassing from the mantle would be much more plentiful than is indicated by estimates attributed to biogenic formation; and perhaps the most intriguing consequence of abiogenic hydrocarbon formation, existing reservoirs could be gradually recharged (albeit at a very slow rate) by continuing formation of oil and gas (Mahfoud and Beck, 1995; Gurney, 1997).
But the alternative explanations of oil and gas origins have not aged well. Once the Russian–Ukrainian theory became better known abroad (starting in the mid-1970s), it was repeatedly dismissed by most of the European and North American petroleum geologists who favor the consensus explanation that excludes any major contribution by abiogenic origins of hydrocarbons. This consensus view is strongly supported by geological and geochemical evidence, and it has been strengthened by the use of the latest analytical methods (Glasby, 2006; Sephton and Hazen, 2013). I will note here half a dozen of major realities that undermine the abiogenic hypothesis.
The upper mantle is too oxidizing to allow the persistence of significant amounts of H2 or CH4; hydrocarbons formed from outgassed methane should be (but are not) concentrated largely along major tectonic discontinuities (faults and convergent zones); modern understanding of fluid and gas migration explains many previously puzzling hydrocarbon occurrences; biomarkers (including porphyrins and lipids) are derived from organic molecules; and isotopic analyses show a match between carbon isotope ratios in hydrocarbons and in terrestrial and marine plants. Lollar et al. (2002) used isotopic analysis of carbon and hydrogen to show a clear distinction between thermogenic and abiogenic hydrocarbons, and due to the absence of appropriate isotopic signatures in economically important reservoirs, they ruled out the occurrence of globally significant abiogenic alkanes.
This does not mean that there are no hydrocarbons of inorganic origin and that we have satisfactory explanation for the formation of all major hydrocarbon deposits. Gold may have overstated his case for abiogenic methane, but he was right when he posited the existence of what he called deep hot biosphere (Gold, 1998), assemblages of extremophilic bacteria living deep underground, up to several km below the Earth’s surface, others deep below the deep-sea bottom. That claim was initially dismissed by the prevailing scientific consensus only to be confirmed later by incontrovertible and amazing findings of such organisms (Reith, 2011). And Milkov’s recent explanation of the formation of giant gas pools in Western Siberia is an excellent illustration of complexities involved in the genesis of hydrocarbons (Milkov, 2010).
Dry gas pools in the northern part of the West Siberian Basin contain about 11% of the world’s conventional gas reserves and account for 17% of current global gas extraction, but none of the proposed hypotheses of their origin (thermogenic gas from deep source rocks, microbial gas from dispersed organic matter, thermogenic gas from coal) are consistent with actual molecular and isotopic composition of extracted gases. Milkov argues that a significant (but unquantified) share of those shallow dry gases is the result of methanogenic biodegradation of petroleum rather than the outcome of thermogenesis. This also illustrates the complexities of postformation hydrocarbon histories. Oil and gas originating in kerogen-rich source rocks will almost invariably migrate through porous and hence permeable reservoir rocks and will be held in place by impermeable traps to hold the liquid in place: we find gas (and oil) only in those places where all of these conditions are appropriately combined.
2.2 WHERE TO FIND NATURAL GAS
Systematic exploration and assessment of fossil hydrocarbons present in the Earth’s uppermost crust resulted in the fundamental division into conventional and nonconventional natural gas resources, that is, the ones that are relatively easy to extract and those that are much more difficult and hence much more costly to recover or that cannot be tapped at all with existing production techniques. The first category of methane-dominated gas mixtures is made up of two distinct resources. Historically, the most common source of gaseous hydrocarbons has been associated gas, that is, a gas dissolved in crude oil and sometimes also forming caps in oil reservoirs.
The world’s largest oil field, Saudi al-Ghawār, is an excellent example of a reservoir containing both liquid and gaseous hydrocarbons: besides producing annually about 250 Mt (10.5 EJ) of oil, it also yields about 21 Gm3 (750 PJ) of associated natural gas (Sorkhabi, 2010). And in 1971, 30 years after the reservoir began producing crude oil, a large pool of nonassociated gas was discovered below the oil-bearing layers at a depth of 3–4.3 km, and this deep reservoir now produces annually about 40 Gm3 (1.4 EJ) of nonassociated gas. And other Middle Eastern oil supergiants are even richer in natural gas.
Al-Ghawār’s oil-to-gas reserve ratio is greater than 30; for al-Burqān, the world’s second largest hydrocarbon reservoir, it is only about 5; for the
Iranian Aghajari, it is about 3; and for Marun (Iran’s second largest oil field), it is less than 1.5. And in the Williston Basin of North Dakota, extracting oil from Bakken oil shale by horizontal drilling and fracturing has been accompanied by so much associated gas that, given the rapid growth of production and lack of adequate transport capacity, gas flaring has been so extensive that nighttime satellite images show large patches of light rivaling such large metropolitan areas as Minneapolis or Denver (see Chapter 7). In other basins, large stores of gas and oil may be found separately: Algerian Hassi R’Mel (discovered in 1958) is a part of a large Saharan Triassic basin that also contains supergiant oil field at Hassi Messaoud.
As the post-WW II demand for natural gas increased and as new large-diameter pipelines enabled long-distance transport and exports, the share of natural gas originating from oil fields began to decline, and currently, most of the world’s natural gas extraction comes from gas reservoirs. Some of them contain almost pure methane, with the world’s largest accumulation of natural gas in a large group of giant fields in the West Siberian Basin being the best example of this high purity (Figure 2.2). These Cenomanian fields (dating to oldest Late Cretaceous epoch of 100.5–93.9 million years ago) contain gas averaging 97.95% CH4, 0.23% C2H6, 1.58% N2, and 0.24% CO2 with traces of He and Ar, a mere 0.019% H2, and no H2S (Milkov, 2010). This composition makes the gas virtually pipeline-ready without any need for processing. But a much more common occurrence is methane with variable but relatively high shares of natural gas liquids (ethane to pentane): the former category, where more than 85% (and as much 95%+) of the volume is CH4, is commonly known as dry gas; when CH4 falls below 85% and the mixture is rich in natural gas liquids, it is wet gas.