scientists agree that a doubling of carbon dioxide abundances, compared to pre-industrial values, will occur within the next century.

The existence of other greenhouse gases is well known. Methane, ozone in the lower atmosphere, and nitrous oxide are also greenhouse gases. Their abundances are also increasing in the atmosphere. The reasons for the increases are only partially clear. The sources of these other gases are not as clear, since the biological mechanisms for their emission characteristics are not completely known at present. While the future atmospheric abundances of these gases cannot yet be predicted reliably, it is clear that most of these gases will substantially add to the greenhouse effect.

Over the past year, it has been discovered that the ozone loss caused in the lower stratosphere by the chlorofluorocarbons (CFC's) reduces the radiative forcing to the lower atmosphere. Current climate models suggest that this ozone loss is possibly causing a cooling contribution to surface temperatures, perhaps offsetting much or all of the direct warming effect of the CFC's. Thus, the current picture is that the CFC's do not have as large a contribution to global warming as was originally thought. Other compounds in the atmosphere, such as pollution aerosols, also may have a surface cooling effect.

Confidence is high in the understanding of the basic radiation physics of trace gases. The above-mentioned gases act to reduce the loss of outgoing thermal radiation to space, thereby increasing the radiation back toward the surface of the planet. Scientists are in solid agreement that the greenhouse gases increase the amount of heat retained by the lower atmosphere (i.e., the "thermal forcing" of the climate system of the lower atmosphere).

The large majority of climate scientists currently believe that the eventual response of the climate system to greenhouse forcing is very likely to be, on the average, a global warming. While it is clear that current science can accurately calculate the thermal forcing of the atmosphere due to increases in the greenhouse gases, it is also clear that estimating the actual response of the climate system to that forcing is a much more difficult task. Several of the key processes that govern that response are quite complex; for example, clouds can both reflect solar radiation and absorb the heat radiated by the Earth's surface.

Many of these processes affect other parts of the system, in what are called "feedbacks". An example of a positive feedback is ice and snow. Warming is amplified when ice and snow melt during warmer periods exposing darker surfaces, which absorb more radiation and add to the warming. Other processes could involve negative feedbacks that could dampen the warming. It is the net effect of all of the feedbacks that determines the new equilibrium climate state that the planet will reach as a result of the increase in radiative forcing by the greenhouse gases.

Scientists have simulated this complex system with computer models and have used these models to estimate the consequences of likely future increases in greenhouse gas abundances. Based on those simulations, most scientists believe that for a doubling of greenhouse gases measured in carbon dioxide equivalents, an eventual global average warming in the range of 1.5-4.5 degrees Celsius is likely as a result of a doubling of the atmospheric abundance of carbon dioxide equivalents. However, based on differences among model results and different approaches to simulating aspects of the global system, some scientists caution that we may not have identified and/or characterized a sufficient number of the key response processes (particularly the feedbacks). As a consequence, the above uncertainty range may be optimistically small. If indeed a warming in the range of 1.5-4.5 degrees Celsius (2.7-8.1 degrees Fahrenheit) were to occur, it would be comparable to or substantially larger than the known temperature changes that have happened naturally in the past several thousand years.

Has a human-induced greenhouse warming been seen already? The large majority of scientists believe that an answer to this question cannot yet be given. Current models predict that, due to the greenhouse gases that are already in the atmosphere, the human-influenced contribution to the global-average surface temperature should be in the range of 0.5 to 1 degrees Celsius (0.9-1.8 degrees Fahrenheit). Has that warming been seen in the temperature record? The answer is not clear. While the surface temperature record shows that there has been an increase of that magnitude over the past several decades, the pattern of that increase-one relatively rapid increase in the 1920's and another in the 1980's-does not match that predicted for the greenhouse effect; namely, a gradual increase in temperature. Therefore, this suggests that there may be other, presumably natural, processes at work that can influence temperature changes of a fraction of a degree Celsius.

As a consequence, scientists now are searching for a greenhouse "signal" whose current magnitude is likely to be comparable to the natural variations of the climate system and to other human-induced variations. This is a challenging task indeed!

