Conventional estimates of net CO2 release from land-use change do not take the possibility of a CO, fertilization effect into account. That is, they do not make any allowance for the possibility that the terrestrial biosphere may be sequestering carbon at an increasing rate. With some uncertainty, it has been suggested that increases in the atmospheric concentration of CO2, and or nitrogen, could act to accelerate the rate at which the terrestrial biosphere stores carbon.

Land-use change in the United States is estimated to be a net sink for carbon. While harvesting of trees is estimated to release 0.35 PgC/yr, this quantity is offset by an estimated 0.46 PgC/yr absorption leaving a net 0.1 PgC/yr net uptake (US, 1992). Emissions of carbon from land-use change are much greater for nations such as Brazil, Columbia, the Ivory Coast, Indonesia, Laos, and Thailand, though great uncertainty surrounds emissions estimates in all cases.

CFC Emissions: Most key nations have either signed or agreed to sign the Montreal Protocol and the subsequent London Amendments. The United States is committed to phase out CFC production by 1996. Global emissions of halocarbons including CFCs amounted to 1.7 Tg/yr in 1990. United States emissions amounted to 0.7 Tg/yr in 1988. Recent findings reported in IPCC (1992) indicate that the indirect global warming effect of CFC emissions through the destruction of O, may be equal in magnitude, but opposite in sign to the direct effects, over the course of 100 years. If this were the case, the net contribution of CFCs to global warming would be zero over the course of 100 years.

Methane Emissions: Estimates of methane emissions have improved over time, although the uncertainty surrounding sources remains greater than with fossil fuel carbon. Manmade emissions are currently thought to derive from energy use (coal mining, and natural gas production, transmission, and distribution), rice paddies, animal husbandry (ruminant livestock and animal waste), anthropogenic waste (domestic sewage treatment and landfills), and biomass burning (including biomass for energy as well as land-use change and crop field burning). Total global anthropogenic emissions were estimated to have been approximately 350 PgCH4/yr. Uncertainty in this estimate is in the range of +50%. United States emissions are estimated to be within the range 0.015 to 0.051 PgCH4/yr.

Nitrous Oxide Emissions: Nitrous oxide emissions are poorly documented, and all major sources and sinks may not have been identified. Anthropogenic emissions are estimated to be in the range 1 to 6 TgN/yr (IPCC, 1992). Nevertheless, anthropogenic emissions are a relatively small contribution to the overall N2O budget. Principal sources of emissions include cultivated soils, biomass burning, stationary combustion, mobile sources, adipic acid production, and nitric acid production. United States emissions have been estimated to be 0.34 to 1.21 TgN/yr.

Sulphur Emissions: Sulphur emissions are a relatively recent addition to the list of RIGs. While the potential for sulphur to play a role in global warming has been known for some time, not until IPCC (1992) did the IPCC recognize that sulphur aerosols could be a major factor in explaining observed global temperature records.3 Unlike other RIGS, the addition of sulphur to the atmosphere tends to cool the surface of the Earth. Unfortunately, sulphur is also an acid rain

3 The IPCC (1992, p.14) concluded that "for clear-sky conditions alone, the cooling caused by current rates of emissions has been estimated to be about 1 W/m2 averaged over the Northern Hemisphere, a value which should be compared with the estimate of 2.5 W/m2 for the heating due to anthropogenic greenhouse gas emissions up to the present."

precursor. Anthropogenic emissions are estimated by IPCC (1992) to range from 71 to 83 TgS/yr in contrast to the range of estimates for natural emissions, 7 to 14 TgS/yr. United States emissions were estimated to be 10 TgS/yr in 1989 (WRI, 1992), but reductions are required under the Clean Air Act Amendments of 1990. We note that sulphur compounds are not covered in the inventory developed for the United States National Action Plan for Global Climate Change (1992).

ANTHROPOGENIC EMISSIONS FORECASTS

The range and variety of estimates of future greenhouse gas emissions vary greatly by gas. IPCC Working Group I (WG1) worked on this problem and presented its results in IPCC (1992).

