Long-Term Energy Technology

Needs and Opportunities for Stabilizing Atmospheric CO2 Concentrations

By Jae Edmonds, technical leader of Economic Programs; Jim Dooley, senior research engineer; and Sonny Kim, senior research scientist, Pacific Northwest National Laboratory.

This paper was prepared for the September 23, 1998, policy conference sponsored by the ACCF Center for Policy Research, and will be published in the Center’s forthcoming book, Climate Change Policy: Practical Strategies to Promote Economic Growth and Environmental Quality.

Executive Summary

Analysis of various greenhouse concentration scenarios and promising technology developments suggests that a portfolio of technologies aimed at carbon mitigation and adaptation will be needed for the twenty-first century. Near-term emissions mitigation can be modest and still meet all but the most stringent caps on atmospheric greenhouse concentrations. Including fuel cells and carbon capture and sequestration technologies among the technology options significantly lowers the cost of emissions mitigation. Some of these technologies such as soil sequestration are currently available; others, however, need more research to reduce their cost and to understand the implications of their deployment on a large scale. Finally, if greenhouse gas concentrations must be stabilized, the recent decline in public and private OECD-nation funding of energy research and development must be reversed to lay a robust foundation for twenty-first-century energy technologies.


The Framework Convention on Climate Change (FCCC; United Nations 1992), ratified by 175 nations, has as its ultimate objective the “stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” (Article 2). However, the treaty and three subsequent conferences of the parties to the Convention artfully dodged several difficult and obvious questions: At what concentration should greenhouse gases be stabilized? How is this to be accomplished? And how much will this cost? These questions are not unrelated, for the answers depend largely on one’s assumptions about the development of future technologies.

It is not surprising that technology plays a large role in constructing the answers. In models, technology can spring into being, fully mature, but real-world technologies are grown from the seeds of research and development (R&D) and must then evolve and compete in a world in which cost and performance matter.

This study addresses two questions important to climate policymakers: What technologies are needed to achieve atmospheric stabilization of greenhouse gas concentrations? And is the existing level of energy technology R&D sufficient to produce these technologies?

However, before turning to technology, it is first necessary to understand the difference between emissions and concentrations, and, second, the unique role carbon dioxide (CO2) plays in any strategy that seeks to use technology to limit either emissions or concentrations.

Stabilizing the Atmosphere

A wide variety of gases, both natural and of human origin, affect the radiative balance of the Earth. CO2 is the most important1 covered by the Kyoto Protocol. Gases such as nitrous oxide and methane exhibit a clear relation between steady-state emissions and steady-state atmospheric concentrations. CO2 does not follow this pattern for two reasons. First, chemical processes in the atmosphere do not remove CO2 in the sense that methane and nitrous oxide are removed. Second, carbon emissions are distributed among three large reservoirs: the atmosphere, oceans, and land. In short, the sum of emissions over a long period of time determines atmospheric CO2 concentration at any future date.

Projecting future concentrations from anthropogenic emissions requires accurate knowledge about all natural carbon flows, and whether these are constant over time. Knowledge about natural flows is evolving. A fixed trajectory cannot be reliably predicted from a particular level of CO2concentrations. Accordingly, the Intergovernmental Panel on Climate Change (IPCC; 1995) and Wigley et al. (1996) have postulated a range of possible trajectories.

In practical terms, modelers must therefore select an emissions trajectory that they believe is credible. This decision, in turn, depends upon the target or ceiling concentration deemed acceptable, the policy instruments available to achieve and maintain the ceiling concentration, and the suite of technologies available to hit the target concentration while providing the energy services needed to sustain the globe’s growing economies. In the current analysis, concentration ceilings ranging from 450 ppmv to 750 ppmv (parts per million by volume) are examined, using Wigley et al. (1996)’s emissions trajectory paths, which approximate cost-effective emission control paths (see Figure 1).

[Figure 1: WRE Emissions Trajectories Consistent With Alternative CO2Concentration Ceilings]


Global Energy Systems With and Without an Emissions Ceiling

Experts are virtually unanimous in their expectation that future technologies will reduce energy intensity and carbon emissions during the twenty-first century. Power generating facilities in the future can be anticipated to be more efficient than at present. Similarly, changes in the composition and efficiency of future activities should provide a source of decreasing energy intensity. While models generally treat technological change as if it materialized by magic, in the real world it is energy R&D that provides the technological catalyst.

