by Vaclav Smil
By 1973, when OPEC’s actions led to the quintupling of global oil price, oil was the world’s leading fossil fuel, accounting for 48% of the global TPES, with coal a receding second at 27% and natural gas rising to about 18%. Oil’s further ascent was checked by OPEC’s continuing oligopolic manipulations (the second round of price rises had more than tripled the 1974 rate by 1981) that spurred the quest for higher conversion efficiency of refined fuels and for their replacement by other energies. Oil’s absolute consumption continued to grow, but by the year 2000, the fuel’s share fell to 38% of the global TPES and as gas supplied about 24% of the global demand. During the first dozen years of the twenty-first century, oil’s share fell further to 33%, and while the global extraction of natural gas rose by nearly 40%, its share of the TPES remained at about 24% due to a rapidly rising combustion of coal (thanks largely to expanded extraction in China and India) that pushed the fuel’s share to roughly 30% of the world’s TPES and had temporarily reversed the long-term energy transition from coal to hydrocarbons (in 2000, they supplied about 62% of the total and in 2012 only 57%).
When looking at the pace of energy transitions, I found some remarkable similarities between the global shift from traditional biofuels to coal and the subsequent substitution of coal by crude oil (Smil, 2010a; Figure 7.1). Coal reached 5% of the global TPES (a share I use as the marker of nonmarginal contribution) by about 1840, and its share was at 10% by 1855, 20% by 1870, 25% by 1875, 33% by 1885, 40% by 1895, and 50% by 1900. That spacing of milestones—15, 30, 35, 45, 55, and 60 years—is very similar to the sequence of identical milestones reached by crude oil as it was displacing coal and biofuels after it reached 5% of the global TPES (15, 35, 40, 50, and 60 years to reach 40%; oil will never supply 50% of TPES).
Figure 7.1 Global fuel transitions.
In comparison, the worldwide transition to natural gas has proceeded at a slower pace: the fuel reached 5% of the global TPES around 1930, 10% 20 years later, and 20% only 45 years later. And depending on the choice of energy conversions (above all the different ways of converting primary electricity to a common energy denominator), it has either reached 25% after some 70 years, or it has yet to reach that level (BP has it at 23.0% in 2012)—while coal claimed 25% of the fuel market 35 years after it reached 5% and oil did so after 40 years. Obvious absence of any acceleration in successive transitions is significant: moving from coal to oil has been no faster than moving from traditional biofuels to coal—and substituting coal and oil by natural gas has been measurably slower than the two preceding shifts.
Scale of the requisite transitions is the main reason why natural gas shares of the TPES have been slower to rise: replicating a relative rise needs much more energy in a growing system. In the year 2000, the global TPES was about five times larger than in 1950 and roughly 10 times the size of the supply in 1900, and hence (even in the absence of any resource constraint), it has become progressively harder for a new source to claim a significant share of the overall demand. As a result, raising oil’s share from 5 to 25% of the global TPES by 1945 called for about 1.6 times more energy than completing the same shift for coal by 1875—but going from 5 to 25% of natural gas required nearly eight times more energy than accomplishing the identical coal-to-oil shift.
Infrastructural challenges are the other obvious explanation as moving solids and liquids is much easier, especially across and among continents, than transporting natural gas. This slower pace means that those virtually always unreliable long-term consumption forecasts have been even more amiss as far as the market penetration of the natural gas has been concerned. But even this slower-than-anticipated shift toward natural gas has been most welcome because the fuel offers a less polluting alternative that is also the best practical means to lower the carbon burden of global energy consumption that still depends overwhelmingly on fossil fuels. Replacing fuel oil by natural gas to heat a house or coal by natural gas to generate electricity is a much easier task than replacing coal or oil by carbon-free renewables because it can be done economically and with assured availability and reliability.
In contrast, even when affordable, solar or wind electricity generation cannot guarantee constant availability—unless backed up by other forms of on-demand generation (gas turbines being one of the best choices!) or by requisitely large storage—but the storage option is practical only on a relatively small scale as we have no means (save for relatively rare and inevitably energy-losing pumped hydro facilities) of storing electricity to meet demand on the scale of hundreds of MW or a few GW (power flows required by today’s large cities). Consequently, all carbon-free renewable alternatives will have limited impact for the foreseeable future.
