Published as: Moriarty, P. and Honnery, D. (2016). Sustainable Energy Resources: Prospects and Policy. Chapter 1 in M.G. Rasul et al. (Eds) Clean Energy For Sustainable Development, Academic Press/Elsevier, London. Chapter 1 Sustainable energy resources: prospects and policy Patrick Moriarty1* and Damon Honnery2 1 Department of Design, Monash University, Caulfield East, Victoria 3145, Australia. *Corresponding author Email: patrick.moriarty@monash.edu 2 Department of Mechanical and Aerospace Engineering, Monash University-Clayton Campus, Victoria, Australia. Glossary BECCS: bioenergy carbon capture and storage CCS: carbon capture and storage CDM: Clean Development Mechanism CO2: carbon dioxide EdF: Electricit́ de France EIA: Energy Information Administration (US) EJ: exajoule (1018 joule) EROI: energy return on energy invested GHG: greenhouse gas GW: gigawatt (109 W) GDP: Gross Domestic Product HANPP: human appropriation of Net Primary Production IAEA: International Atomic Energy Association IMF: International Monetary Fund IPCC: Intergovernmental Panel on Climate Change ITER: International Thermonuclear Experimental Reactor MWe: megawatt (106 W) electric NRC: National Research Council (US) NOx: oxides of nitrogen NPP: Net Primary Production OECD: Organsation for Economic Cooperation and Development OTEC: ocean thermal energy conversion PV: photovoltaic RE: renewable energy RCP: Representative Concentration Pathways SCC: social cost of carbon SOx: oxides of sulphur SRM: solar radiation management. STEC: solar thermal energy conversion TW: terawatt (1012 W) U-235: uranium isotope 235 W/m2: watt/square metre ZJ: zettajoule (1021 joule) Abstract Global primary energy use is presently dominated by fossil fuels. But given the environmental resource and environmental challenges facing these fuels, alternatives must be sought. This chapter looks at each energy group—fossil fuels, nuclear energy and renewable energy—in turn. We argue that neither carbon sequestration nor geoengineering are feasible solutions for the climate effects of fossil fuels. Nuclear plays only a minor role in world energy production, and official forecasts show little change in market share. Renewable energy also faces a variety of challenges, particularly because the most promising sources, wind and solar energy, are intermittent, and will eventually require expensive energy storage. However, present fossil fuel use is very heavily subsidised, suggesting that removal of these subsidies would greatly reduce energy demand to a level renewable energy could meet. Keywords Carbon sequestration; Earth energy flows; energy selection; energy subsidies; fossil fuels; geoengineering; intermittency; nuclear energy; renewable energy. 1.1 INTRODUCTION According to the analysis of Marchetti [1], over the long term energy sources replace each other in a regular fashion. Thus the millennia-long dominance of biomass ended in the 19th century, and was replaced by coal, which in turn was replaced by oil. But given the problems faced by these fossil fuels, we argue that the world could well see an eventual reversion to renewable energy (RE). But at least for the coming decades, RE must compete with its rival energy sources, fossil fuels and nuclear energy, for share in the global energy market. At present, fossil fuels dominate global energy supply as they have done for over a century, and are only very slowly losing share in global commercial (i.e. excluding fuel wood) energy consumption. The primary energy output in 2014 is shown in Table 1.1 [2]: Table 1.1 Primary commercial energy output, 2014 (EJ). The global values for energy shares (and energy use per capita) conceal large differences between nations. A number of considerations are important in selecting the energy types used in a given region or country:  Local availability of the energy resource. Using locally available energy resources can improve energy security, save foreign exchange, and provide local employment opportunities. For example, bioethanol production in the US and  Brazil is regarded as a means of raising rural incomes and employment. The costs of each energy type, which includes construction, operation and maintenance, fuel costs, and decommissioning. As will be shown below, fossil fuels enjoy massive subsidies, although all three energy groups are subsidised to some extent, either through monetary subsidies of various types, or because external costs are not paid. An obvious and important unpaid external cost is the CO2 emissions from fossil fuel combustion, although the carbon pricing schemes  being introduced in some countries partly address this issue. Environmental considerations. These may range from loss of scenic amenity (as in resident opposition to wind turbines), air pollution from coal-burning power stations, or fears about radiation leakage or reactor accidents in the case of nuclear power. In fact, public opposition has been a major factor in limiting the growth (or  even outright moratoria) for nuclear power in OECD countries in recent decades. Available financial and technical resources. For example, much of Africa’s hydro and geothermal energy potential remains undeveloped because of lack of financial resources. Also many countries do not yet have the technical capacity to start a nuclear power program. A further point is that for some countries the entire national grid may be too small to support even a single nuclear reactor for baseload power. These considerations are often in conflict with each other. Many claim that the large US corn ethanol program, while undoubtedly beneficial to farmers, is not economic. Also for the US, shale gas has helped reduce US energy imports and so improved energy security, but some argue that the economics of shale gas are fragile, and that production will drop dramatically in a few years [3]. Different countries give different weight to these factors, which helps explain why the energy mix varies so widely from country to country. In this chapter we first examine in turn the prospects for the two rival fuel groups to RE. We examine the future difficulties these rival fuels are likely to face in a changing world, including the issues of climate change and possible resource depletion. We then use this analysis as a basis for evaluating the prospects for RE, paying particular attention to which RE types are likely to exhibit the greatest growth in the coming decades. In the final section we look at the policies required to best encourage the needed growth in RE in what may well be an era of continuing financial constraints. We find that removing the vast subsidies to fossil fuels represents the single most effective policy for RE development. 1.2 FOSSIL FUELS Fossil fuels occupy an entrenched position in the global fuel mix, having dominated energy supply for over a century. In 2014, global consumption was 176.3 EJ, 162.5 EJ and 130.9 EJ for oil, coal and gas respectively [2]. A number of energy researchers doubt that this level of fossil fuel energy use can be maintained for much longer. Schindler [4], for example, regarded oil supply as having been on an (undulating) plateau since 2004, with output decline imminent. For coal, he envisaged that ‘global coal production will peak around 2025 at about 30% above the current production rate – this being the upper boundary of the possible development.’ He considered that growth in natural gas can potentially continue for another 5-15 years, also rising to a peak about 30% beyond the current global production rate. Höök & Tang [5] reached similar conclusions, and further argued that global depletion would impose limits on fossil fuel carbon emissions, and hence on their climate change impact. In contrast, McGlade & Ekins [6] thought that ‘globally, a third of oil reserves, half of gas reserves and over 80 per cent of current coal reserves should remain unused from 2010 to 2050 in order to meet the target of 2 °C’. Clearly, for these authors, global climate change concerns, not fossil fuel depletion, is the decisive factor. If such levels of fossil fuel reserves did remain unused, it would have serious and global financial implications. Other authorities also assume few supply constraints on fossil fuel use in the coming decades. The US Energy Information Administration (EIA) [7] regularly publishes forecasts for global energy consumption by fuel type. Apart from a Reference scenario, the EIA scenarios include High and Low Economic Growth cases, and High and Low Oil Price cases. The High and Low Economic Growth cases are the most and least favourable cases for fossil fuel energy use (and energy growth in general). The EIA forecast that coal, natural gas and (all) liquid fuels will still account for between 77.3% and 80.3% in 2040. The latest Intergovernmental Panel on Climate Change (IPCC) reports [8, 9] assumed that carbon sequestration can allow fossil fuels to supply carbon-free or green energy. The integrated assessment modelling by van Vuuren et al [10] examined four Representative Concentration Pathways (RCPs) for the IPCC. The four RCPs were termed 2.6, 4.5, 6 and 8.5, where these numbers