However, the relationship between R&D and technological change has proved difficult for researchers to predict.2 The current analysis does not attempt to develop a cause-and-effect representation for R&D and energy technology. Rather, we consider changes in the global energy system that might be anticipated under various policy regimes aimed at stabilizing atmospheric concentrations, and, more importantly, we attempt to identify those areas in the energy system where technical performance will be at a premium.

Coal, Oil, and Gas Production

We start by using the MiniCAM integrated assessment model to explore two future scenarios, factoring in interactions between population, economic activity, energy, agriculture, land use, greenhouse gas emissions, and atmospheric dispositions. The first scenario, Coal Bridge to the Future (CBF), assumes a transition from the present domination of oil and gas to a future dominated by coal (the dominant energy resource globally in terms of reserves). The second, Oil and Gas Forever (OGF), assumes that oil and gas can be economically attractive in the future, continuing their domination as fuels.

The principal difference between the CBF and OGF scenarios is the technology assumed for extracting unconventional oil and gas resources. In the OGF scenario, unconventional oil and gas resources, including methane hydrates (clathrates), are initially available at oil prices of $20 per barrel, with technological change lowering costs at 1 percent per year. In the CBF scenario, these resources are available at $50 per barrel with technological progress proceeding at only 0.5 percent per year. While the same resources are available in all scenarios, differences in technology assumptions determine relative economic performance.

Both of these energy worlds evolve against a background of continued productivity improvement in energy production, transformation, and end-use. Both scenarios are interesting to climate policymakers. The world can either continue its economic development using oil and gas as the core fuels, or it can transition to coal as the core fuel. In both cases, other energy forms such as solar energy play important and even growing roles. However, the backbone of the world’s energy system remains fossil fuels.

Our two reference cases, CBF and OGF, exhibit continued growth in fossil fuel emissions that eventually is viewed as inconsistent with the stabilization of CO2 concentrations. However, because we do not know at what level concentrations will be stabilized, we analyze constraints that result in 450 ppmv, 550 ppmv, 650 ppmv, and 750 ppmv concentrations as well as the reference case of unrestrained carbon accumulation. We compute the cost of achieving each of these concentration objectives under a policy regime that assumes all nations participate, regardless of economic development, and that there may be compensating transfers of income among nations. This strategy minimizes the cost of emissions mitigation at every point in time over a longer period of time. However, it should be noted that the cost of stabilizing carbon concentrations early in the twenty-first century would be significantly greater than doing so gradually. Cost also would be higher if a significant portion of the world did not participate in this effort. Here is a look at how emissions, concentrations, land-use, and other factors essential to the world’s energy system might change and various energy technologies develop.

Commercial Biomass Energy

In the MiniCAM model, modern commercial biomass energy is produced in the context of overall agriculture and land-use management decision making. Land is partitioned between managed and unmanaged components. An expansion of the managed component typically has carbon emissions consequences. Thus the expansion of commercial biomass production implies that land-use patterns must change. Land must be acquired, with some of the land coming through competition with other activities within the managed land system and some coming at the expense of net intrusion into unmanaged ecosystems.

The cost of producing commercial biomass varies with the state of biomass technology, the level of production (land costs depend upon overall demands for land), the technology for producing competing agricultural and sylvacultural products, and the demands for competing agricultural and sylvacultural products.

Electric Power Generation

Fossil fuel electric power generation is assumed to improve continuously between 2000 and 2095. It reaches an efficiency of transformation of 0.67 by 2095. Other forms of power generation also improve. The busbar cost of power declines rapidly between 1997 and 2035 but remains relatively stable thereafter. In 1997 the cost is assumed to be $0.52 per kilowatt hour but has fallen to $0.06 per kilowatt hour by 2035. Further decreases are modest, with the final cost reaching $0.05 per kilowatt hour in 2095.

Nuclear power is assumed to be phased out in Western Europe and North America, but remains an option in the remainder of the world. In regions where it is available, costs are assumed to decline at an average annual rate of 0.5 percent per year. Fusion power is not considered in this exercise, but holds potential for contributing to a low-carbon-emission future. Hydro-electric power is limited by available resources.

The two base or reference scenarios paint different pictures of the global energy system, even though total primary energy and total carbon emissions are similar (see Figure 2). In CBF, the transition from conventional oil and gas to coal is accompanied by an increase in the price of oil and natural gas during the first half of the twenty-first century. This leads to a lower future energy consumption in end-use applications but relatively high primary energy demands and growing carbon emissions.

