each appear to be absorbing ~2.0 Pg C y-1, leaving about 1.7 Pg C y unaccounted for. Most of this "missing carbon" is probably going into the terrestrial biosphere primarily in the Northern Hemisphere. The CO2-fertilization effect is, probably, also contributing to the increased capture of C in terrestrial ecosystems.

In its Second Assessment Report the Intergovernmental Panel on Climate Change (IPCC, 1996) estimated that it may be possible over the course of the next 50 to 100 years to sequester 40 and 80 Pg of C in cropland soils (Cole et al., 1996; Paustian et al., 1998; Rosenberg et al., 1998). Reference to Table 1 shows that, if this is so, agricultural soils alone could capture enough C to offset any further increase in the atmospheric inventory for a period lasting between 12 and 24 years. These calculations are still crude and cannot be taken as certain, but they do suggest a potential to offset significant amounts of CO2 emissions by sequestering C in the soils of lands currently in agricultural production. Of course, there is additional C sequestration potential in the soils of managed forests and grasslands (which we do not address here). And, as is discussed below, there is a very large potential for C storage in the soils of degraded and desertified lands. However, a caution needs to be raised here: unless alternatives to fossil fuels are found, the energy demands created by growing populations and rising standards of living could greatly increase CO2 emissions over the next century and the capacity of agricultural soils to sequester carbon could be exhausted to little long-term effect.

The carbon content of the atmosphere can be stabilized either by decreasing the rate at which greenhouse gases are emitted to the atmosphere or by increasing the rate at which they are removed from it. It was well recognized that photosynthesis, by fixing C in standing and below ground portions of trees and other plants, provides a powerful means of removing CO2 from the atmosphere and sequestering it in the biosphere. The Kyoto Protocol establishes the concept of credits for C sinks (Article 3.3) but allows credits for only a limited list of activities including afforestation and reforestation (Article 3.4). As of this writing, the Protocol does not allow credits for sequestration of C in soils except, perhaps (indeed, this is not yet clear), for carbon accumulating in the soils of afforested and reforested land. Although the capacity for doing so clearly exists, sequestration in agricultural soils is not now permitted to produce C sequestration credits under the Kyoto Protocol. This mitigation option was set aside in the Kyoto negotiations ostensibly because of the perceived difficulty and cost of verifying that C is actually being sequestered and maintained in soils. However, the soil carbon sequestration option is specifically mentioned in Article 3.4 for possible inclusion at a later time and will be discussed at COP VI in the Hague this fall.

Another way of looking at the potential role of soil C sequestration is shown in Figure 1, produced with the integrated assessment model MiniCAM 98.3 (Edmonds et al, 1996a,b; Rosenberg et al., eds. 1999). The top line in the figure represents the anticipated increase in carbon emissions to the atmosphere from the year 2000 to the end of the 21st century under a MiniCAM "business-as-usual" scenario. It also shows a more desirable emissions trajectory that allows atmospheric [CO2] to rise from its current level and stabilize at a maximum of 550 ppmy by 2035 (Wigley et al., 1996). Annual C emissions are allowed to increase at first but then are lowered steadily to reach a level in 2100 between 6–7 Pg C y-1. For the upper emissions line to be brought down to the desired level will require great changes from our current energy systems. The caption of Figure 1 identifies some of the technologies that will create such change in the 21st century. Increased efficiency in the uses of fossil fuels, development of non-carbon emitting fuels, improvements in power generation, a greater role for biomass, solar, wind, and nuclear energy and other technological advances will ultimately be needed to mitigate climate change. Figure 1 shows that soil C sequestration can play a very strategic role but cannot, in and of itself, solve the problem. Soil C sequestration alone could make up the difference between expected emissions and the desired trajectory in the first 3-4 decades of the 21st century, buying time for development of the new technological advances identified above. The calculations shown in Figure 1 are based on the assumption that from 2000 to 2100 agricultural soils sequester C at global annual rates ranging from 0.4 to 0.8 Pg y, with rates twice as great in the initial years and half as great in the later years. It is further assumed that the full potential of soil C sequestration is

1 Estimates of soil sequesterable carbon in agricultural soils are more conservative in a Special Report of the IPCC Summary for Policymakers, 2000, entitled "Land Use, Land-Use Change, and Forestry", For example, assuming 30 percent of the global agricultural soils are managed with practices that increase C sequestration, the annual net change in C stocks in agricultural soils in 2010 would be 125 Mt C per yr. However, improved management on only 10 percent of global grazing lands would sequester 240 Mt C per yr.

realized without any additional net cost to the economy-not unreasonable in view of the known benefits of organic matter in soils. In addition, by allowing time for new technologies to be developed and for existing facilities to live out their design lifetimes, the costs of an avoided tonne of carbon emissions over the next century can be cut approximately in half.