refer to the climate forcing (in W/m2) compared with that for the pre-industrial era. Fossil fuels were assumed to supply anywhere between 478 and 1200 EJ by the year 2100, compared with 2014 global consumption of 467 EJ [2, 10]. If the low production estimates of the more pessimistic researchers proves accurate, this would mean that RE would eventually be left with only one major competitor, nuclear power. However, most think that global non-conventional resources of fossil fuels are large (see, for example, [2, 9]). Such non-conventional resources—tar sands, oil shales and various forms of ‘tight’ gas, have much lower energy return on energy invested (EROI) values than conventional fossil fuels, and accordingly have much higher CO2 and other environmental costs, as well as higher economic costs per EJ of energy delivered to the consumer. Such high costs will unfortunately not necessarily prevent them being exploited. Nevertheless, fossil fuels are a finite resource; sooner or later the world will need to shift to alternative energy sources. At present, despite much talk about the need to drastically limit CO2 emissions, consumption of fossil fuels, and with it their CO2 emissions, are still growing [2]. Assuming that the annual supply of fossil fuels is adequate to meet demand over the coming decades, then only concerns about climate change (and, to a lesser extent, regional air pollution) will curb their use. But many see high consumption of fossil fuels continuing even in a carbonconstrained world, because of two possible technological solutions to the CO2 emissions problem, carbon sequestration and geoengineering. 1.2.1 CARBON SEQUESTRATION Two general approaches are possible for carbon sequestration: biological and mechanical. With biological sequestration, the approach is to enhance soil carbon or carbon storage in biomass, particularly by afforestation and reforestation. Since the carbon stored in biomass is estimated to have fallen by around 45% over the past two millennia, such carbon sequestration would merely help restore the status quo ante [11]. Bio-sequestration is also thought to be fairly cheap compared with other carbon mitigation alternatives; Marshall [12] gave a cost of $20-$100 per tonne of CO2 captured, with a potential of around three billion tonnes annually (3 Gt/year) (and as much as 6 Gt/year for bioenergy carbon capture and storage (BECCS). Total fossil fuel carbon emissions in 2014 were 9.7 Gt, or 35.5 Gt of CO2 [2]. Nevertheless, this approach faces two serious problems. The first is the question of how much carbon could potentially be sequestered. Although Marshall suggests that potentially several billion tonnes of carbon could be sequestered annually by the end of this century, other researchers have suggested much lower potentials. Putz & Redford [13] have argued that maximising carbon storage may conflict with biodiversity conservation. Mature forests are better for biodiversity maintenance, but actively managed forests can store more carbon. Smith & Torn [14] regard estimates such as those of Marshall for biosequestration as far too optimistic. They argue than even 1.0 Gt/year of carbon sequestration from combined afforestation and BECCS would represent ‘a major perturbation to land, water, nitrogen, and phosphorous stocks and flows.’ The reason for their pessimism is that only marginal land would be available, given humanity’s already high demands on Earth’s Net Primary Production (NPP). Such land would need major inputs of water and fertiliser for the necessary biomass growth. The second problem concerns the net climate change effects of such tree planting. On the one hand, forest growth in all regions will draw down CO2 from the atmosphere, with positive climate mitigation benefit. But on the other hand, Keller, Feng & Oschlies [15] and Arora & Montenegro [16] have shown that increasing forest area in boreal regions will lower the albedo of such regions, because tree foliage absorbs more insolation than snow-covered ground. Climate forcing (in W/m2) is thereby raised. Reforestation in tropical areas does not lower albedo, hence reforestation—or rather, preventing further deforestation—should be a priority. Arora & Montenegro have claimed that per unit area, tropical afforestation can give three times the warming reductions compared with boreal or temperate region afforestation. Mechanical sequestration— carbon capture and storage (CCS)—can also take two forms. First, capturing CO2 from the exhaust stacks of large fossil fuel power plants or oil refineries, followed by burial in, for example, disused oil and gas fields, or saline aquifers. Second, direct air capture (DAC) of CO2, again followed by burial. DAC is not limited to national emissions, or even emissions from that year, and can be done anywhere, although areas with good wind speeds will help CO2 absorption. It can also usually be carried out closer to CO2 burial sites than is the case for exhaust stack capture. The crucial disadvantage is that it is very energy intensive [17, 18], because of the low CO2 concentration in ambient air. The 2011 report by the American Physical Society [19] summed it up succinctly: ‘In a world that still has centralized sources of carbon emissions, any future deployment that relies on low carbon energy sources for powering DAC would usually be less costeffective than simply using the low-carbon energy to displace those centralized carbon sources. Thus, coherent CO2 mitigation postpones deployment of DAC until large, centralized CO2 sources have been nearly eliminated on a global scale.’ The burial phase of mechanical sequestration is also not without its problems. Zoback & Gorelick [20] concluded that: ‘(…) there is a high probability that earthquakes will be triggered by injection of large volumes of CO2 into the brittle rocks commonly found in continental interiors. Because even small- to moderate-sized earthquakes threaten the seal integrity of CO2 repositories, in this context, large-scale CCS is a risky, and likely unsuccessful, strategy for significantly reducing greenhouse gas emissions.’ A further problem with the long-term integrity of CO2 storage is its potential conflict with ‘fracking’ for natural gas in shale formations. As Elliot & Celia [21] have pointed out, CO2 sequestration needs a deep permeable formation, overlain with an impermeable one to provide a good caprock. Shale formations are ideal for this purpose, and potentially provide a large storage capacity in the US. However, they showed that 80% of this storage capacity overlaps with ‘potential shale-gas production regions’, and the fracturing of the shale for gas extraction would conflict with sequestration. Both these papers have proved controversial, but mainly for the extent of the problems they identify, rather than their existence. 1.2.2 GEOENGINEERING Geoengineering can be defined as the planned modification of the environment on a very large scale, often globally. The idea is not new, but has received recent attention because of the perceived urgent need to avoid dangerous climate change, and has even been cautiously endorsed by the Royal Society in the UK [22]. Advocates have stressed that conventional mitigation methods, such as energy efficiency improvements and greater use of non-fossil fuel energy sources, have not so far stemmed CO2 emissions. The most discussed form of geoengineering would mimic the cooling effects of major volcanic eruptions such as Mount Pinatubo in 1991, and involves placing sulphate aerosols in the lower stratosphere to increase Earth’s albedo. Marine cloud brightening is an alternative strategy for increasing albedo. The albedo represents the Earth-averaged percentage of short-wave insolation reflected directly back into space, and is presently around 30%. By increasing Earth’s albedo, SRM can counteract the global warming resulting from greenhouse gases absorbing and re-emitting long wave radiation back to Earth’s surface. Since the term geoengineering is sometimes taken to include large-scale biosequestration, we will use the more specific term solar radiation management (SRM). SRM would produce a number of benefits. Because the aerosols would be rained out within a year or so, continuous placement of aerosols—perhaps by using civilian airliner flights or military aircraft— would be needed. But this is also an advantage of aerosol placement, since if unanticipated side effects were discovered, the project could be quickly terminated. Further, the temperature reduction benefits would appear within a year, as happened with the Mount Pinatubo natural global cooling. Another important benefit of this approach to SRM is its reported very low (annual) costs [23], especially compared with other climate mitigation methods. SRM also retains the benefits from CO2 fertilisation, which would be lost with CO2 removal policies discussed above. Nevertheless, there are a number of serious problems already recognised with SRM, and perhaps also presently unknown ones. First, emissions of CO2 increase the CO2 content of the oceans as well as that of the atmosphere. The result—which, unlike