[Figure 2: Reference Case Energy Emissions for the CBF and OGF Scenarios]

In the OGF scenario, oil and gas prices remain low because usable resources are never exhausted. The lower energy prices imply a higher level of final energy consumption, but the lower carbon content of the primary energy inputs to the system leave carbon emissions similar to those of the CBF scenario. We also note that the lower energy prices imply a smaller contribution by renewable energy forms.

Carbon Emission Reductions

The requirement to reduce carbon emissions (shown in Figure 1) changes energy-use patterns relative to the base case shown in Figure 2. As the CO2concentration constraint tightens from 750 ppmv to 450 ppmv, changes in the global energy system become more pronounced. For example, in the CBF case the scarce resources of conventional oil and gas continue to be used in all but the 450 ppmv ceiling, where a slight decrease in cumulative use is observed. However, coal use is constrained to mitigate carbon emissions.

In the OGF case the scale of change in the energy system is somewhat greater than in the CBF case. More of the world’s energy system is based on oil and gas in OGF than in CBF and less on coal, the changes that occur in the energy system being required to satisfy a carbon concentration ceiling that consequently requires reductions in the scale of both coal and gas production. Renewable, nuclear, and conservation (defined as the reduction in scale of the energy system relative to the reference case) all increase under CO2 concentration ceiling cases. Conservation provides about half of the change in the energy system and reduces the scale of that system considerably.

Note that the bulk of emissions mitigation occurs in the second half of the next century for all scenarios and that no mitigation is required before 2005 (see Table 1). For example, under the 450 ppmv scenario, a cumulative total of 211 billion metric tons of carbon must be removed from the emission stream over the 2005-2050 time period. Note also that the lower the concentration constraint, the higher the cumulative emissions mitigation and the greater the proportion of emissions mitigation that occurs in the first half of the next century. The difference between cumulative emissions mitigation in the first half of the twenty-first century with a 450 ppmv ceiling and a 750 ppmv ceiling is almost an order of magnitude.

Table 1 Cumulative Emission Mitigation Required for Alternative CO2Concentration Constraints Over the Period 2000-2095
(Billions of metric tons of carbon)

Time period & scenario

450 ppmv 550 ppmv 650 ppmv 750 ppmv
2005-2050 CBF 211 87 43 25
2005-2050 OGF 208 84 40 22
2050-2095 CBF 668 480 344 264
2050-2095 OGF 652 464 328 248


Moreover, while the harvesting of commercial biomass would benefit farmers by increasing the demand for crops, this trend would raise food prices, cause unmanaged lands to be brought into production, and increase land-use carbon emissions (relative to the reference case) because the average carbon content of biomass crops is lower than that of unmanaged ecosystems.

The value of carbon increases over time for all the carbon concentration ceilings. Costs start modestly, at about $9 per metric ton (CBF, 750 ppmv) to $105 per metric ton (OGF, 450 ppmv) in 2020, and rise substantially by the end of the century, to about $125 per metric ton (CBF, 750 ppmv) to some $1,179 per metric ton (OGF, 450 ppmv) in 2095 (see Figure 3).

[Figure 3: Value of Carbon for Alternative Atmospheric CO2 Constraints, 2005-2095]

Allowing Carbon Capture and Sequestration

The foregoing discussions assumed that reducing carbon emissions meant reducing use of fossil fuels. This ignores the possibility that carbon from fossil fuels could be captured and sequestered. Introducing the possibility of large-scale carbon capture and sequestration radically changes both the cost and the character of carbon mitigation. Our study considers carbon capture from power plants and the production of hydrogen from fossil fuel feed stocks. Costs for these systems are broken down into energy and capital costs. In addition to the cost of carbon capture there are costs of sequestering and transporting the carbon.

Electric Power Production: The capture of CO2 from the waste stream of a plant requires energy. We assume that the efficiency of carbon capture will increase with time, i.e., new and improved technologies and processes will come on-line that will reduce the energy penalty associated with powering the capture systems. Herzog et al. (1997) state that the eventual integration of these systems into the overall design of new fossil fueled power plants-such as integrated gasification combined cycle power plants-holds forth the promise of reducing the cost of CO2capture significantly. Further, recent research indicates that targeted basic science programs could lead to advancements that over time would improve the performance and reduce the costs of these systems.

Fuel Cell Technology: Fuel cells are a complementary technology to carbon capture and sequestration. They allow hydrogen to be employed as a fuel for transportation as well as to produce heat and power. In the absence of fuel cell technology, the transportation sector depends on batteries and fuels from biomass for emissions mitigation. Fuel cells are assumed to be capable of delivering electricity both to stationary and mobile applications.

