What can be done to slow global warming? Huge new sources of carbon-free power may be needed. But other options also exist, and with so many uncertainties dogging predictions of technology and climate, choosing the best portfolio is hard.
The Earth is getting warmer, probably as a side-effect of human industry. The main culprit is carbon dioxide (CO2), a by-product of burning fossil fuels for energy. On page 881 of this issue1, Hoffert et al. examine one plausible scenario for future energy demand. They conclude that to stabilize the concentration of CO2 in the atmosphere at twice pre-industrial levels will require a vast increase in the supply of carbon-free energy sources such as solar, wind and nuclear power.
In order to explore where policy-makers have leverage, Hoffert et al. express CO2 emissions as the product of four variables: CO2 released per unit of energy; energy consumed per unit of economic output; economic output per person; and the number of people.
There are good reasons not to meddle with the last two variables. Ideally, policy should limit CO2 concentrations while doing minimum harm to the economy. Nor can politicians do much more to curb world population, which will probably stabilize at about 11 billion people2. Many countries may even attempt to increase their numbers in order to reverse the strains on productivity and social security that result when populations shrink and grey.
That leaves policy-makers two options: 'decarbonize' the energy system by lowering the quantity of carbon emitted per unit of energy3, or improve energy efficiency (cutting the primary energy needed per unit of economic output).
Both of these numbers have improved over the past 150 years. On average, worldwide, the energy consumed per unit of economic output has declined at about 1% per year4. Hoffert et al. show that, if that rate remains constant, then stabilizing atmospheric CO2 at 550 parts per million will require about 15 terawatts (TW) of carbon-free power by 2050, and even more thereafter. For comparison, today the world consumes about 13.5 TW of power altogether, of which no more than 3.4 TW are carbon free (Fig. 1). Creating so much carbon-free power would require a decrease in the carbon emission per energy unit of about 1% per year. The historical rate has been a more leisurely 0.3% per year3. The gap, conclude the authors, is huge.
" Figure1 Global energy consumption in 1996, totalling 13.5 terawatt hours.About 25% is carbon free. Of the rest, coal releases 790 million tonnes of carbon per terawatt hour; oil 610; gas 470. "Figure 2 Three ways to curb carbon: nuclear fusion (top left), a carbon-free technology that is still not commercially viable; tree plantations (top right), which can remove carbon from the atmosphere or provide fuel wood with practically no net carbon emission; and energy-efficient light bulbs. Deciding on which are best depends on many factors, including expected future costs
This arithmetic is hardly controversial, but it leaves boulders unturned. Most importantly, Hoffert et al. omit economics from the analysis and so cannot explore the trade-offs that real policy must confront -- such as the balance between carbon-free power and improved energy efficiency. For example, if policies to improve energy efficiency could accelerate the 1% annual decline in energy use per unit of economic output to 1.5%, then the carbon-free energy required would be cut in half (see Fig. 3 on page 883), and the urgency of investing in new power sources would wane. The energy system is full of fat: the useful energy output of many devices, such as illumination or locomotion, is only a few per cent of total energy input5. So large savings are quite possible.
Hoffert et al. may overstate the need for carbon-free power for another reason. Their baseline is the 1992 "a" scenario published by the Intergovernmental Panel on Climate Change, which assumes that coal, the most carbon-intensive fossil fuel, remains one of the main energy sources. So carbon emissions would be high, and stabilizing atmospheric CO2 requires a lot of carbon-free power. But it may well be that even without a deliberate policy to slow global warming, energy production comes to be dominated by less carbonaceous natural gas, or by carbon-free nuclear and biomass fuels6, yielding much lower carbon emissions and less need to supply still more carbon-free power.
Politicians should also consider options for withdrawing carbon from the atmosphere by planting trees, improving soil management and disposing of CO2 captured from the smokestacks of power plants before it is released to the atmosphere. That would lower the need for carbon-free power and is also likely to be a cost-effective option, especially during the next few decades, before radically new energy systems are available (Fig. 2).
Hoffert et al. are right to emphasize that to curb global warming will entail massive technological change. Research and development is needed to make new technologies viable, so it is worrisome that public energy-related R&D is declining in nearly every industrialized country7,8. In the United States -- the biggest spender on energy R&D -- the total funding fell by 40% from 1985 to 1994, and worse was to come: gas and electricity companies cut basic research by two-thirds9 from 1995 to 1996, as restructuring and deregulation of energy markets led them to concentrate on short-term returns. Society as a whole prospers by the new concepts that emerge from basic research, but competitive companies have little incentive to make the necessary long-term investments. Nevertheless, efforts in some countries to reverse the trend are bearing fruit.
To slow global warming will require policies that penalize carbon emission and encourage higher investment in energy R&D. Those policies can be more effective if guided by economic and engineering analysis, but only if the models used to assess long-term global-warming costs and benefits include an improved representation of technological change10. Research is also needed to identify the best policies for directing R&D. Hoffert et al. suggest something on the scale of the Manhattan Project or Apollo Program as models. But unlike those publicly funded crash programmes, an effective energy R&D programme must heed market conditions -- after all, the market will dictate which technologies are eventually adopted.
And there is an even more fundamental question to be answered: what should be the objective of global-warming policy? Many studies assume that we should aim to stabilize atmospheric CO2 at 550 parts per million, a convenient number that is about twice the concentration before the industrial revolution, and about 50% more than today's. But there is little solid evidence to justify that choice of number; nor is it clear whether any single number can express the damage that could result from global warming. For now, taking action on global warming is akin to buying insurance with an unknown premium against unknown hazards. Investing in new technology may be a good hedge against that uncertain future, but exactly what to do is still unclear.
David G. Victor is at the Council on Foreign Relations, 58 East 68th Street, New York, New York 10065, USA.
1.Hoffert, M. I. et al. Nature 395, 881-884 (1998).
2.Lutz, W., Sanderson, W. & Sherbov, S. Nature 387, 803-805 (1997).
3.Nakienovi, N. Daedalus 125, 95-112 (1996).
4.Manne, A. & Richels, R. Buying Greenhouse Insurance: The Economic Costs of CO2 Emission Limits (MIT Press, Cambridge, 1992).
5.Nakienovi, N. et al. Energy 18, 401-409 (1993).
6.Nakienovi, N., Gr,bler, A. & McDonald, A. (eds) Global Energy Perspectives (Cambridge Univ. Press & World Energy Council, 1998).
7.International Energy Agency, IEA Energy Technology R&D Statistics: 1974-1995 (IEA, Paris, 1997).
8.Dooley, J. J. Energy Policy 26, 547-555 (1998).
9.Morgan, M. G. & Tierney, S. F. Issues in Science and Technology 15, 81-87 (1998).
10.Gr,bler, A. Technology and Global Change (Cambridge Univ. Press, 1998).