How realistic are the estimates of potential soil C sequestration on which the economic modeling is based? The IPCC estimates for cropland assume the restitution of up to two-thirds of the soil C released since the mid-19th century by the conversion of grasslands, wetlands and forests to agriculture. The experimental record confirms that C can be returned to soils in such quantities. Some examples: carbon has been accumulating at rates exceeding 1 Mg ha-1 y in former U.S. crop lands planted to perennial grasses under the Conservation Reserve Program (CRP) (Gebhart et al, 1994). Soil C increases ranging from 1.3 to 2.5 Mg ha-1 y have been estimated in experiments on formerly cultivated land planted to switchgrass (Panicum virgatum), a biomass crop (preliminary data, Oak Ridge National Laboratory). Further, there have been a substantial number of experiments over the last two or three decades with minimum tillage and no-till management of farm fields demonstrating that such practices lead to increases in soil C content (Lal et al., 1998a; Nyborg et al., 1995; Janzen et al., 1998).

Despite these indications that needed quantities of C can be sequestered in agricultural soils there are still important questions to be answered. Among them 4 appear to be critical: (1) Can methods be developed to increase still further the quantities of C that accumulate in soils and, perhaps more importantly, the length of time during which the C resides in soils? (2) Can opportunities for soil C sequestration be extended beyond the currently farmed lands to the vast areas of degraded and desertified lands worldwide. (3) Can we develop quick, inexpensive and reliable methods to monitor and verify that carbon is actually being sequestered and maintained in soils? and (4) What are the policy and economic problems associated with implementation of soil carbon sequestration programs worldwide?

A workshop to explore these questions was organized by the Pacific Northwest National Laboratory, the Oak Ridge National Laboratory and the Council for Agricultural Science and Technology and was held in December of 1998 in St. Michael's, MD. The workshop was attended by nearly 100 Canadian and U.S. scientists, practitioners and policy-makers representing agricultural commodity groups and industries, Congress, government agencies, national laboratories, universities and the World Bank. Support for the workshop was provided by the Environmental Protection Agency, the U.S. Department of Agriculture, the Department of Energy, the Monsanto Company and NASA.

Some general conclusions of the workshop are given here.

• New Science. The potential for carbon sequestration in all managed soils is large and progress can be made using proven crop, range and forest management practices. But this potential might be made even greater if ways can be found to restore more than the two-thirds of the carbon that has been lost from conversion to agriculture and perhaps even to exceed original carbon contents in some soils and regions. This would involve a search for ways to effect greater, more rapid and longer-lasting sequestration. Promising lines of research are evolving that could lead to an improved understanding of soil C dynamics and the subsequent development of superior C sequestration methods. These studies aim to: improve understanding of the mechanisms of C stabilization and turnover in soil aggregates; improve description of the various carbon pools and transfer among them to better model the dynamics of soil organic matter; improve understanding of landscape effects on C sequestration and how it might be controlled through precision farming; apply genetic engineering to enhance plant productivity and favor C sequestration; and better understand the environmental effects of soil C sequestration (e.g., erosion, nutrient leaching, emissions of other greenhouse gases).

• The Soil Carbon Sequestration / Desertification Linkage: It is estimated that there are some 2 billion hectares of desertified and degraded lands worldwide, 75 percent of them in the tropics, with degradation most severe in the dry tropics. The potential for carbon sequestration on these lands is probably even greater than on currently farmed lands. Improvements in rangeland management, dryland farming and irrigation can add carbon to soils in these regions and provide the impetus for changes in land management practices that will begin the essential process of stabilizing the soil against further erosion and degradation with concomitant improvements in fertility and productivity. Erosion control, agricultural intensification, forest establishment in dry regions, and biomass cultivation appear to offer the greatest potential for increased se

questration on degraded lands. Soil carbon sequestration offers a special opportunity to simultaneously address objectives of two United Nations Conventions-the Framework Convention on Climate Change and the Convention to Combat Desertification.