Soil Carbon: Over the course of the next 50 to 100 years, according to the IPCC estimates (Cole et al. 1996), between 40 and 80 billion metric tons of fossil fuel carbon emissions might be offset in croplands alone by applying soil carbon sequestration techniques. The estimates for cropland assume the restitution of up to two-thirds of the soil carbon released since the mid-nineteenth century by the conversion of grasslands, wetlands, and forests to agriculture. The experimental record confirms that carbon can actually be returned to soils in such quantities: carbon has been accumulating at rates exceeding one metric ton per hectare per year in former croplands planted to perennial grasses under the Conservation Reserve Program, through which farmers establish permanent vegetative cover on environmentally sensitive cropland.

Managed forests, wetlands, and rangelands provide further opportunity for significant carbon storage. For example, when agriculture is converted (or allowed to revert) to forest vegetation in systems with very little management to improve growth, soil carbon may accumulate at rates ranging from near zero to seven metric tons per hectare per year.

Afforestation and Reforestation: The forestry sector is modeled explicitly in MiniCAM. We have made no attempt to model policies to expand the stock of forest carbon, though the model can easily accommodate an increase in the demand for forests. The increase in the stock of carbon resulting from afforestation and reforestation policies represents a one-time removal of carbon from the atmosphere, and cannot be sustained unless the forests are harvested and the land replanted. In the latter case the problem of removing the harvested products from contact with the atmosphere arises. The problem can be overcome if the harvested materials are used as a commercial biomass feedstock and as a substitute for fossil fuel energy.

Broadening the Technology Options 

Our analysis shows that the introduction of carbon capture and sequestration from central power facilities, and the introduction of hydrogen fuel cells as an option for both power generation and transport, would enable the economy to rely less heavily on carbon-neutral technologies such as commercial biomass harvesting and solar power (which are at an early stage in their technological development) to achieve a particular concentration level. For example, carbon sequestration at power plants and fuel cell use for electric power generation and transportation could cut the present discounted cost of satisfying the 550 ppmv constraint by more than 60 percent. Moreover, if all the potential sequestration options discussed above were combined, costs for keeping under the 550 ppmv ceiling could be reduced more than 70 percent (see Figure 4).

[Figure 4: The Effect of Carbon Capture and Sequestration Technologies on the Cost of Meeting Alternative CO2 Concentration Constraints-CBF]

The costs of alternative emissions concentration ceilings are also highlighted in Figure 4. For example, under the base case, which assumes no sequestration of carbon, lowering the target from 750 ppmv to 650 ppmv doubles the cost, from $266 billion to $529 billion. Similarly, going from a 650 ppmv to 550 ppmv ceiling almost triples the cost, while dropping from 550 ppmv to 450 ppmv more than quadruples the cost of achieving the concentration ceiling. Conversely, going from a ceiling of 450 ppmv to 550 ppmv reduces the costs by a factor of four.

Energy R&D Trends

Substantial investment in energy R&D is justified today, regardless of which concentration ceiling policymakers may eventually select. This R&D investment is needed both to develop new, large-scale technologies that now are not part of the global energy system, and to deploy that technology widely in the coming decades.

However, energy R&D has declined during the past decade in many OECD nations. Various reasons help explain this decline, including changes in productivity of investments, declining energy prices, and changes in institutions and markets. Regardless of the reasons, this downward R&D investment trend-which is present in both the public and private sector-reflects the current assessment of the costs and benefits associated with continued development of climate technologies. This trend cuts across national lines.

For example, U.S. investment in energy R&D has declined substantially over the last decade. Total (public and private) U.S. expenditures for energy R&D declined from $6.6 billion to $4.4 billion (expressed in constant 1995 dollars), a real decline of 34 percent.

Similar reductions can be seen over the period 1985-1995 in public sector support for energy R&D in the key countries known to perform such research (see Figure 5). Approximately 96 percent of the industrialized world’s public sector energy R&D is carried out in only nine countries (number in parentheses shows country’s rank in terms of public sector support for energy R&D in 1995): Canada (6), France (3), Germany (4), Italy (5), Japan (1), the Netherlands (7), Switzerland (8), the United Kingdom (9), and the United States (2) (International Energy Agency 1997). All of these countries except Japan and Switzerland cut their public sector investments in energy R&D over this period.