Thus, regarding whether a greenhouse warming has or has not been seen, the "jury is still out." Recent studies indicate that unequivocal detection could be several or more years away.

Current greenhouse-warming models do not have the ability to predict the climate of a particular region or the climate of a given year. Those who construct climate models clearly state that, while they can simulate many of the global features of the Earth system very well, the models are not yet sufficiently realistic to yield reliable greenhouse-warming predictions of climate features on regional scales. Similarly, they cannot yet predict the climate of a particular year. This means that the global climate models cannot determine if the U.S. midwestern drought of 1988 was due to the greenhouse effect. Nor can they predict what the climate will be for the next few years. However, many scientists do believe that the models can reliably predict that, because of the greenhouse effect, impacts like the U.S. midwestern drought will become increasingly likely in coming decades.

The above summary gives the current viewpoints of the international scientific community regarding the state of the science of the human-influenced greenhouse effect. While I hope that this current status report will be useful to the members of the subcommittee, I would now like to summarize an assessment process that will provide, within the next few years, an updated view of the improvements in the understanding of global warming. Namely, the Intergovernmental Panel on Climate Change (IPCC) of the World Meteorological Organization (WMO) and the United Nations Environment Programme (UNEP) are planning an international state-ofknowledge review of climate change.

A. Content and Timetable

The main features of this Scientific Assessment of Climate Change are as follows: -There are three companion WMO/UNEP assessments taking place: (i) science; (ii) impacts; and (iii) greenhouse gas emission scenarios and other cross-cutting is

sues.

-They will be prepared internationally. For example, the United Kingdom will continue to lead the science assessment and the United States will lead the impacts assessment.

-The main features of the scientific Assessment of Climatic Change are the following:

-It will be a peer-reviewed scientific document done by the best scientists available and submitted to the IPCC. It will be accompanied by a summary directed to Government officials and will be suitable for IPCC release to the private sector and the public.

-Its scope will be the full climate-change phenomena, namely: climate forcing agents; past climate record; processes involved in climate change; climate model formulation; tests of climate models; predictions of past and future climate, and physical and biological responses to climate change.

-A special emphasis throughout will be an identification of gaps in the current knowledge and a better quantification of uncertainties.

-The full scientific assessment of the state of understanding of climatic change will be available in 1995. An update on a few special topics will precede it in 1994. Of what value are these scientific assessments to decision makers? The answer is "considerable", but it may be useful to elaborate here on the fundamental reasons why:

The assessments are single concise statements from the large majority of the scientific community. In these assessments, the major representatives of the scientific community will speak at one time and place regarding the knowns and unknowns of climate change, including global warming. Where full consensus exists, it will obviously be stated. Where differences of opinion exist, they too will be stated, since they are a measure of current scientific uncertainty. The assessment can be a common reference point for decision makers, rather than having to resort to the views of an individual scientist.

They are international scientific statements. The assessments provide all nations with a common basis of scientific input for their decision making, as opposed to several national statements.

The scientific scope is comprehensive. With these assessments, decision makers have available a single homogeneous summary of the current scientific understanding of the whole climate-change phenomenon, ranging from the natural and humaninduced causes of climatic change to the physical and biological responses to that change. This is much more useful than separate reviews of components of the phenomenon done at different times and perhaps for different purposes.

Both natural and human-induced climate change are considered. In contrast to considering only the potential perturbation of climate by human activities, the assessments place that predicted change into the context of the observed and predicted

changes that are a natural part of the climate system. The comparison of the two will afford an immediate and straightforward insight into the significance of the predicted human-induced perturbations.

The focus on identifying gaps in knowledge and in quantifying uncertainties aids risk analysis. The difference between: (a) "The predicted range of possibilities is from X to Z"; and (b) "The prediction Y" is, with regard to decision making, a highly relevant difference. The first statement explicitly reflects the existence of uncertainties in the prediction.

While the 1995 assessment will usefully summarize the current state of understanding, it is clear that key problems related to global warming will need further elucidation. Therefore, it will prove to be crucially important over the coming years to continue to improve, test, and assess our understanding of the phenomenon, as it is embodied in numerical greenhouse-forcing and climate-response models.