There is still considerable unexploited hydroenergy potential in Asia, Africa, and Latin America, but there are only few remaining prospects for large-scale hydroelectricity generation in North America and Europe. New renewables above all solar photovoltaics and wind have been capturing higher shares of national electricity generation in some countries (Germany being the leader among large affluent economies), but their market penetration has been proceeding at a relatively slow pace when measured as a share of the TPES: in 1990, 88% of the world’s TPES came from fossil fuels; in 2012, the share was 87% (Smil, 2014). In addition, nuclear generation has stalled, or is declining, in most affluent countries; its future is made even more uncertain by the Fukushima disaster and continuing problems with the site’s cleanup. Natural gas is thus the best option for near- and midterm decarbonization of the global TPES.
As already described (in Chapter 5), recent advances in LNG transportation have finally made a global gas market (akin to the long-established oil trade) an increasingly appealing economic option and have greatly improved the prospects for accelerated market penetration, but an even more important factor to determine the eventual outcome will be the extent and the pace with which natural gas can penetrate a key energy market in modern economies, that of transportation fuels. At the same time, we cannot assume that much higher dependence on natural gas will have only positive environmental impacts, and that is why I will assess all of the fuel’s environmental consequences in the closing segment of this chapter.
7.1 FUEL SUBSTITUTIONS AND DECARBONIZATION OF ENERGY SUPPLY
Inevitably, the universal process of energy transitions—including the sequence of fuel substitutions from wood to coal to oil to natural gas and higher reliance on primary electricity—has displayed many nation-specific variations. Some countries had never went through a coal stage as they moved from wooden age to economies based on crude oil, other nations (most notably China) remain still highly dependent on coal, and yet others rely on exceptionally high shares of hydroelectricity. But different transition paths have eventually the same important outcome as societies benefit from higher efficiency of final energy conversions, from less pollution associated with combustion from reduced intensity of carbon emissions. Gradual decarbonization of energy supply is an especially desirable trend as it helps to moderate the human interference in the global carbon cycle and slow down the rise atmospheric CO2.
Wood, the dominant energy source of the premodern and early modern (1500–1800) world, is mostly cellulose, hemicellulose, and lignin. Overall carbon content of these biopolymers averages about 50%, with a relatively narrow range of 46–55% (Lamlom and Savidge, 2003; Cornwell et al., 2009). Wood contains only about 5% of hydrogen, as does bituminous coal whose carbon content is around 65%. Atomic H:C ratio of wood is thus about 1.4 and of bituminous coal typically around 1.0, and hence, it would seem that the transition to coal did not result in any decarbonization of fuel use. But because a large share of wood’s hydrogen atoms is never oxidized (due to hydroxyl radicals that escape in early stages of combustion), the effective H:C ratio of wood is typically less than 0.5, and a shift to coal results in producing lower CO2 emissions per unit of fuel energy, and in practice, the reduction has been even greater due to higher combustion efficiencies of better designed stoves or boil
ers burning coal.
Liquid fuels derived from crude oil average 86% carbon and 13% hydrogen, and both gasoline and kerosene, with atomic H:C ratios of 1.8 (nearly twice as high as bituminous coal), resulted in a major shift toward decarbonization—and the gain is even greater when burning methane whose atomic H:C ratio is obviously 4.0. Specific carbon emissions thus decline from about 30 kg/GJ of wood to as little as 25 kg/GJ of excellent bituminous coal and 20 kg/GJ of refined liquid fuels, and burning of natural gas will release only 15.3 kg C/GJ (IPCC [Intergovernmental Panel on Climate Change], 2006). Obviously, global energy transitions (whose major features were quantified in this chapter’s opening section) have resulted in a persistent decline of specific carbon emissions (Figure 7.2).
Figure 7.2 Decarbonization of global energy supply.