climate change, is not contested—is steady acidification of the ocean waters. This acidification could inhibit, or at least slow, calcification in a variety of marine organisms, such as coral and foraminifera [24]. Unlike carbon sequestration, SRM would not address this problem. Second, climate forcing from CO2 from fossil fuel combustion and land use change would continue, but would be offset by matching aerosol placement. However, if serious problems appeared with SRM, and the climate forcing offset was terminated, temperatures would subsequently rise rapidly because of the sudden increase in net climate forcing. Ecosystems would have difficulty adjusting to such unprecedented rates of temperature change. Also, the low annual cost of SRM may be more than offset by the need to continue aerosol placement for the lifetime of excess atmospheric CO2, which could take until the year 3000 [25]. Third, with SRM, it may be possible to keep global average surface temperatures at their present (or lower) level, or globally averaged precipitation, but not both [25]. It is possible for example, that the Asian monsoons could be adversely affected, with grave consequences in an already water-stressed world. Fourth, given the problems as well as the benefits of SRM, there may well be intractable problems in obtaining global political consensus for action. Some countries will gain net benefit, others net losses, from a global SRM program, and so it is unlikely that the net losers will agree to SRM. A fifth problem is moral hazard. Because of the advantages of SRM—its low annual cost, its short lead times for implementation, and the ability to rapidly terminate SRM if serious unforeseen consequences arise—it could prove very attractive if other methods fail to avert global climate change. Conventional mitigation, such as the replacement of fossil fuels by RE, will be a slow and expensive process, if only because remaining fossil fuel reserves would be deemed worthless. Also, successful mitigation requires concerted effort by all large emitting countries. But global SRM could be implemented by one group of countries, or even one major country. One response to the problems discussed is to limit SRM to certain regions, and at certain times of the year only, with the aim of mitigating specific climate problems. Thus the proposal to concentrate aerosol placement in the Arctic stratosphere, in order to prevent further loss of sea ice and to arrest Greenland icecap melting. But modelling by Tilmes et al. [26] concluded that: ‘A 4 times stronger local reduction in solar radiation compared to a global experiment is required to preserve summer Arctic sea ice area.’ With the necessary high aerosol concentations needed, the effects would spread well beyod the Arctic. There are also proposals to limit SRM to within national boundaries, such as proposals to paint urban roads and roofs white, cover deserts with reflective material, and even change crop reflectivity, to increase local or regional albedo [22]. But even these more modest measures will either have only minor global cooling potential, or will face serious environmental problems. Given the remaining uncertainties, it would be unwise to rely on this untried technology. As the National Research Council [25] put it: ‘Intervening in the climate system through albedo modification therefore does not constitute an “undoing” of the effects of increased CO2 but rather a potential means of damage reduction that entails novel and partly unknown risks and outcomes.’ Perhaps for this reason the latest IPCC reports do discuss SRM in some detail, but unlike CCS, do not include SRM in any of their future scenarios. 1.2.3 DISCUSSION The view of many official organisations, such as the EIA and the IPCC, is that fossil fuels will continue to dominate global fuel supply for many decades. The implicit assumption is that dwindling reserves (or very high extraction costs) will not limit demand. But as Anderson [27], has stressed, CCS is an untested technology at the very large scale needed, so it would be unwise to base future energy policy around it. At present, only about 10 million tonnes of CO2 are sequestered annually, whereas several billion tonnes would be needed for CCS (or air capture) to be a major mitigation technology. Since CO2 capture from exhaust stacks can only offset around one quarter of all GHG emissions [24], air capture would need to be deployed on a very large scale. Its very high energy costs would thus lead to an even more rapid depletion of fossil fuels. 1.3 NUCLEAR ENERGY The first nuclear energy plants went online in the 1950s decade. After experiencing fairly rapid growth in the two following decades, growth slowed after the Chernobyl accident in 1986. In fact nuclear’s share of the global electricity market peaked in 1996 at 17.6%, and by 2014 had fallen to 10.8% [2], although this share can be expected to rise a little as Japan restarts its reactors in the wake of the 2011 Fukushima accident. Whereas all of the 200+ countries in the world use some fossil fuels, only 31 have nuclear power, mostly countries in the OECD, although China, Russia and India have important nuclear programs. So in most countries nuclear has zero share of the electricity market, but at the other extreme is France, where its share is presently 78.4% [2]. What are the future prospects for nuclear energy? One view is that nuclear power output will be limited by dwindling uranium supplies. Dittmar [28] has argued that even modest growth in nuclear power output will soon be constrained. His forecast, based on historical data from existing and former uranium mines, was that annual global production will soon peak at around 58 kilotons (kt), and that by 2030 will have declined to only about 41 kt. Output will not be enough to sustain even a modest growth rate of nuclear power production of 1% annually, well below the forecast growth rate for global electricity [7]. Other researchers envision either breeder reactors (perhaps using thorium) or even fusion reactors overcoming any possible uranium fuel constraint [29, 30]. World reserves of thorium are thought to be around four times those for uranium [31]. Breeder reactors were early on recognised as necessary to extend limited uranium supplies, as they can convert the fertile isotope uranium-238 (U-238) into fissile plutonium-239, compared with conventional reactors which can only use fissile U-235. (The U-235 isotope only forms 0.7% of naturallyoccurring uranium, with U-238 accounting for nearly all the remainder.) In conventional ‘once-through reactors’, using fuel enriched to around 2-4% U-235, about 99% of the potential energy content goes unused, as the current plan is to bury the spent fuel rods after treatment. However, experience with full-scale breeder reactors have shown that they are difficult to operate. France’s 1200 MWe Superphénix breeder reactor only operated for a decade at low reliability before being permanently shutdown in 1996, and Japan’s Monju reactor, after being shut down from 1995-2010, may not operate again [32]. Hopes for fusion energy are mainly placed in the International Thermonuclear Experimental Reactor (ITER) presently under construction in France, and financed by a multinational consortium. But the date for completion remains uncertain after repeated postponments, and costs have tripled since initial estimates, with further rises likely [33]. And even if successful in its aims, it will still not demonstrate that commercial fusion energy is feasible. Another reason why growth in growth nuclear output will likely be at a low level is that the present reactor fleet is ageing. According to an analysis by Froggatt & Schneider [34]: ‘the unit-weighted average age of the world operating nuclear reactor fleet continues to increase and by mid- 2014 stood at 28.5 years.’ They further add that over 170 of the global 388-strong reactor fleet have run for 30 years or more, and 39 of these for over 40 years. Thus, a substantial reactor building program will soon be needed merely to maintain nuclear power’s present output. Also, worldwide construction costs and construction times appear to be rising. Even official projections for nuclear power do not envision large growth rate increases. The EIA [7] forecast that globally, nuclear output will increase by 2.4% and 2.6% annually between 2010 and 2040 in the low and high economic growth cases, respectively. The International Atomic Energy Association (IAEA) [35], an organisation charged with promoting nuclear energy, have forecast the share of nuclear energy in global electricity production out to 2050. They envisaged this share rising from 12.3% in 2011 to between 12.8 and 13.9% by 2020, but thereafter declining to between 5.0 and 12.2% by 2050. This decline may be in recognition of the ageing reactor fleet discussed earlier. A third forecast, based on the integrated assessment modelling by van Vuuren et al [10] on the four RCPs, showed nuclear power in the year 2100 supplying between 4.1% of global energy in the worst case (RCP 6.0), and 11.3% in the most favourable case (RCP 8.5). The most serious nuclear accidents in recent decades have occurred in the US (Three Mile Island, 1979), the former USSR (Chernobyl, 1986), and Japan (Fukushima, 2011), all technologically sophisticated nations. In each case, the accident had major repercussions for nuclear power worldwide. Given that some technically advanced nations are phasing out their nuclear power programs, major global growth in nuclear energy will necessarily mean programs in countries with lower nuclear expertise and regulation. Nuclear power also must soon face the decades-old problem of waste disposal. The conclusion that can be drawn from this brief survey is that nuclear energy cannot be expected to supply more than its present share of global primary energy, and could supply much less, given widespread public opposition. 