Fossil Fuel Carbon: The fossil fuel resource base provides no constraint on future atmospheric CO2 release. The present atmospheric stock of carbon is approximately 740 PgC (1988). The estimated resource of fossil fuels is huge by comparison. While the carbon content of conventional oil and natural gas resources is only slightly more than half as large as the current atmospheric stock of carbon, coal resources are an order of magnitude larger. The carbon content of unconventional oil resources is 55 times larger than the current atmospheric stock of carbon. The pool of carbon available for combustion might be constrained to 4000 PgC by considering only those resources recoverable under present technologies. Even this severely constrained resource definition provides no physical constraint on climate change from fossil fuel use. Approximately 80% of the coal resource base is thought to be in three areas: the United States, former Soviet Union, and China. There are approximately 800 PgC in the form of coal, recoverable with today's technologies, within the jurisdictional boundaries of the world's other countries.

A great deal of attention has been placed on the potential future fossil fuel carbon emissions. Edmonds et al. (1992) reviewed 30 selected reference case trajectories from 18 analyses of fossil fuel carbon emissions for comparison to the IPCC 1990 reference cases. These studies included analyses of uncertainty. It is highly likely that without intervention to reduce emissions, emissions will rise over the course of the next century. The rate at which emissions will rise is highly uncertain. Relatively moderate emissions trajectories can place 1000 PgC into the atmosphere by the year 2100. Even if fossil fuel carbon emissions were stabilized immediately at 1990 rates, 6.1 PgC/yr, cumulative emissions would reach approximately 650 PgC over the period to 2100. Note that only about half of that amount would be expected to accumulate in the atmosphere because carbon is removed in a very complex process referred to as the carbon cycle, a subject discussed in other testimony presented today. It is interesting to note that the stabilization of emissions at present levels implies that global average emissions per person must be roughly cut in half over the course of the next century.

The range of fossil fuel emissions scenarios is driven by the supply and demand for energy. These in turn reflect patterns of economic and population growth, changes in end-use energy intensity, the availability of inexpensive fossil fuels such as oil and natural gas, and the availability and cost of competing sources of energy services such as conservation, renewables, nuclear power, and fusion energy. In the uncertainty analysis conducted by Reilly et al. (1987) three factors were disproportionately important in shaping business as usual future emissions scenarios: the rate of growth of labor productivity (directly related to GNP growth), the rate of exogenous end-use energy intensity improvement (i.e., the non-price induced change in energy productivity brought about by changes in technology and changing composition of product mix), and the income elasticity of

demand for energy in developing nations (i.e., the relationship between economic growth and demand for market energy). Other factors, such as population growth were less influential in shaping the variation in future emissions trajectories.

The lower ranking of population growth in determining overall emissions growth stems from two factors. First, changes in adult population are much more important in determining overall emissions than changes in total population. Only when population reaches adult age does it begin to have a major impact on national output and therefore on national energy consumption. Changes in fossil fuel emissions lag behind any change in population growth rate. Furthermore, the variation in possible population growth scenarios is not as great as the variation in factors such as economic growth.

Land-Use Change: Estimates of net carbon emissions from land-use change under "business-asusual" conditions begin with the handicap that present emissions rates are highly uncertain. Unlike fossil fuel carbon emissions, cumulative emissions from land-use change are bounded by the total stock of carbon stored in above ground living matter, approximately 560 PgC. None of the IPCC scenarios released as much as half that amount. The maximum rate of net carbon emissions from land-use change in the set of IPCC scenarios was approximately 3 PgC/yr, and in the later years of some scenarios, net emissions were negative. Land use is determined by a somewhat different set of factors than energy use. The demand for land depends on the productivity of land under alternative uses, technological options for producing land services, the availability of capital to implement technological alternatives, the value of goods produced by the land, and policy options. CFCS: The least complicated of the emissions are those of the chlorofluorocarbons (CFCs). These gases are being phased out under the Montreal Protocol and subsequent agreements. Emissions in both the near- and long-term will likely be substantially below 1990 levels. Furthermore, as we discussed earlier, the indirect effect on global warming of CFCs, through the pattern of ozone destruction, may lead to a cooling over the course of a century which is equal in magnitude to the direct warming that has already taken place, but has been hidden by the CFCs themselves.

Substantial uncertainty surrounds the pattern of future production, use and character of CFC substitutes. At present hydrogenated chlorofluorocarbons (HCFCs) and hydrogenated fluorocarbons (HFCs) are most likely replacements of CFCs, though HCFCs are also scheduled to be phased out. While these gases have much shorter lifetimes than the CFCs, they are nonetheless radiatively active and can affect global warming. Ultimately a "third generation" of compounds must be developed to replace CFC, HCFC, and HFC compounds. But this "third generation" of compounds have not yet been developed and it is impossible to know their greenhouse characteristics.