"These models are, of course, only as good as the accuracy and completeness with which their components represent the relevant processes of the "real" world. Some of the shortcomings of our understanding of these processes are clear now and hence define some of the priority tasks that need research emphasis. other required tasks are associated with establishing a better observational system that could not only provide additional input to the models, but also could signal the arrival of a greenhouse warming. Resolving these shortcomings will require collaborative research among several disciplines, including atmospheric and biological sciences.

Some of the highest-priority needs are the following:

-Better characterization of the cloud feedback mechanisms, such as observations and theories that can relate the radiative effects of cloud-scale processes to planetary-scale processes.

-Identification of other significant feedback processes, such as biogenic-emissions/cloud-formation interactions.

-Better experimental and theoretical understanding of the biological, chemical, and physical processes that control the emissions and uptake of the radiatively important trace gases other than carbon dioxide.

-Defining the trends in the atmospheric abundances of likely additional greenhouse gases, such as lower-atmospheric ozone and stratospheric water vapor.

-Characterizing the trends and spatial variations of climate-sensitive properties, such as temperature and ozone, in the middle-to-upper stratosphere.

-Better global coverage of the observations of key response variables, such as surface temperature and albedo at global oceanic locations and a better monitoring record of sea level.

-Establishing more accurate decadal trends in the radiative forcings, such as cloud cover, in order to develop and test improved models of key feedback systems. -Improving the understanding (that is, the observations, process studies, and modeling) of major subcomponents of the climate system, such as the coupled oceanatmosphere of the equatorial South Pacific, since these submodels are part of the basis of eventual global coupled models.

Better characterization of the processes that determine the thermal inertia of the oceans, such as large-scale vertical motions.

Improving the quality of, and learning to interpret better, the long-term record of past climate change in order to develop and test our century-scale models, since the modelers clearly cannot wait for the future centuries of data in order to do so. New data management systems will need to be developed and existing data management systems will need to be improved to handle larger volumes of data, facilitate scientific analyses of retrospective data, and accurately document the historical (century scale) and contemporary evolution of the climate system.

The U. S. Global Change Research Program has responded to these uncertainties by directing research toward four high-priority foci: (i) climate modeling and prediction; (ii) global water and energy cycles; (iii) global carbon cycle; and (iv) ecological systems and population dynamics. In doing so, the U.S. Program is combining the collective efforts of Government and academic scientists, in collaboration with scientists from other countries, on these high-priority multidisciplinary research thrusts.

Global climate has often changed substantially and frequently with severe effects over geological time. There is every reason to believe that such natural change will continue in the future. There is now the likely prospect that human activities will add to climate change in the future. There are two implications for us to consider. They both have separate, different, and equally important associated policy questions:

-Natural variation: How do we accommodate to it?

-Human-induced variation: Do we need to mend our ways?

Policymakers will be addressing both questions. Scientists must assist with both answers. Improved answers require a better understanding of the fundamental processes of the ocean/atmosphere climate system, which is a challenging task. Nevertheless, the fundamental understanding of natural processes that relate to the wellbeing of mankind are almost always cost effective. Regarding our environment and what it means to us all, it is the price of ignorance that we cannot afford.

Mr. Chairman, this concludes my prepared remarks. If I have, while seeking the goal of brevity, strayed from one of clarity, I would be pleased to answer any questions that you or the members of the subcommittee may have.

Mr. SHARP. We are pleased now to hear from Dr. Edmonds.

STATEMENT OF JAE EDMONDS

Mr. EDMONDS. Thank you, Mr. Chairman, and members of the subcommittee, for the opportunity to offer testimony on global climate change this morning. As you are aware, I have been asked to testify on the topic of present and future greenhouse and related gas emissions.

I would like to ask that my prepared statement be included in the record of this morning's hearings. But as time is short, I would propose to talk from a briefer document, which I circulated to the subcommittee earlier entitled "Key Points."

Mr. SHARP. That will be fine.

Mr. EDMONDS. I made eight key points and buttressed them with figures, which you can see by turning over the first page of the "Key points" document.