1.4 RENEWABLE ENERGY RE sources differ from their main competitor, fossil fuels, in several important ways. The energy from fossil fuels resides in chemical bonds embodied in matter, enabling fossil fuels to be stored above ground, or left underground until needed. On the other hand, RE energy (except for bioenergy and to some extent for geothermal energy) exists only as flows—so RE energy not used is lost forever, and with it the chance to reduce carbon emissions. Also, RE sources, except for hydro, are estimated to have a much lower EROI than present fossil fuels [36]. But RE has important advantages over fossil fuels. As shown earlier, there are two major question marks over the future of fossil fuels. First, there are doubts about how long present, let alone increased production levels, can be maintained at anywhere near present unit costs. Second their combustion produces the long-lived major greenhouse gas, CO2, much of which will remain in the atmosphere for thousands of years [9]. As already shown, technical fixes in the form of carbon sequestration and SRM face several serious problems, probably explaining why neither has been taken up, although discussion on both go back several decades. Finally, the combustion products NOx, SOx, and particulates produce serious air health problems, particularly in the densely populated megacities of the world. While RE sources can also produce GHGs and air pollutants, their emissions output per unit of energy are far smaller than for fossil fuels. At first glance it might appear that rising use of RE on its own will not cut fossil fuel use. Over the past half-century, global RE output has grown in step with fossil fuel use [2]. But when the experience of individual countries is examined, a different picture emerges. A number of European countries, including Germany and the UK, have experienced long-term declines in fossil fuel consumption along with rising RE output. Nevertheless, the various RCP scenarios assumed that, globally, RE output would grow in step with growth in both total energy and fossil fuel energy output, with fossil fuel CO2 emissions greatly reduced through CCS and especially BECCS [10, 37]. 1.4.1 EARTH ENERGY FLOWS AND RE POTENTIAL The energy potentially available to us in the form of RE comes from three sources: the sun, Earth’s interior, and tidal energy (see Figure 1.1). By far the largest is the low wavelength radiation from the sun. As already discussed, about 30% is reflected, unchanged, back into space from our planet. The remainder, about 3.9 million EJ/year, or 3900 ZJ/year (ZJ = zettajoule = 1021 joule), is absorbed by the land, oceans, and clouds. It is this energy which ultimately drives the atmospheric circulation and hydrological systems, and through photosynthesis, plant growth. The energy diverted to atmospheric circulation is subject to a wide range of estimates, but for the entire atmosphere might be 1200 TW, or roughly 38,000 EJ/year [38]. Most of this energy is accounted for by the high-altitude jet stream, and is not available (at least in the foreseeable future) as a human energy source. About one third of Earth’s insolation is diverted to drive the Earth’s hydrological system [39]. However, the power of all the world’s river runoff is a vastly smaller amount, about 3 TW or 95 EJ/year [40]. Figure 1.1 Simplified diagram of annual Earth energy flows. The second energy source is a result of the mutual gravitational attraction between Earth and its much smaller but close-by moon, and to a lesser extent, between Earth and the sun. This tidal energy amounts to about 76 EJ/year. Almost all of this energy is dissipated in the oceans, and fortunately for us, mainly along coastlines. The third energy source is internal to Earth. Geothermal energy is simply the residual heat left over from the violent impacts involved in the formation of our planet around 4.5 billion years ago, together with heat energy derived from the slow radioactive decay of various isotopes of uranium, thorium and potassium both in the Earth’s core and crust. (The 235uranium isotope also presently provides the fuel for nuclear fission energy.) Compared with insolation intensity at the top of the atmosphere of 1366 watt/square metre (W/m2), geothermal output averaged over the Earth’s surface is only about 0.08 W/m2 or about 1300EJ/year. Fortunately, energy flows are much higher near regions of high tectonic activity, such as plate boundaries. In the very long term, all three energy flows are only temporary. The sun, a main sequence star, will expand in volume to become a red giant in a billion years or so, engulfing the Earth. Similarly, tidal energy is slowly decreasing, and will eventually fall to zero when the moon and our planet become locked. Geothermal energy will likewise eventually dwindle away; the primeval heat is slowly being lost from Earth, and the radioactive elements are slowly decaying. But compared to fossil fuels, where the difference in remaining lifetimes between the ‘peak theorists’ and the more optimistic experts is measured in mere decades, these flows can be regarded as permanent. Geothermal energy is a partial exception. Geothermal plants are more economical if the accumulated heat in the field is ‘mined’ at a rate faster than replenishment. Fields can then take several decades to recover. Compared with global commercial primary energy use in 2014 of 541.3 EJ [2], the theoretical availability of the various RE sources, as just discussed, are very high. But a series of constraints limit the actual amount of each RE source that can actually be tapped. The first major constraint is that not all areas of Earth with suitable RE flows can be developed for energy. The deep oceans, the ice caps and high mountain ranges are obviously unsuitable, but some areas may be off-limits for various environmental reasons, while other areas may simply be too distant from energy markets. Together these land constraints limit RE theoretical potential to the geographical potential [38]. Apart from the use of passive solar energy for space heating and cooling, and wind for drying clothes and crops, Earth’s energy flows are not usually used in their crude form. Instead, the natural flows are converted by devices such as photovoltaic (PV) cells or wind turbines into more useful forms of energy, usually electricity. Such conversion is far less than 100% efficient, entailing further energy losses. The energy that can be harvested from the various conversion devices from geographically suitable areas using currently available technology is termed the technical potential. Again, not all technical potential is necessarily economic potential, which de Vries et al. [42] define as: ‘The economic potential is the technical potential up to an estimated production cost of the secondary energy form which is competitive with a specified, locally relevant alternative.’ But as we discuss in detail later in this chapter, large energy subsidies make economics alone a poor guide for selecting energy types. Many researchers have argued that the technical potential for RE is so vast that it will not possible constrain any conceivable global energy use level (see, for example, [43, 44]). However, the published literature on the technical potential on the main RE types in use today and for the foreseeable future—solar, wind, hydro, biomass and geothermal—show a range sometimes spanning two orders of magnitude [45, 46]. Only for hydroelectricity are estimates fairly tightly constrained at around 30-60 EJ. Table 1.2 shows the upper and lower limits for technical potential reported for each of the five leading RE types, as published since the year 2000. For geothermal energy, only the electricity potential is shown, but estimates for the global technical potential for lower temperature heat are very large, with estimates as high as 310,000 EJ. One of the reasons for the large range in global RE potential is that few RE technical potential estimates are based on EROI. The EROI is the ratio of the lifetime output energy to input energy for RE device (for construction, operation and maintenance over the life of the project, and finally decommissioning), both measured in compatible units, for example, primary energy units. The acid test for any new energy project is that the EROI must be greater than 1.0. If it is less, the energy project is an energy sink, not a net addition to energy supply. Only in the development stage of a new energy source can an EROI < 1.0 be tolerated. Table 1.2 Range of global RE technical potential estimates (EJ) The main conclusion from the data in Table 1.2 is that the technical potential for the leading RE sources is not known with any accuracy. However, even the minimum values for RE potential are many times the current RE annual production. Further, apart from the biomass values in Table 1.2, the figures are for electricity, and should be multiplied by 2.6 [2] to better indicate potential in primary energy terms. 