• Monitoring and Verification: There is opposition to using soil carbon sequestration in the Kyoto Protocol calculations. One cause of the opposition is the perception that it will be difficult, if not impossible, to verify claims that carbon is actually being sequestered in the soils of fields around the world that may eventually number in the millions. It is currently possible to monitor changes in soil carbon content, but current methods are time-consuming and expensive and are not sensitive enough to distinguish year-to-year changes. If there are to be international agreements allowing soil sequestration to figure into a nation's carbon balance, agreed-upon means of verification will be required. Improved methods for monitoring changes in soil organic carbon might involve spatial integration based on process modeling and geographical information systems, application of high-resolution remote sensing, and continuous direct measurements of CO2 exchange between the atmosphere and terrestrial ecosystems. There may very well be a market for new instruments that can serve as "carbon-probes". These verification and monitoring methods will have to be developed or tailored to operate at different scales (e.g., the field, the region). Verification of changes in soil C in individual fields will rely on laboratory analyses of soil samples or, perhaps a few years from now, on carbon probes. Estimates of soil C changes at the regional scale will be made with the aid of simulation models. High resolution remote sensing and GIS will be used to extrapolate C sequestration data from field observations and modeling results and aggregate them to still broader regions and to track trends in C sequestration with time.

Implementation Issues and Environmental Consequences: The prospect that carbon may become a tradable commodity has not gone unnoticed in the agricultural and forestry communities. Beneficial land-management practices might be encouraged if credit toward national emissions targets could be gained by increasing the stores of carbon on agricultural lands. However, uncertainty about the costs, benefits and risks of new technologies to increase carbon sequestration could impede their adoption. Financial incentives might be used to encourage adoption of such practices as conservation tillage. Government payments, tax credits, and/or emissions trading within the private sector are also mechanisms that could be employed to overcome farmer reluctance. Despite uncertainty of many kinds, the process is beginning. We do not yet fully understand the social, economic and environmental implications of incentives that lead to a widespread adoption of soil carbon sequestration programs. Most foreseeable outcomes appear benign-for example an increased commitment of land to reduced tillage practices. Another likely outcome would be increased effort aimed at restoration of degraded lands and for retirement of agricultural lands into permanent grass or forest cover. Continuation and/or expansion of Conservation Reserve programs might also be encouraged and lead to improved management of residues in agricultural harvests. All of these actions have the potential of reducing soil erosion and its negative consequences for water quality and sedimentation. In addition, since increases in soil organic matter content increase water-holding capacity, irrigation requirements could be reduced. Conversion of agricultural lands to grasslands or forests would expand to provide wildlife habitat. Reduced soil disturbance and, possibly, more efficient use of fertilizers could alter the volume and chemical content of runoff from agricultural lands. This would in turn reduce water pollution and improve water quality and the general ecology of streams, rivers, lakes and aquifers in these regions for use by non-agricultural water consumers.

But negative effects are also possible. Programs designed to move agricultural lands into forestry could negatively affect the traditional forest sector, leading to either deforestation of traditional parcels or reduced levels of management and lessened C sequestration. Such actions might offset much of the benefit of sequestering C in agricultural soils. Expanded use of agricultural lands for C sequestration might compete with the use of agricultural lands for traditional food and fiber production. The result might well be decreased production, increased consumer prices for crops, meat and fiber and decreased export earnings from agriculture. Reduction in intensity of tillage often leaves more plant material on the soil surface. Conservation tillage has been found to require additional use of pesticides to control weeds, diseases and insects. Increased use of pesticides may have detrimental effects on ecological

systems and water quality. Conversion of croplands to grasslands decreases emissions of N2O and increases oxidation of CH4, another strong greenhouse gas.

Discussions at the workshop took place with recognition that there is no "free lunch", even in the case of such an apparently benign activity as soil carbon sequestration. The production, transport and application chemical fertilizers, manures and pesticides and the pumping and delivery of irrigation water needed to increase plant growth and encourage C sequestration all require expenditures of energy and, hence, the release of CO2 from fossil fuels. It is clearly necessary to determine to what extent the energy costs of the practices used to increase C_sequestration actually reduce the net carbon-balance benefits. Professor Michael Schlesinger of Duke University brought this question to sharp focus in an invited critique of the "New Science" issue paper at the St. Michael's workshop and subsequently in a Forum article for Science (Schlesinger, 1999). Other analysts (e.g. Izaurralde et al., 2000) take issue with his assertions in that article that nitrogen fertilization, the application of manures and irrigation in semi-arid regions have associated carbon costs that effectively negate any net carbon sink resulting from these practices. Aside from their arguments with the details of Schlesinger's calculations, these analysts make the critical point that no-one seriously believes that agricultural soils will ever be managed for the primary purpose of C sequestration. Fertilizers, manures, chemicals and irrigation water will continue to be used primarily for the production of food, fiber and, increasingly in this new century, for the production of biomass as a substitute for fossil fuels.