When expressed in kg C/GJ of the global TPES, the rate declined from nearly 28 in 1900 to just below 25 in 1950 and to just over 19 in 2010, roughly a 30% decrease (Smil, 2013a). Among the affluent countries, the drop has been largest in France (due to its nuclear commitment, not because of natural gas), the US decline was nearly 40% (due to gas and nuclear), but China’s only about 25% (due to continuing high reliance on coal). When this worldwide decarbonization is quantified in terms of H:C ratio of fossil fuels, its global mean rises from 1.0 in 1900 to 1.6 by 1950 and 1.8 by 1980 and 1.9 by the year 2000. Subsequent slight reversal (to almost 1.8 by 2012) was caused by China’s massive addition of new coal extraction capacities.
This means that decarbonization of the global TPES has been proceeding at a slower rate than has been expected. Most notably, during the mid-1990s, Ausubel foresaw the global H:C mean of 3.0 in 2010 (the actual rate was just 1.83) and the global CH4 economy (H:C ratio of 4.0) shortly after 2030, and more recently, he still expected that CH4 will supply 70% of the global TPES soon after 2030 (Ausubel, 2003). That is not going to happen, but how far the substitutions will go remains uncertain, and I will assess the best evidence in the closing chapter. In any case, declines in the specific carbon intensity have yet to be translated into absolute decreases of global CO2 emissions (CDIAC, 2014).
The verdict is clear: rising consumption of natural gas has been a key cause of decarbonizing the global TPES, but the recent growth of gas supply could neither prevent further growth of carbon emissions nor slow down their growth to such an extent that the world would avoid going above 450 ppm of CO2 in decades to come. Much has been expected of natural gas, but there is only so much it can deliver in the world of rising energy demand. And, obviously, a much higher share of carbon-free energies (be it solar, wind, or other renewable modes of electricity generation and nuclear fission) will be needed to carry eventual decarbonization beyond the carbon limits inherent even in a pure CH4 system.
As expected, national experiences have shown a wide range of outcomes, from rapid market penetration in smaller countries with abundant domestic resource of natural gas to still very low shares of the TPES in some of the world’s largest energy consumers heavily dependent on imports. There is no better example of a rapidly executed shift to natural gas than the Dutch exploitation of the supergiant Groningen field. As already noted, the field was discovered in July 1959, and in that year, domestic coal dominated the country’s primary energy supply (about 55%) followed by imported crude oil (about 43%), and natural gas supplied less than 2% of the total supply (UNO [United Nations Organization], 1976).
Groningen field began producing in December 1963, and as its output began to rise, the Dutch government decided in December 1965 (when the natural gas share stood at 5% of primary energy supply) to end all coal extraction in old Limburg fields (going back to the sixteenth century) in no more than 10 years. This deliberate social dislocation (Limburg mines employed 45,000 miners with 30,000 directly related jobs) was eased by giving the state mining company a 40% share in the gas development and helping it to reinvent itself as a producer of chemical goods and later into nutrition, pharmaceutics, and materials (DSM, 2014). By 1971, Groningen gas was providing half of the country’s energy demand, and by 1975, it had stabilized at just short of 50%, while coal (mainly for coking) sank to less than 3%. And because during the early 1970s it was widely believed that nuclear energy will dominate the supply in the long term, it was also decided to maximize exports to the neighboring countries, and they had quadrupled between 1970 and 1980 to more than 40 Gm3 by 1980.
The shift to natural gas was swiftly accomplished as its share in the nation’s TPES rose from less than 5% in 1965 to 33% by 1971. That was an unprecedented speed of substitution: after reaching the 5% mark, it took only less than 6 years to reach 33% (in comparison, the United States took 50 years to go from 5 to 25%, while the USSR needed 20 years to go from 20 to 40%). Extraction (further spurred by the belief that exports should be maximized before nuclear energy, at that time a very promising mode of future energy supply, would weaken the demand) peaked in 1977 at 82.3 Gm3. The supply share during the 1980s and the 1990s leveled off at about 40% as the output became regulated (nuclear electricity did not take over!) in order to extend Groningen’s lifespan. Meanwhile other, smaller, onshore and offshore fields began to augment the field’s production, and by 1990, they supplied more than half of the Dutch gas as Groningen’s output declined from the peak of more than 80 Gm3 to less than 30 Gm3 in the year 2000—but natural gas was still at 40% of the Dutch TPES in the year 2000 and at nearly 37% in 2012 (BP [British Petroleum], 2014a).