The first of these Dr. Albritton has already made, and that is that there are multiple greenhouse gases. Not only are there greenhouse gases, but there are also related gases. Greenhouse gases are the ones which are themselves transparent to incoming sunlight and yet trap and re-emit in the infrared spectrum, warming the surface of the Earth.

Related gases affect the concentrations of greenhouse gases, even if they aren't themselves major greenhouse gases, and those include the carbon monoxide and the NOx compounds, NO and NO2. In addition, as we have already heard, there is the sulfur aerosols which have the opposite effect to most greenhouse gases; that is, their presence in the atmosphere tends to cool. And then finally there are the man-made greenhouse gases, the chlorofluorocarbons and the chlorofluorocarbon substitutes.

If you look at the first figure, what I have done is I have taken 1988 global emissions of greenhouse and related gases and I have used coefficients called greenhouse warming coefficients to weight their relative contribution to global warming, and I have done this to get a feel for the relative contributions of each one of these gases to the atmosphere.

You will note that the largest single contributor to the radiative forcing of the 1988 emissions is carbon dioxide and that the other gases in combination would probably make up about an equal contribution, except for the fact that the CFC's, as we have just heard, have an indirect effect, which is a cooling effect, which to a first approximation as best science currently understands the phenomena, is roughly equal in magnitude to the warming effect, leaving them a net non-contributor. And so I have colored those clear and have begun to remove those wedges from the pie chart. But you

can see from this that carbon dioxide is the single most important of gases from the perspective of radiative forcing.

The second point I wanted to make, and that is illustrated on the second figure, is that human emissions of greenhouse and related gases are large relative to natural levels. This figure shows human emissions as a percent of the total annual flux around the year 1990, and you can see that for methane, CH4, carbon monoxide, the NO, compounds, sulfur dioxide, and for the CFC's half or more of all of the fluxes to the atmosphere in that year were contributed by human activities. Only the nitrous oxide compounds had a contribution from human activities below the 50 percent level.

I have left carbon dioxide off of this figure due to the complexities of the carbon cycle and the explanation which would be required to explain any number that I put down there. Let me say that the human activities which emit those, emit carbon dioxide to the atmosphere are extremely important.

And that leads me to the fourth point, which is that fossil fuel use is the largest source of carbon dioxide emissions to the atmosphere. In 1990 there were approximately 7.4 pentagrams or gigatons of carbon emitted to the atmosphere in the world. Fossil fuel use accounted for, as best we can assess, 6 of those, with landuse change being the next most important at 1.3 pentagrams of carbon per year, and cement being a trivial contributor at 0.1. So the fossil fuel use is the largest source of carbon dioxide to the atmosphere.

The fifth point, which is illustrated on the third figure, is the United States is the largest emitter of fossil fuel carbon at approximately 21 percent in the year 1990. This has declined radically from the contribution of the United States in 1950 where that contribution was approximately 40 percent of the world's total. And, as you can see, the United States, while the single largest contributor, has been a shrinking share over time.

The sixth point is that the United States has a relatively high per capita emission rate. Of the top five emitters shown here, the United States, the former Soviet Union, China, Japan, India, you can see that the United States has the highest per capita emissions rate at approximately 5.3 tons of carbon per person per year, and that is significantly higher than the global average of 1.1.

China accounts for more than 10 percent of the world's emissions, and has per capita emissions and of only 0.6 tons per person per year, although that is growing very rapidly as a consequence of Chinese rapid economic growth. And India, of course, is much lower still at 0.2 tons per person per year.

The seventh point, which is illustrated on the fifth figure, is that the United States alone is not going to be able to control global fossil fuel carbon emissions. This figure shows a typical business as usual fossil fuel carbon emissions scenario, starting in the year 1975 and running out to the end of the next century. It is very similar to the middle-to Case A of the IPCC 1992 assessment, and it is very similar in nature to a wide array of scenarios that have been constructed over time.

You can see that by the year 2010 the growth in emissions from the rest of the world-that is, outside the United States-has grown to the point at which those emissions, even if United States