References:

Cole, C. V., C. Cerri, K. Minami, A. Mosier, N. Rosenberg, and D. Sauerbeck. 1996. Agricultural options for mitigation of greenhouse gas emissions. Chapter 23. Climate Change 1995: Impacts, Adaptations and Mitigation of Climate Change pp. 745-771. Report of IPCC Working Group II, Cambridge University Press, 880 pp. Edmonds, J. A., M. Wise, H., Pitcher, R. Richels, T. M. L. Wigley, and C. MacCracken. 1996a. An integrated assessment of climate change and the accelerated introduction of advanced energy technologies: An application of MiniCAM 1.0. Mitigation and Adaptation Strategies for Global Change. 1:311–339.

Edmonds, J. A., M. Wise, R. Sands, R. Brown, and H. Kheshgi. 1996b. Agriculture, Land-Use, and Commercial Biomass Energy: A Preliminary Integrated Analysis of the Potential Role of Biomass Energy for Reducing Future Greenhouse Related Emissions. PNNL-11155. Pacific Northwest National Laboratory, Washington, DC.

Gebhart, D. L., H. B. Johnson, H. S. Mayeux, and H. W. Pauley. 1994. The CRP increases soil organic carbon. Journal of Soil and Water Conservation. 49:488–492. IPCC. 1996. Climate Change 1995: The Science of Climate Change. Report of Working Group I. Cambridge University Press, New York. P. 4.

Izaurralde, R.C., W.B. McGill, and N.J. Rosenberg. 2000. Carbon cost of applying nitrogen fertilizer. Science 288:811-812.

Janzen, H. H., C. A. Campbell, R. C. Izaurralde, B. H. Ellert, N. Juma, W. B. McGill, and R. P. Zentner. 1998. Management effects on soil C storage on the Canadian prairies. Soil Till. Res. 47:181-195.

Lal, R., L. Kimble, R. Follett, and B. A. Stewart, eds. 1998a. Management of carbon sequestration in soil. Adv. Soil Sci., CRC Press, Boca Raton, Florida.

Nyborg, M., M. Molina-Ayala, E. D. Solberg, R. C. Izaurralde, S. S. Malhi, and H. H. Janzen. 1998. Carbon storage in grassland soil and relation to application of fertilizer. Management of carbon sequestration in soil Adv. Soil Sci., CRC Press, Inc., Boca Raton, Florida. Pp. 421-432

Paustian, K., C. V. Cole, D. Sauerbeck, and N. Sampson. 1998. Mitigation by agriculture: An overview. Climatic Change. 40:135–162.

Rosenberg, N. J., C. V. Cole, and K. Paustian. 1998. Mitigation of greenhouse gas emissions by the agricultural sector: An introductory editorial. Climatic Change. 40:1-5.

Rosenberg, N. J., R. C. Izaurralde, and E. L. Malone, eds. 1999. Carbon Sequestration in Soils: Science, Monitoring and Beyond. Proceedings of the St. Michael's Workshop, December 1998. Battelle Press, Columbus, Ohio. 199 pp.

Schlesinger, W.H. 1999. Carbon sequestration in soils. Science 284:2095.

United Nations. 1992. United Nations Framework Convention on Climate Change. United Nations, New York.

United Nations. 1997. Report of the Conference of the Parties on its Third Session. Held at Kyoto from December 1-11, 1997. Kyoto Protocol, FCCC/CP/1997/7/Add.1, United Nations, New York.

Wigley, T. M. L., R. Richels and J. A. Edmonds. 1996. Economic and environmental choices in the stabilization of atmospheric CO2 concentrations. Nature. 379:240-243.

Figure 1. Global Carbon Emissions Reductions: WRE 550 (Wigley et al., 1996, 550 ppmv atmospheric CO2 concentration). This figure shows a hypothetical path to carbon emissions reductions from MiniCAM's business as usual (BAU) emissions pathway to the WRE 550 concentration pathway, under a scenario in which credit for soil carbon sequestration is allowed. Soil sequestration of carbon alone achieves the necessary net carbon emissions reduction in the early part of the century. From the middle of the century on, further emissions reductions must come from changes in the energy system (such as fuel switching and the reduction of total energy consumption).

The CHAIRMAN. Thank you, Dr. Rosenberg.

Ms. Mesnikoff, Dr. Romm in his written statement said: "The fundamental relationship between energy use and economic growth in the United States has been changed permanently by the spread of new economy technology to every corner of our lives." Do you agree with that statement?

Ms. MESNIKOFF. I think I would agree with that statement, and I think that, going back to the issue of the need to use more coal, I think we need to begin to look at the fact that most of the homes in this country still use incandescent lightbulbs, which are tremendously inefficient as compared to the compact fluorescent bulb, which would save about 400 pounds of coal over its lifetime.