1.4.2 PRESENT AND FUTURE USE OF RE Table 1.3 shows the global RE electricity output from various sources for 1990 and 2012 in TWh. For completeness, it also includes the category ‘ocean energy’; at present, nearly all of this is the output of the tidal power station on the Rance Estuary in France. Although the growth in wind and solar energy has been rapid, hydro still dominates RE electricity production, and will for decades to come. What are the prospects for the various RE sources in the coming decades? Table 1.3 Global RE electrical output in 2012 (TWh). Although biomass has only minor electric power output, it dominates RE, with perhaps 50 EJ worldwide, mainly fuel wood burnt at low efficiency in industrialising countries. Along with oil, it is the only energy source that is used in virtually all countries. But although many see a very large technical potential, its future is uncertain because the human appropriation of the Earth’s terrestrial NPP (HANPP) is already very large, with estimates as high as 40% [11]. As human population increases, and with it the demand for food, forage for livestock, fibres (cotton, wool), and timber, HANPP can only rise. Already, given the unprecedented high extinction rates, the natural world is under stress. Adding another heavy levy on NPP, in the form of bioenergy, will add to this stress, and may even be counterproductive, in the sense of decreasing the absolute levels of biomass available for humans [51]. As Table 1.2 shows, even optimistic estimates for geothermal electricity potential are still small, although many times the existing output from the 45 countries that presently operate geothermal electricity plants [52]. The real potential lies in direct use of geothermal heat, which has a very large potential, and is available in many more countries. Hydro is already heavily exploited, with few suitable sites left in OECD countries; most of the remaining potential is in Asia, Africa and South America. Already about 125 countries have hydro plants [52]. Although to a lesser extent than biomass, hydro development can also be at the expense of other ecosystem services [53, 54]. In any case the total technical potential is probably no larger than about 30 EJ. Wind turbines presently operate in around 100 countries, and many more have at least some technical wind potential [52]. Wind turbines are available in several standard sizes, and wind farms can also vary in size, from a single turbine to several hundred. They can thus be utilised by countries with small grids, or even by single households. Another advantage of wind energy is that it is compatible with some existing land uses, such as crops or grazing. Their biggest disadvantage, which they share with solar energy, is the intermittent nature of their output. Ways of overcoming this problem are discussed in the next section. Modern solar energy conversion devices are a relatively recent addition to electricity production (Table 1.3), but output is growing very rapidly. Although solar thermal energy conversion (STEC) is only suitable at utility scale, photovoltaic cells (PV) can be installed as large arrays by utilities with 100 MW or more output, or by individual households, with output measured in kW. Germany is a leader in rooftop PV installations. Together with a storage battery, they are also very useful for off-grid households, such as the majority of tropical African households. So far we have only considered active solar energy, the type delivered by dedicated conversion devices like PV cell arrays or solar hot water heaters. But passive solar energy is an important, if relatively neglected RE source, with high technical potential and low costs. It can be used for heating and cooling buildings, and for lighting. Solar heating and cooling have been used for millennia, and are most effective if they are incorporated into the design of buildings and the selection of building materials. However, some passive solar heating and cooling practices can be backfitted to existing buildings. Similarly, solar lighting can be as simple as drawing the curtains in a room in the morning, but light can also be channeled to illuminate interior rooms with no windows, such as in commercial buildings. It is often difficult to separate out passive solar energy from building energy conservation. In any case, to be effective it requires the active participation of the occupants in opening and closing doors, windows and sun shades at appropriate times. What do official projections see as the future for RE in the coming decades? The four RCPs considered by van Vuuren et al [10] foresaw all RE primary energy growing to between roughly 135 EJ to 335 EJ—but only by the year 2100. Their share of global primary energy would then be between about 16% (RCP6) and 37% (RCP2.6). These low absolute and % values for RE occur because of the major emphasis the RCP scenarios place on CO2 removal through CCS and BECCS. 1.4.3 COPING WITH INTERMITTENT RE SOURCES It is generally agreed that the greatest technical potential for RE lies with solar and wind energy. But these two energy sources, along with wave energy (if ever commercialised), are intermittent sources of energy, which could represent a barrier to their large-scale uptake in electricity grids. Electricity grids have always had to deal with variable demand—for instance, demand is much lower late at night than in the mornings or early evenings. They cope with such fluctuations by having standby power units of known output which can be rapidly brought on line to meet rises in load. But intermittent electricity adds a further uncertainty—this time on the electricity supply side. There are several ways of overcoming such intermittency as wind and solar inevitably assume progressively larger shares of electricity in a given grid. One approach is simply to build overcapacity, so that even during low periods of wind and/or insolation, there is sufficient power to meet demand. But such an approach is wasteful, as electricity will have to be discarded at times of peak intermittent RE output. Another approach tries to avoid this waste by expanding the grid, both to include more non-intermittent electricity sources such as hydroelectricity, but also to help even out the fluctuations from wind/solar by drawing on a wider area. Chatzivasileiadis, Ernst & Andersson [54] have discussed the Desertec proposal, which envisaged bringing both solar and wind energy thousands of km from the deserts of North Africa and the Middle East to Europe. Apart from the cost, another disadvantage with this proposal is the question of energy security for European countries. Seboldt [55] has even proposed a truly global grid spanning both northern and southern hemispheres and the various time zones. This approach would even out seasonal and diurnal insolation fluctuations, but energy security and cost problems would remain. Another method for dealing with intermittency is energy storage. Pickard [56] has argued that an RE electrical grid will need very large storage capacity. California, a leader in RE electricity, has even mandated that the state must have 1.32 GW of storage capacity by the year 2020 [57]. Storage could take the form of pumped hydro storage, compressed air, batteries, and even conversion to another energy carrier such as hydrogen. Some storage of intermittent RE cannot be avoided even with a global grid, because not all final energy demand is for electricity—aircraft and freight ships cannot be feasibly run on electricity. Converting intermittent electricity to hydrogen or methanol to fuel such uses, followed by storage and transport to final users, will also entail high energy costs. Grid expansion and energy storage (or, alternatively, the need for back-up fossil fuel energy) are to some extent substitutes. Steinke, Wolfrum & Hoffmann [58] looked at the trade-offs between the two for a 100% intermittent RE electricity grid in Europe. They found that a European-wide grid expansion could cut the back-up energy needed from 40% to 20% of annual electricity consumption. Only for a truly global grid (or satellite solar power transmitted to Earth receiving stations [55]) could the need for energy storage or back-up power for the electricity grid be completely avoided. Finally, because of the intermittent nature of power output, transmission lines for RE must be of higher capacity that those for fossil fuel power of the same output [59]. 