IPCC Special Report on Emissions Scenarios

Other reports in this collection

4.4.6. Resource Availability

Section 3.4 in Chapter 3 reviews energy resources and technologies. Here existing reserves (identified quantities recoverable at today's prices and with today's technologies), resources that have yet to be discovered or that need foreseeable techno-economic progress to become available in the future, and other occurrences of hydrocarbons in the Earth's crust are considered. Oil, gas, and uranium occur in deposits that need to be located, and the exploration for new resources is related to the needs for production over the next few decades rather than to a need to define what might ultimately be available for exploitation. Thus the ultimate resource base is uncertain. Coal, on the other hand, occurs in seams over wide areas and very little exploration is needed to give an estimate of potentially available resources. Whether or not they could be mined with given technologies and economics remains the most important uncertainty. Finally, new renewable sources of energy are dependent on ongoing technological development and cost reductions.

The conventional oil industry is relatively mature and the question is at what point in the 21st century will the current reserves start to run out. However, unconventional resources are also available - shale oil, bitumen, and heavy oil. These are starting to be exploited and they will extend current conventional oil reserves. The gas industry is less mature and much more remains to be discovered, particularly in areas that do not currently have the infrastructure to utilize gas and consequently exploration has been unattractive. Additionally, large amounts of unconventional gas have been identified, some of which are already in commercial production (e.g., in the US). Also, huge quantities of natural gas are believed to exist as methane hydrates on the ocean floor (see Chapter 3) and it is possible that technology to exploit these will be developed at some stage. For uranium and thorium, the amount of exploration to date has been very limited, and hence the possibilities of discovering new deposits are enormous. It is likely that even a major expansion of the nuclear industry will not be limited by the amount of available uranium or thorium. With coal, the question is not one of discovery but one of economics, accessibility, and environmental acceptability.

To consider future resource availability as a dynamic process, however, does not resolve the inherent uncertainties in terms of future success rates of hydrocarbon exploration, technology development for either non-conventional fossil resources or non-fossil alternatives, or future energy prices. Therefore, these uncertainties are explored by adopting different scenario assumptions that range from low to (very) high resource availability (see Table 4-4), consistent with the interpretation of the various scenario storylines presented in Section 4.3. This scenario approach is especially important given that hydrocarbon occurrences are the largest storage of carbon. IPCC WGII SAR (Watson et al., 1996) estimates the size of the total carbon "pool" in the form of hydrocarbon occurrences to be up to 25,000 GtC. How much of this eventually could become atmospheric emissions is at present unknown, and depends on the future evolution of technology, prices, and other incentives for future hydrocarbon use and their alternatives.

Given that long-term emission scenarios invariably rely on quantification by formal models, an important distinction needs to be made between assumptions concerning the ultimate resource base and projected actual resource use. Typically, assumptions on the ultimate resource base enter models as exogenously specified constraints - cumulative future production simply cannot exceed values specified as the resource base. Actual resource use, or what is frequently termed the "call on resources" conversely depends on numerous other factors represented in models, such as:

Their complex interplay results in scenarios of future cumulative resource use being the most appropriate indicator, as opposed to exogenously pre-specified resource-base constraints, especially in view of the multi-model approach adopted to develop the SRES scenarios. Table 4-10 and Figures 4-8 to 4-10 summarize the results for the four SRES marker scenarios and of the ensemble of SRES scenarios for their respective scenario families and scenario groups (in the case of the A1 scenario family). It is evident that, in the absence of climate policies, none of the SRES scenarios depicts a premature end to the fossil-fuel age. Invariably, cumulative fossil-fuel use to 2050 (not to mention 2100) exceeds the quantities of fossil fuels extracted since the onset of the Industrial Revolution, even though the "call on" fossil resources differs significantly across the four marker scenarios. This increase is higher in the scenarios that explore a wider domain of uncertainty on future fossil-resource availability.