[Figure 5: Public Sector Support for Energy R&D in Selected Nations, 1985-1995
(Millions of constant 1995 U.S. dollars; percentage indicates real growth from 1985 to 1995)]


We see no evidence that a single energy technology will sufficiently address climate change concerns during the twenty-first century. Instead, a portfolio of technologies that addresses both carbon mitigation and adaptation will be needed.

Moreover, emissions mitigation can remain modest in the near term, unless the goal is to sharply constrain atmospheric concentrations of man-made greenhouse gases. Indeed, the bulk of emissions mitigation in all of the scenarios we examined occurs in the second half of the twenty-first century. This implies that the near-term value of a metric ton of carbon is relatively small, but could rise steadily.

With respect to technology development, we believe that the availability of carbon capture and sequestration technologies on a global scale would significantly reduce the long-term cost of emissions mitigation. These technologies could be an important part of a long-term technology strategy to address climate change. However, it is important to note that they reduce cost by expanding the number of mitigating emission options. Capture and sequestration technologies and fuel cells are likely to meet only a fraction of total mitigation needs. Most reductions will need to come from increased use of nuclear power, renewable energy, and conservation technologies.

The recent decline in energy R&D investment is a major obstacle facing those who would stabilize atmospheric greenhouse gas emissions by the end of the twenty-first century. A robust energy technology foundation will be needed in the decades ahead if the concentration of greenhouse gases in the atmosphere is to be stabilized. 


  1. Other radiatively important gases include methane, nitrous oxide, other nitrogen compounds, carbon monoxide, and sulfate aerosols. Important anthropogenic gases include the chlorofluorocarbons (CFCs), hydrofluorocarbons (HFCs), and perfluorocarbons (PFCs).
  2. Some authors have examined the effect of induced technological change on the timing of emissions mitigation. See, for example, Goulder and Schneider (1998), Nordhaus (1997), and Grubb (1996). This literature breaks induced technological change into two types: learning-by-doing and induced R&D. The general conclusions reached by these researchers is that the presence of induced technological change tends to move mitigation activities from the near-term to the far-term when an optimal global tax policy is implemented (Goulder and Schneider 1998; Nordhaus 1997). The presence of learning-by-doing has an ambiguous effect on the timing of emissions abatement. Whether this effect shifts mitigation to the near-term or far-term depends upon the particular parameterization chosen (Goulder and Schneider 1998; Nordhaus 1997; Grubb 1996).


  • Cole, V., C. Cerri, K. Minami, A. Mosier, N. Rosenberg, and D. Sauerbeck. 1996. Agricultural options for mitigation of greenhouse gas emissions. InClimate Change 1995: Impacts, Adaptations and Mitigation of Climate Change: Report of IPCC Working Group II, Chapter 23, 745-771, eds. R.T. Watson, M.C. Zinyowera, and R.H. Moss. Cambridge: Cambridge University Press.
  • Goulder, L.H. and S. Schneider. 1998. Induced Technological Change, Crowding Out, and the Attractiveness of CO2 Emissions Abatement. Working paper. Stanford University.
  • Grubb, M. 1996. Technologies, Energy Systems, and the Timing of CO2Emissions Abatement: An Overview of Economic Issues. London: Energy and Environmental Programme, Royal Institute of International Affairs.
  • Herzog, H., E. Drake, and E. Adams. 1997. CO2 Capture, Reuse, and Storage Technologies for Mitigating Global Climate Change. Cambridge, Mass.: Energy Laboratory, Massachusetts Institute of Technology.
  • IPCC. 1995. Climate Change 1994: Radiative Forcing of Climate Change and an Evaluation of the IPCC IS92 Emissions Scenarios. Eds. J.T. Houghton, L.G.M. Filho, J. Bruce, H. Lee, B.A. Callander, E. Haites, N. Harris, and K. Maskell. Cambridge: Cambridge University Press.
  • International Energy Agency. 1997. IEA Energy Technology R&D Statistics: 1974-1995. Paris: International Energy Agency.
  • Nordhaus, W.D. 1997. Modeling Induced Innovation in Climate Change Policy. Presented at the National Bureau of Economic Research Summer Institute, PEG Workshop, Snowmass, Colo., 13-14 August 1997.
  • United Nations. 1992. Framework Convention on Climate Change.
  • Wigley, T.M.L., R. Richels, and J.A. Edmonds. 1996. Economic and Environmental Choices in the Stabilization of Atmospheric CO2Concentrations. Nature 379 (6562): 240-243.