1.5 PROSPECTS AND POLICIES FOR RE Previous sections have characterised the situation for rivals to RE in the global energy market, and argued that the futures for both fossil fuels and nuclear energy are far from certain. We then discussed the technical potential of RE, as well as some of its advantages and disadvantages. In this section, we look at policies which, if adopted, could best aid a more rapid uptake of RE globally. We argue that by far the most important action would be to remove the vast subsidies presently given to fossil fuels. 1.5.1 FOSSIL FUEL ENERGY SUBSIDIES MUST BE CUT As mentioned earlier, all energy sources receive subsidies, sometimes very heavy ones, depending on the country and the fuel type. But present subsidies for alternative fuels pale beside the global subsidies for all fossil fuels as calculated by the International Monetary Fund (IMF) for 2015 [60]. The IMF considers two types of subsidies. The more easily calculated type consists of consumer subsidies, defined as the difference between the international price and the price charged to the consumer; this subsidy the IMF estimate at less than 20% of the total. Most of this subsidy went to consumers in the oil-exporting countries themselves. Darmstadter [61] has pointed out that in Venezuela, motorists were recently paying only 10 cents per gallon (2.6 cents/litre) for petrol. The second, larger, but less well-defined subsidy, consists of negative externalities that energy use inflicts on society and the environment (such as the health effects of air pollution, or the cost of CO2 emissions), and accounts for over 80% of the total. The breakdown of these subsidies are shown in Table 4. The total subsidy amounted to 6.5% of global Gross Domestic Product, even with the large drop in energy prices in 2015. But large as these values are, they may still be underestimates. As an example of a less obvious subsidy that could be considered, Delucchi & Murphy [62] looked at the reduction in just the US military expenditures that would be possible if there were no oil in the Persian Gulf states, and found that roughly $27–$73 billion (2004 values) per year could be saved in the long-run. Table 1.4 Estimated subsidies to fossil fuels, 2015 ($US, 2015). The costs of climate change impacts from fossil fuels may even be under-estimated in the IMF study. The values calculated there for the negative externalities from fossil fuel CO2 emissions (the social cost of carbon (SCC) in US$ per tonne carbon)) used the values from the US government Interagency Working Group on Social Cost [63]. This Working Group estimated that ‘a tonne of carbon dioxide emitted now will cause future harms worth US$37 in today’s dollars’ (Revesz et al. 2014). However, the latter authors and others (e.g. [65-67]) have criticised such values as being far too low. Indeed, Ackerman & Stanton [65] have argued that values of SCC of $1000 or even much higher could easily be justified. If this were the case, SCC would dominate the total cost of negative externalities, and total subsidies overall would rise steeply. Subsidies to nuclear and RE were not considered in the IMF study. Nuclear energy subsidies were fairly small in 2015. However, in the past, nuclear power received very large subsidies, which were vital for its commercial introduction. According to a study by Ward [68] ‘it is estimated that the USA nuclear energy sector received financial support to the tune of $15.3 per kWh in the first 15 years of its development (1947-1961), compared to wind energy which received just $0.46 per kWh in its first 15 years.’ US support for the nuclear energy was thus around 30 times the rate for wind energy. One possible defence of consumer energy subsidies is that it promotes equity. But as Edenhofer [69] has stressed: ‘Energy subsidies are typically captured by rich households in low income countries and do little to support the poor.’ Similarly for the indirect subsidies: climate change impacts of fossil fuel use will heavily impact on low-income communities [70], and urban air pollution, largely from fossil fuel combustion, is likewise more serious in lower-income areas [71]. The most important point about these huge energy subsidies is that they lead to massive overuse of energy in general. If consumer prices were not subsidised, and the full health and environmental (eg global climate change) costs were met by users, global energy use would be far smaller. In other words, it does not particularly matter if RE cannot meet present (or future projected) global energy levels. RE will in future be the major energy source (in terms of energy share) partly by absolute growth in RE output, but also by a rapid reduction in fossil fuel use. The most important energy policy—and one that would be of the largest benefit to RE—would be the removal of fossil fuel subsidies. 1.5.2 POLICIES NEEDED TO SUPPORT RE Unruh [72] has introduced the notion of carbon ‘lock-in’, the idea that, as he puts it: ‘industrial economies have become locked into fossil fuel-based energy and transportation systems through path-dependent processes driven by technological and institutional increasing returns to scale.’ A more recent insight on carbon lock-in is provided by Davis, Caldeira & Matthews [73]. They calculated the cumulative CO2 emissions emitted if all fossil fuel power stations and vehicles operating in 2010 continued in use until their economic lives ended. Their estimates centered on roughly 500 Gt of CO2, about 14 times the 35.5 Gt emitted by all fossil fuels in 2014 [2]. At the least, drastic reductions in fossil fuel use would entail the premature retirement of much of the fossil fuel power station and vehicle fleets. Fossil fuel energy subsidies will probably not be heavily reduced any time soon, given not only the carbon lock-in just discussed, but also the vested interests and economic power of fossil fuel producers. Another factor is the popularity of direct fuel subsidies to consumers, such as low petrol prices for motorists. We have already discussed the high subsidies for nuclear power in its early years. Promotion of RE will inevitably require some subsidies if it is to become the dominant energy provider. Koseoglu et al. [74] have grouped the vast number of the various specific policies possible into two general approaches:  market instruments for encouraging greater RE use, the approach favoured by Germany and many other countries;  emphasis on, and support for, R&D for RE, the approach favoured in California and some other US states. However, most countries use a mix of policies to support RE growth, and further, vary in the level of direct government control. Germany, a leader in RE, has relied heavily on feed-in tariffs, which vary for different forms of RE. They are much higher for off-shore than for on-shore wind electricity, reflecting its higher costs. Support for PV electricity was in turn nearly five times higher in 2009 than for on-shore wind, perhaps because insolation levels in Germany are relatively low. The German experience is considered as effective in supporting rapid RE growth, but not very cost-effective. There is also the danger of ‘technological lock-in’ with technology-specific tariffs. As an alternative to supporting RE market applications, as in Germany, governments can also subsidise R&D for RE. Koseoglu et al. found that supporting R&D was a better use of scarce resources for immature (and rapidly developing) RE technologies. For more mature technologies such as hydro, however, this approach will be of limited effectiveness. In China, and to a lesser extent in other lower income countries, the Clean Development Mechanism (CDM) has been important in encouraging the growth of RE. The CDM allows these countries to offset the CO2 emissions from industrial countries. It has been criticised for being relatively ineffective, insofar as in many cases the low-carbon projects (such as hydroelectricity projects) would have been undertaken in any case. Others like Newell [75], have gone much further, and argued that market mechanisms such as carbon markets will only have limited success in reducing carbon emissions. There are evidently advantages and disadvantages with each of the approaches used, and no global recommendation is possible; it can and should vary from country to country, depending, among other factors, on the particular RE source involved and level of technological development in the country. PV cell arrays and solar water heaters have been installed on the roofs of millions of private residences worldwide. Similarly geothermal heat pumps are rapidly expanding in use in many countries. But private residences cannot install a hydro, STEC, or geothermal plant. Different support is needed for households compared with utility scale plants. The support needed for bioenergy is likewise very different in low-income compared with high-income countries. Firewood and crop residues are the cheapest fuel in poor countries, which explains their very high level of use, even if this level of use is environmentally unsustainable. Incentives are needed, not to encourage but to reduce this form of