For non-fossil resources, like uranium and renewable energies, future resource potentials are primarily a function of the assumed rates of technological change, energy prices, and other factors such as safety and risk considerations for nuclear power generation. Generally, absolute resource constraints do not become binding in the marker scenarios or other scenarios. The contribution of these resources is substantially below the physical flows identified in Section 3.4, and therefore results mainly from scenario-specific assumptions concerning technology availability, performance, and costs. These are summarized in Section 4.4.7. A1 Scenarios

Energy resources are taken to be plentiful by assuming a large future availability of coal, unconventional oil, and gas as well as high levels of improvement in the efficiency of energy exploitation technologies, energy conversion technologies, and transport technologies. The grades of energy resources used in the model differ on the basis of extraction costs. When combined with the level of improvement in efficiency of exploitation technology (expressed as the rate of improvement in marginal production costs), the graded costs of energy-resource exploitation determine the energy production costs (prices) and hence the ultimate resource extraction quantities. For A1, large amounts of unconventional oil and natural gas availability were assumed. Cumulative (1990 to 2100) extraction of oil ranges between 15 and 30 ZJ in the A1 scenarios (A1B marker, 17 ZJ); for gas the range is between 23 and 48 ZJ (A1B marker, 36 ZJ) and for coal the range is between 8 and 50 ZJ (A1B marker, 12 ZJ). Resource availability and reliance uncertainties are also explored through additional scenario groups. Three of these (A1C, A1G, and A1T) explore more extreme patterns of reliance on particular resources and technologies compared to the more "balanced" tendencies described in the A1B scenarios, including the A1B marker. As discussed in Chapter 3 and Section 4.3.1, this characteristic of the A1 scenario family stems from the interpretation of technological change and resource availability as being cumulative and path dependent.

Table 4-10:Cumulative hydrocarbon use, historical data from 1800 to 1994 (Nakicenovic et al., 1993, 1996; Rogner, 1997) and range for SRES scenarios (markers and range across all scenarios) for the four scenario families and their scenario groups. The numbers in brackets give minimum and maximum values of scenario variants. Note in particular the large variation within the A1 scenario family as a result of its branching out into four scenario groups, each with a different reliance on particular resource categories and technologies that range from carbon-intensive developments to decarbonization. A1C and A1G have been combined into one fossil-intensive group A1FI in the SPM (see also footnote 1).

World Cumulative Hydrocarbon Use, in ZJ (1,000 EJ)



(12.6-44.4) A1 Scenario Groups

Besides the A1B marker scenario group, alternative pathways unfold within the A1 family, according to diverging technology and resource assumptions (Figures 4-8 to 4-10). Two of these groups (A1C and A1G) were merged into one fossil-intensive group (A1FI) in the SPM. The more detailed information on these two groups is presented here, in Chapter 5 and Appendix VII (see also footnote 1).

The coal-intensive scenario group A1C is restricted mainly to conventional oil and gas, which results in the lowest cumulative oil and gas use (15 to 19 ZJ) of all scenarios; it is even slightly lower than in the B2 scenario, which has much lower energy demand. As such, the scenario illustrates the long-term GHG emission implications of quickly "running out of conventional oil and gas" combined with rapid technological progress in developing coal resources and clean coal winning and conversion technologies. As a result, cumulative coal use is very high - between 48 and 62 ZJ (median, 60 ZJ) between 1990 and 2100.

Conversely, oil and gas resources are assumed to be plentiful in the world of scenario group A1G because of the assumed development of economic extraction methods for unconventional oil and gas, including methane clathrates. Cumulative oil and gas extraction amounts to 76 to 88 ZJ, about twice as high as in the A1C scenario group. Mainly this reflects current perceptions that radical technological change needs to occur to translate a more significant portion of the resource base of unconventional oil and gas into potentially recoverable reserves, a development evidently also cross-checked by possible developments in non-fossil alternatives. Cumulative coal extraction in A1G is relatively low at 15 to 38 ZJ (median, 19 ZJ) across the scenarios of this scenario group.