Methane: Because the sources of CH, are uncertain, forecasts of emissions are also uncertain. Most forecasts simply project the rate of accumulation in the atmosphere to continue. More recently attempts have been made to estimate future emissions based on estimates of energy (coal mining, gas production, and landfills) and agricultural activities (rice cultivation and ruminant livestock production), which in turn are determined by assumptions about population, economic growth, tastes, and the assumed natural gas resource base. Studies that have attempted to develop emissions forecasts from forecasts of underlying human activities, assuming business as usual conditions, yield growing CH, emissions (for the period to 2050) ranging from 0.5%/yr (EPA, 1989) to 1.25%/yr (Rotman et al., 1989). IPCC (1992) developed a series of methane emissions

trajectories which range from a scenario with only 15% emissions growth by the year 2100 to a case in which total emissions more than approximately double by 2100.

Nitrous Oxide Emissions: Forecasts of future NO emissions vary greatly. Those which were constructed prior to the discovery of a sampling artifact, which occurred in early energy related emissions coefficients, link future emissions of N,O primarily to the use of coal. Emissions growth rates for such studies as Rotman et al. (1989) and Mintzer (1987) for the period to 2050 produce rates of growth of emissions that range between 0.50 and 1.75 %/year. The U.S. EPA (1989) uses a much lower emission coefficient for fossil fuel N,O with a consequent lower rate of emission growth, 0.3%/year. IPCC (1992) developed a series of N2O emissions trajectories which range from a scenario with almost no growth in emissions through the next century to a case in which total emissions approximately double (reach 19 TgN/yr in 2100). Forecasts show a relatively small role for N2O in determining future radiative forcing, generally contributing 5% and almost always less than 10% of total radiative forcing, because there are vastly fewer molecules of N2O in the atmosphere relative to CO2. This result is in spite of the fact that an individual molecule of N2O is estimated to be approximately 250 times more efficient in absorbing infrared radiation than a molecule of CO2 and has an extremely long average residence time in the atmosphere, 100 to 175 years. These factors are generally counterbalanced by low emissions, which grow at rates similar to those forecast for other greenhouse gases.

Sulphur Emissions: The first major study of global sulphur emissions developed using an integrated model was performed by the IPCC. The range of future emissions depends greatly on policies to reduce acid rain throughout the world. In the IPCC (1992) low case, emissions actually decline globally by approximately 20% by the year 2100. In the highest emissions case, emissions grow by more than 150% in that same period. Unlike the other RIGS, sulphur has a cooling effect on global temperature, leading to the irony that improving local and regional air quality may in fact "unmask" some of the greenhouse effect hidden by the cooling influence of sulphur aerosols.

SUMMARY

Greenhouse gases are effectively transparent to incoming sunlight but absorb infrared radiation escaping to space and thereby warm the surface of the Earth. Human activities result in the release of greenhouse and related gases in such quantities that they are changing the composition of the atmosphere. Greenhouse and related gases which human activities release include carbon dioxide (CO2), methane (CH4), carbon monoxide (CO), nitrous oxide (NO), nitrogen oxides (NO), chlorofluorocarbons (CFCs) and sulphur dioxide (SO2).

Carbon dioxide is the most important of the greenhouse-related gases. Approximately 7.4 PgC/yr are released globally by fossil fuel burning and land-use change. Fossil fuel burning is presently the dominant source of emissions. The United States is the largest single emitter of CO2 from fossil fuel use, with annual carbon emissions exceeding 5 tonnes per person, as well as being a major contributor to the release of all other gases. However, the United States is actually a net absorber of CO2 from land-use change.

Future projections of global emissions under business as usual conditions show stable or growing anthropogenic emissions to be highly likely for most greenhouse and related gases over the long term. United States emissions of fossil fuel carbon are anticipated to rise under business as usual

assumptions, but the United States' relative role in future emissions is expected to decline over time as developing nations pursue economic growth objectives. There are no important natural constraints on the release of fossil fuel carbon. Carbon release from land-use change can be large, but cannot attain the scale of fossil fuel use. The Montreal Protocol and subsequent amendments are expected to reduce global emissions. The Clean Air Act Amendments of 1990 implement the United States phase-out of CFCs. These emissions trends in conjunction with recent findings of a counterbalancing indirect (cooling) effect of CFC emissions through O, depletion result in a lessened concern about future greenhouse contributions by CFCs.

Thank you, Mr. Chairman. I would be pleased to answer questions.