bioenergy use, through the widespread use of more fuel-efficient cooking stoves 1.5.3 WHICH SOURCES OF RE SHOULD BE SUPPORTED? There are a very large number of possible sources of RE, apart from the five most commonly used. These include   ocean thermal energy conversion (OTEC)  wave energy  higher altitude wind energy collected from turbines on air balloons or even kites  ocean currents  tidal energy osmotic energy at the fresh water/salt water interface at river estuaries. The question arises as to whether to support these RE sources in hopes of a breakthrough, or to avoid spreading limited R&D resources too thinly, and concentrate on improving the mainstream RE types. Sometimes, novel energy sources can be ruled out, at least as major energy sources, by using energy analysis. For example, OTEC will have a thermal efficiency of only 3-4% [76], given that the temperature difference between tropical surface waters and ocean depths is at most around 20 °C, and a 1000m pipe will be needed to draw up deep colder water as a heat sink. It is possible small-scale OTEC plants, with distilled water as a co-product, could be feasible at coastal locations on small islands. However, for large scale energy production, the OTEC plants would need to move over the ocean to maintain a temperature differential, and the electricity generated converted into an energy carrier such as ammonia or hydrogen, stored, and periodically shipped to shore. When these conversion, storage and transport energy costs are considered along with the low Carnot efficiency, it is doubtful than any net energy would result. However, it is possible to select which RE types should be generally supported. Solar energy is the obvious candidate, for two important reasons. First it has by far the highest technical potential of any RE source. Second, there are likely to be further technical breakthroughs, not only leading to reduced production costs for existing PV cell types, but also to novel PV materials. As for other new RE technologies, it will be important to choose materials (as well as installation sites) that do not compromise environmental sustainability in general [46, 77, 78]. For RE to be a major force in energy production, the intermittent RE sources, wind, solar and perhaps wave energy, will need to supply most RE. But as, we have seen, this will inevitably mean than large amounts of energy storage will be needed. Conversion of intermittent RE electricity to an alternative energy carrier such as hydrogen or methanol, then storage, followed by re-conversion to electricity, will greatly reduce the net energy output from these sources, and hence lower the EROI and raise costs. Other approaches are of course possible, such as pumped hydro and battery storage, are further possibilities if the output is also to be electricity. The scope for further conventional pumped storage is low [56]. A large variety of battery types are being investigated, with lithium batteries the most popular for consumer electronics and electric vehicles. Lemmon [57] has argued that ‘(…) batteries cannot provide rapid (less than a second) high-power responses and supply energy for long periods. Batteries degrade and are expensive to replace.’ Instead, he argued for new types of fuel cells which ‘can be modified to store energy and produce liquid fuels such as methanol, thanks to breakthroughs in materials and designs.’ But Soloveichik [79] in a recent review, has argued that flow batteries can overcome many of the problems of other batteries, such as limited life, even if they are not suitable for mobile applications. As with different RE sources, it is likely that different applications will need different energy storage systems. 1.6 DISCUSSION As output of RE rises to make it the dominant energy source, it may prove necessary to move away from the fossil fuel-era idea of energy available in any desired amount, at any time, at any place. Demand management of energy is not new: for many decades off-peak energy (usually at night) has been cheaper than peak rates. With the advances in information technology, it is now possible to price electricity according to instantaneous supply. Further, some load-shedding is possible at times of low supply without causing disruption to domestic users. For example, freezers, refrigerators and hot water systems can be turned off for limited periods without any ill-effects. But we may need to go well beyond this, and change policies not directly connected to the energy sector. Future RE potential may well be located at sites remote from existing grids. One possibility is to move some industry to these locations, as is already done with aluminium smelters, sometimes being located near cheap hydropower. Population growth could also be encouraged in these areas. This shift would be especially valuable for lowtemperature geothermal heat energy, which has a very large global potential. The problem is that it is neither energetically nor economically feasible to pipe such heat more than a few kilometres. In the US at least, such geothermal heat (mainly in the western Rocky Mountain states) is poorly matched to population [80]. The climate change problem will need to be decisively tackled in the coming decade or two. We have argued that attempts at ‘greening’ fossil fuels through CO2 capture are likely to be marginal, or in the case of geoengineering, unlikely to be attempted at all. There will probably be a large mismatch between the rate at which fossil fuel use will have to be reduced, and the maximum rate at which RE can be introduced as replacement energy. At present, except for solar energy, which is still growing exponentially from a small base, all other RE sources are at best growing only linearly [2]. RE sources will likely only come to dominate the future energy mix if absolute global energy levels are greatly reduced. REFERENCES [1] [2] [3] [4] [5 [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] 18] Marchetti C. On energy systems historically and in the next centuries. Global Bioethics 2009; 22(1-4): 53-65. BP. BP statistical review of world energy 2015: BP London, 2015. Inman M. The fracking fallacy. Nature 2014; 56:28-30. Schindler J. The availability of fossil energy resources’ in Factor X: policy, strategies and instruments for a sustainable resource use, eds M Angrick et al, Eco-Efficiency in Industry and Science 29: Springer Dordrecht NL, 2014, pp. 19-38. Höök M, Tang X. Depletion of fossil fuels and anthropogenic climate change—a review. Energy Policy 2013; 52: 797–809. McGlade C, Ekins P. The geographical distribution of fossil fuels unused when limiting global warming to 2 °C. Nature 2015; 517: 187-190. Energy Information Administration (EIA). International energy outlook 2013: US Dept. of Energy, 2013. (http://www.eia.gov/forecasts/archive/ieo13/pdf/0484(2013).pdf). Intergovernmental Panel on Climate Change (IPCC). Climate change 2014: mitigation of climate change: Cambridge University Press Cambridge UK, 2014. Intergovernmental Panel on Climate Change (IPCC). Climate change 2014: synthesis report: Cambridge University Press Cambridge UK, 2015. van Vuuren DP, Edmonds J, Kainuma M, Riahi K, Thomson A, Hibbard K et al. The representative concentration pathways: an overview. Climatic Change 2011; 109: 5–31. Schramski JR, Gattie DK, Brown JH. Human domination of the biosphere: rapid discharge of the earthspace battery foretells the future of humankind. Proceedings of the National Academy of Sciences 2015. (doi: 10.1073/pnas.1508353112). Marshall M. Transforming earth. New Scientist 2013; 12 October: 10-11. Putz FE, Redford KH. Dangers of carbon-based conservation. Global Environmental Change 2009; 19: 400–401. Smith LJ, Torn MS. Ecological limits to terrestrial biological carbon dioxide removal. Climatic Change 2013; 118: 89–103. Keller DP, Feng EY, Oschlies A. Potential climate engineering effectiveness and side effects during a high carbon dioxide-emission scenario. Nature Communications 2014; 5(3304). (doi: 10.1038/ncomms4304). Arora VK, Montenegro A. Small temperature benefits provided by realistic afforestation efforts. Nature Geoscience 2011; 4: 514-518. House KZ, Baclig AC, Ranjanc M, Van Nierop EA, Wilcox J, Herzog, HJ. Economic and energetic analysis of capturing CO2 from ambient air. Proceedings of the National Academy of Sciences 2011; 108(51): 20428-20433. Moriarty P, Honnery D. A human needs approach to reducing atmospheric carbon. Energy Policy 2010; 38: 695–700. [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] 30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] American Physical Society (APS). Direct air capture of CO2 with chemicals, 2011. (http://www.aps.org/policy/reports/assessments/upload/dac2011.pdf). Zoback MD, Gorelick SM. Earthquake triggering and large-scale geologic storage of carbon dioxide. Proceedings of the National Academy of Sciences 2012; 109(26): 10164–10168. Elliot TR, Celia, MA. Potential restrictions for CO2 sequestration sites due to shale and tight gas production. Environmental Science and Technology 2012; 46:4223-4227. Royal Society. Geoengineering the climate: science, governance and uncertainty: Royal Society London, 2009. McClellan J, Keith DW, Apt, J. Cost analysis of stratospheric Albedo modification delivery systems. Environmental Research Letters 2012; 7(034019): 8pp. Moriarty P, Honnery D. Rise and fall of the carbon civilisation: Springer London, 2011. National Research Council (NRC). Climate intervention: reflecting sunlight to cool earth: National Academies Press Washington DC, 2015. Tilmes S, Jahn A, Kay JE, Holland M, Lamarque J-F. Can regional climate engineering save the summer Arctic sea ice? Geophysical Research Letters 2014; 41: 880–885 (doi:10.1002/2013GL058731). Anderson K. Duality in climate science. Nature Geoscience 2015 (doi:10.1038/ngeo2559). Dittmar M. The end of cheap uranium. Science of the Total Environment 2013; 461–462: 92–798. Moriarty P, Honnery D. Intermittent renewable energy: the only future source of hydrogen? International Journal of Hydrogen Energy 2007; 32: 1616–1624. Moriarty P, Honnery D. What energy levels can the Earth sustain? Energy Policy 2009; 37: 2469-2474. Bagla P. Thorium seen as nuclear’s new frontier. Science 2015; 350: 726-727. Dittmar M. Nuclear energy: status and future limitations Energy 2012; 37: 35-40. Clery D. New review slams fusion project’s management. Science 2014; 343: 957-958. Froggatt A, Schneider M. Nuclear power versus renewable energy: a trend analysis. Proceedings of the IEEE 2015; 103(4): 487-490. International Atomic Energy Association (IAEA). Energy, electricity and nuclear power estimates for the period up to 2050: IAEA Vienna, 2012. Hall CAS, Lambert JG, Balogh SB. EROI of different fuels and the implications for society. Energy Policy 2014; 64: 141–152. van Vuuren DP, Stehfest E, Elzen MG, Kram T, Vliet JV, Deetman S. et al. RCP2.6: exploring the possibility to keep global mean temperature increase below 2 °C. Climatic Change 2011; 109: 95–116. de Castro C, Mediavilla M, Miguel LJ, Frechoso F. Global solar electric potential: A review of their technical and sustainable limits. Renewable & Sustainable Energy Reviews 2013; 28: 824–835. Hafele W. Energy in a finite world: a global systems analysis: Ballinger Cambridge MA, 1981. Makarieva AM, Gorshkov VG, Li B-L. Energy budget of the biosphere and civilization: rethinking environmental security of global renewable and nonrenewable resources. Ecological Complexity 2008; 5: 281–288. Trenberth KE. An imperative for climate change planning: tracking Earth’s global energy. Current Opinion in Environmental Sustainability 2009; 1: 19–27. de Vries BJM, van Vuuren DP, Hoogwijk MM. Renewable energy sources: their global potential for the first-half of the 21st century at a global level: an integrated approach. Energy Policy 2007; 35: 2590– 2610. Lu X, McElroy MB, Kiviluoma J. Global potential for wind-generated electricity. Proceedings of the National Academy of Sciences 2009; 106: 10933–10938. Jacobson MZ, Delucchi MA. Providing all global energy with wind, water and solar power, part 1: technologies, energy resources, quantities and areas of infrastructure, and materials. Energy Policy 2011; 39: 1154–1169. Moriarty P, Honnery D. What is the global potential for renewable energy? Renewable & Sustainable Energy Reviews 2012; 16: 244–252. Moriarty P, Honnery D. Can renewable energy power the future? Energy Policy 2016; 93: 3-7 (doi: 10.1016/j.enpol.2016.02.051). Field CB, Campbell JE, Lobell DB. Biomass energy: the scale of the potential resource. Trends in Ecology & Evolution 2008; 23(2): 65-72. de Castro C, Mediavilla M, Miguel LJ, Frechoso F. Global wind power potential: physical and technological limits. Energy Policy 2011; 39: 6677–6682. Moriarty P, Honnery D. Preparing for a low-energy future. Futures 2012; 44: 883–892. Electricit́ de France (EdF). Worldwide electricity production from renewable energy sources, Fifteenth Inventory, 2013. (http://www.energies-renouvelables.org/observ-er/html/inventaire/pdf/15e-inventaireChap01-Eng.pdf). [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] 64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] Kleidon A. The climate sensitivity to human appropriation of vegetation productivity and its thermodynamic characterization. Global and Planetary Change 2006; 54: 109–127. World Energy Council (WEC). World energy resources: 2013 survey: WEC London, 2013. (Also earlier surveys.) Moriarty P, Honnery D. Is there an optimum level for renewable energy? Energy Policy 2011; 39: 27482753. Chatzivasileiadis S, Ernst D, Andersson G. The global grid. Renewable Energy 2013; 57: 372-383. Seboldt W. Space- and Earth-based solar power for the growing energy needs of future generations. Acta Astronomica 2004; 55: 389–399. Pickard WF. Smart grids versus the Achilles’ Heel of renewable energy: Can the needed storage infrastructure be constructed before the fossil fuel runs out? Proceedings of the IEEE 2014; 102(7): 10941105. Lemmon JP. Reimagine fuel cells. Nature 2015; 525: 447-449. Steinke F, Wolfrum P, Hoffmann C. Grid vs. storage in a 100% renewable Europe. Renewable Energy 2013; 50: 826-832. Buijs P, Bekaert D, Cole S, van Hertem D, Belmans R. Transmission investment problems in Europe: going beyond standard solutions. Energy Policy 2011; 39: 1794–1801. Coady D, Parry I, Sears S, Shang B. How large are global energy subsidies? IMF Working Paper WP/15/105, 2015. Available at (https://www.imf.org/external/pubs/ft/wp/2015/wp15105.pdf). Darmstadter J. The International Monetary Fund confronts global energy subsidies. Environment: Science and Policy for Sustainable Development 2013; 55(5): 41-44. Delucchi MA, Murphy JJ. US military expenditures to protect the use of Persian Gulf oil for motor vehicles. Energy Policy 2008; 36: 2253– 2264. Interagency Working Group on Social Cost of Carbon. Technical update of the social cost of carbon for regulatory impact analysis under executive order 12866, US Government, Washington DC, 2013. Available at (http://go.nature.com/vzpkkb). Revesz RL, Howard PH, Arrow K, Goulder LH, Kopp RE, Livermore MA et al. Improve economic models of climate change. Nature 2014; 508: 173-175. Ackerman F, Stanton EA. Climate risks and carbon prices: revising the social cost of carbon. Economics 2012; 6(10) (http://dx.doi.org/10.5018/economics-ejournal.ja.2012-10). Aldy JE. Pricing climate risk mitigation. Nature Climate Change 2015; 5: 396-398. Moore FC, Diaz DB. Temperature impacts on economic growth warrant stringent mitigation policy. Nature Climate Change 2015; 5: 127-131. Ward P. Unfair aid: the subsidies keeping nuclear energy afloat across the globe. Nuclear Monitor 2005; 630-631. (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.363.641&rep=rep1&type=pdf). Edenhofer O. King Coal and the queen of subsidies. Science 2015; 349(6254): 1286-1287. Moriarty P, Honnery D. Future cities in a warming world. Futures 2015; 66: 45-53. Hajat A, Hsia C, O’Neill MS. Socioeconomic disparities and air pollution exposure: a global review. Current Environmental Health Reports 2015; 2: 440–450. Unruh GC. Escaping carbon lock-in. Energy Policy 2002; 30: 317–325. Davis SJ, Caldeira K, Matthews HD. Future CO2 emissions and climate change from existing energy infrastructure. Science 2010; 329: 1330–1333. Koseoglu NM, van den Bergh JCJM, Lacerda JS. Allocating subsidies to R&D or to market applications of renewable energy? Balance and geographical relevance. Energy for Sustainable Development 2013; 17: 536–545. Newell P. The political economy of carbon markets: the CDM and other stories. Climate Policy 2012; 12: 135–139. Buigues G, Zamora I, Mazón AJ, Valverde V, Pérez FJ. Sea energy conversion: problems and possibilities 2006: 8 pp. (http://icrepq.com/icrepq06/242-buigues.pdf). Vidal O, Goffé B, Arndt N. Metals for a low-carbon society. Nature Geoscience 2013; 6: 894-896. Moriarty P, Honnery D. Reliance on technical solutions to environmental problems: caution is needed. Environmental Science & Technology 2015; 49: 5255−5256. Soloveichik GL. Flow batteries: current status and trends. Chemical Reviews 2015; 115: 11533−11558. Lienau PJ, Ross H. Final report: low-temperature resource assessment program, 1996. (http://geoheat.oit.edu/pdf/db7.pdf). Figure 1.1 Simplified diagram of annual Earth energy flows Sources: [24, 41]. Table 1.1 Primary commercial energy output, 2014 (EJ). Energy source Energy output % of total (EJ) energy Fossil fuels 467.2 86.3 Nuclear energy 24.0 4.4 All RE sources 50.1 9.3 Source: [2] Note: EJ = exajoule = 1018J. Table 1.2 Range of global RE technical potential estimates (EJ) RE source Technical potential range (EJ) Biomass 27-1500 Geothermal1 1.1-22 Hydro1 Solar 1 Wind1 1 19-95 63.0-15,500 31.5-3000 electric output Sources: [38, 43, 45, 48]. Table 1.3 Global RE electrical output in 2012 (TWh). RE source Elec output (TWh) 1990 2012 Biomass 88.9 326.2 Geothermal 36.1 70.4 Hydro 2163.3 3663.4 Solar 0.4 104.5 Wind 3.6 534.3 Ocean 0.5 0.5 All RE 2292.8 4447.5 sources Sources: [2, 49, 50]. Table 1.4 Estimated subsidies to fossil fuels, 2015 ($US, 2015). Energy type Subsidy US$ % of total billions subsidy coal 3147 59.4 oil 1497 28.2 natural gas 510 9.6 electricity 148 2.8 all energy 5302 100 Source: [60].