As a result of fast technological progress in post-fossil alternatives in the technology-dynamic A1T scenario group, the call on oil and gas resources is comparatively modest - cumulative extraction to 2100 ranges between 36 and 46 ZJ, quite similar to the A1C scenario group. The main difference is that because of the improvements in non-fossil alternatives the call on coal resources remains modest - cumulative coal use of 4 to 12 ZJ (median: 10 ZJ) in A1T is the lowest of all the scenarios. A2 Scenarios

Figure 4-8: Cumulative oil resource use 1990 to 2100 in the SRES scenario families, including the four scenario groups within the A1 scenario family. The bars show the spread of total oil extraction over all scenarios in the respective scenario family; the resultant medians and the values of the respective marker scenarios are also shown. A1C and A1G have been combined into one fossil-intensive group A1FI in the SPM (see also footnote 1).

Figure 4-9: Cumulative gas resource use 1990 to 2100 in the SRES scenario families, including the four scenario groups within the A1 scenario family. The bars show the spread of total gas extraction over all scenarios in the respective scenario family; the resultant medians and the values of the respective marker scenarios are also shown. A1C and A1G have been combined into one fossil-intensive group A1FI in the SPM (see also footnote 1).

Figure 4-10: Cumulative coal resource use 1990 to 2100 in the SRES scenario families, including the four scenario groups within the A1 scenario family. The bars show the spread of total coal extraction over all scenarios in the respective scenario family; the resultant medians and the values of the respective marker scenarios are also shown. A1C and A1G have been combined into one fossil-intensive group A1FI in the SPM (see also footnote 1).

Resource availability assumptions for the A2-ASF world are generally rather conservative, essentially that current conventional estimates of petroleum resource availability are not expanded29. Unconventional hydrocarbons, such as methane clathrates and heavy oils, do not come into large-scale use. As a result, coal resource use is the highest among the SRES marker scenarios. The ASF marker scenario quantification of oil, natural gas, and coal resource availability reflects the Rogner (1997) estimates for conventional oil and coal resource availability and the recent IGU (1997) estimates for conventional gas reserves (optimistic scenario, see Chapter 3). Resource extraction costs in the ASF depend on the resource "grade" and vary from US$2.6 to 5.2 per GJ for oil (in 1990 dollars), from US$1.2 to 4.6 per GJ for gas, and US$0.7 to 6.0 per GJ for coal. Harmonized and Other A2 Scenarios

The primary energy structure of the A2 family scenarios is also reflected in the cumulative fossil fuel resource use, characterized by an increasing reliance on coal resources (see Figures 4-8 to 4-10). The cumulative oil use varies by a factor of two across the A2-family, between 11 and 24 ZJ (median, 18 ZJ; A2 marker, 17 ZJ). Cumulative gas use ranges between 20 and ZJ 36 (median, 23 ZJ; A2 marker, 25 ZJ). The higher end of the range of gas resource use occurs in the A2G-IMAGE scenarios, which explored the scenario sensitivity to assuming that a significant fraction of methane hydrate occurrences become technically and economically recoverable in an A2 world. Given the regional orientation of the A2 scenario storyline and the resultant quest for energy independence, the possibility of tapping even currently "exotic" fossil resources certainly merits such a scenario sensitivity analysis. The opposite end of the resource availability spectrum is explored in the MiniCAM scenarios of the A2 scenario family. First, methane clathrates are assumed not to become available. As a result, the call on resources focuses on coal (A2-MiniCAM) or, in a scenario sensitivity analysis, more on unconventional oil and gas (A2-A1-MiniCAM). The range of reliance on coal resources is thus an inverse image of the range of oil and gas resource availability. Cumulative coal extraction varies between 22 and 53 ZJ (median, 35 ZJ; A2 marker, 47 ZJ) across the scenarios of the A2 scenario family. This picture mainly represents what used to be termed "conventional wisdom" in much of the scenario literature (including the previous IS92 scenario series). Importantly, while the probabilities of alternative developments of fossil and non-fossil resource availability cannot be assessed at present, the multi-model, multi-scenario approach described here demonstrates that the uncertainties in fossil resource availability might be much larger than assumed a decade ago. This finding also reflects the results of IPCC WGII SAR (Watson et al., 1996). B1 Scenarios