Domestic, and international, consequences of the Dutch natural gas extraction have been truly transformative. All space heating (household, commercial, institutional) was converted to Groningen gas, as was nearly all industrial processing and (a very important consideration for a small economy that is the world’s second largest exporter of agricultural products by value) all heating of the country’s extensive greenhouses. They add up to the world’s largest area of heated cultivation (occupying more than 10,000 ha or roughly five times as much as land devoted to potatoes) and produce vegetables (in terms of value dominated by red and yellow peppers), fruits, and flowers, about 40% for export (TNO [Toegepast Natuurwetenschappelijk Onderzoek], 2008).
Moreover, part of CO2 from the clean-burning natural gas is not emitted outside, but it is used to enrich the atmosphere inside greenhouses (to about 1,000 ppm compared to the ambient level of 400 ppm) in order to increase the rate of growth and greenhouse productivity (Hicklenton, 1988; NGMA [National Greenhouse Manufacturers Association], 2014). Exports of the Dutch gas helped the neighboring EU countries to reduce air pollution by displacing coal and fuel oil, to increase combustion efficiency of household heating and industrial processing, and to lessen the dependence on Russian natural gas exports or on even more expensive LNG from the Arab countries—while earning the Netherlands on the order of $10 billion a year (Trading Economics, 2014).
Comparison of the British and the Dutch experience illustrates how national specificities affect the pace of fuel substitutions. Both countries profited from major natural gas discoveries (the first British North Sea gas discoveries were made in 1965), but the Dutch accomplished in six years (going from 5 to 33%) what the United Kingdom took 25 years to do: the North Sea gas began to supply more than 5% of the British TPES in 1971, and it reached 33% only after 26 years in 1997. Then its contribution peaked at 39% in the year 2000 and declined to less than 35% by 2012. Explanations of this disparity are obvious. When the transition began, the British energy was nearly four times larger than the Dutch requirement; the country’s electricity generation was dominated by large coal-fired plants that could not be suddenly shut down; increasing share of electricity was coming from nuclear generation pioneered by British reactor designs; developing offshore resources was more challenging than putting Groningen on stream; and undersea pipelines and longer land lines had to be built to bring the gas to the British market.
Inevitably, the transitions have been even slower when the gas had to be imported from overseas: as already noted, Japan began its
LNG imports in 1969; they reached 5% of the country’s TPES in 1979, 14% by the year 2000, and (following the temporary closure of all nuclear power plants) 22% by 2012. That is a remarkably high share for which the country had to pay by incurring its first prolonged trade deficits since the early 1980s (Trading Economics, 2014). And it is almost certain that the imports will have to increase in the future because in the near term even a vigorous promotion of renewable conversion would not make for the losses of electricity-generating capacity due to the post-Fukushima (March 2011) shut down of Japan’s nuclear power plants.
Recent surge in LNG imports put Japan far ahead of China whose natural gas extraction contributed only about 2% of the country’s TPES until the end of the twentieth century. Subsequent tripling of domestic natural gas extraction and the rise of both pipeline (from Turkmenistan) and LNG imports (mostly from Qatar, Australia, Indonesia, and Malaysia) pushed the gas share to roughly 7.5% by the year 2012. That is the lowest relative contribution among the world’s major economies as even India was slightly ahead in 2012, with nearly 9%. And although large LNG projects (now underway or in planning stages) will further increase the import capacity, the relative contribution of natural gas will continue to rise only slowly as long as the country’s coal expansion will continue: it is now targeted to go to 4.8 Gt by 2020, nearly 40% up from 3.5 Gt in 2013 (Xinhua, 2013).