Assumptions on the fossil fuel resource-base used in the B1 marker scenario quantification are based on the estimates of ultimately recoverable conventional and unconventional fossil resources described in Rogner (1997). The capital output ratio of resource exploitation is assumed to rise with progressive resource depletion, but this is counteracted by learning curve effects in the marker scenario quantification provided by the IMAGE model. Regional estimates of the exploitation costs of conventional and unconventional resources of Rogner (1997) were used to construct long-term supply cost curves as of 1971. These values, rather than absolute upper bounds on resource base availability, define future resource availability in the IMAGE model. The supposed availability of huge non-conventional occurrences of oil and natural gas, with a geographic distribution markedly different from the distribution of conventional oil and gas, has significant implications for fuel supply and trade patterns in the long term. For coal resources, Rogner's (1997) estimates were also adopted; of the total of 262 ZJ, 58 ZJ belong to the categories of proved recoverable, additional recoverable, and additional identified coal resources. The production costs of coal were assumed to rise with increasing depth and rising labor wages, but these costs are largely offset by mechanization (in underground mining) and economies of scale (in surface mining). Harmonized and Other B1 Scenarios

The call on oil resources in the scenarios that comprise the B1 scenario family ranges between 11 and 20 ZJ, with a median of 17 ZJ (B1 marker, 20 ZJ). For gas the range is 15 to 33 ZJ (median, 20 ZJ; B1 marker, 15 ZJ), and for coal the corresponding range is between 3 and 27 ZJ (median, 11 ZJ; B1 marker, 13 ZJ). An overview is given in Figures 4-8 to 4- 10. B2 Scenarios

The availability of fossil energy resources in the B2 marker scenario is assumed to be conservative, in line with the gradual, incremental change philosophy of the B2 scenario storyline. Consequently, oil and gas availability expands only gradually while coal continues to be abundant. Assumed oil and gas resource availability does not extend much beyond current conventional and unconventional reserves. Through gradual improvements in technology, a larger share of unconventional reserves and some additional resource categories are assumed to become available at improved costs over the 21st century. The availability of oil and gas, in particular, is limited compared to the estimated magnitude of global fossil resources and occurrences (Watson et al., 1996). This translates into relatively limited energy options in general and extends also to non-fossil energy options. Harmonized and Other B2 Scenarios

Alternative B2 scenario implementations assumed similar order of magnitudes of resource availability as the B2-marker scenario, except for B2 High-MiniCAM. The resultant cumulative resource use (1990-2100) ranged between 9 and 23 ZJ (median, 17 ZJ; B2 marker, 19 ZJ) for oil, between 18 and 27 ZJ (median, 21 ZJ; B2 marker, 27 ZJ) for gas, and between 12 and 55 ZJ (median, 21 ZJ; B2 marker, 13 ZJ) for coal (see Figures 4-8 to 4-10). The largest uncertainties relate to different interpretations of the more gradual changes under a "dynamics-as-usual" philosophy that characterizes the B2 scenario storyline. One group of scenarios (including the B2 marker) assumed a gradual expansion in the availability of conventional and unconventional oil and gas, whereas another group of scenarios adopted more conservative assumptions (akin to the A2 and B1 scenario families)30. All else being equal, lower resource-availability assumptions for oil and natural gas lead to a higher reliance on coal and non-fossil alternatives and explain, together with technology assumptions, the differences in emissions between alternative B2 scenario quantifications discussed in Chapter 5.

Other reports in this collection