chance to narrow the uncertainties to where, again, most scientists would say they are credible is in my view decades because it takes that long to couple models of atmosphere with models of ocean to include some of the extra processes that we know are important, to connect them to land surface processes, to put in the effects of forests and grasslands, which are very important in mid-continents, to the climate, not just to farmers or ecologists, and then at the same time to get the satellite, and other observational programs. We are looking at a multi-decadal process before we are really going to have strong consensus on the details of the question.

Now, that having been said, what is the basis for the concern? Why is that many people, myself included-as I said in my written testimony, I am one of those who has serious worries about this problem, and personally in my own value system believe it is one worth a prudent response. I think I can summarize that in a number of ways, and that is what I want to try to do in the 5 or so minutes I have here.

In my testimony there is a Figure 1. And some of you may not have it, and certainly the people behind me do not. So, I will just describe it briefly. I actually believe I used it in this committee when I testified in 1988. It shows three possible scenarios for the future. There is an upper scenario which shows a dramatic warming, more than 5 degrees Celsius, somewhere around the middle of the next century. Then there is a very mild scenario which shows, somewhat akin to the Marshall Institute's prediction, something like a half a degree C warming by the end of the 21st century. Then there is a "middle" scenario which is very close to the IPCC's so-called "best guess."

By the way, I have never liked that term "best guess" because I also don't know how we have a basis for knowing what best is. I don't know quite how to choose within that range where the truth actually lies. I guess it is maybe more probable, but "best" is a strange word.

Nevertheless, if you look over that wide range of uncertainty, this figure, which was put together by an international assessment several years before IPCC and is consistent with those from the U.S. National Academy of Sciences and other national and international groups for the last 15 years, has temperature changes anywhere from about a half degree in a century up to almost 5 to 10 degrees in a century.

Now, what are the bases of the uncertainties? It is not just the missing physics. That's maybe a third of the uncertainty in that graph. There are two other uncertainties that are equally impor

tant.

One is what will happen to people. We have to make what we call human behavioral assumptions. How many people will there be in the world? What will their standard of living be? What kind of energy and other technologies will they use for agriculture, energy supplies? What will their supply and demand technologies be, and what level of affluence will occur and so forth? So, these issues all play at least a factor of 2 uncertainty, maybe a factor of 3.

Then what will the biosphere do? When you add CO2, as we heard from an earlier witness-I guess it was Mike MacCracken

CO2 is a fertilizer. It makes plants grow. Remember, there is some high school biology. It increases photosynthesis. But at the same time, there is about as much living organic matter in the trees as there in the air, but there is about twice as much carbon in the soils as there is in the air. If the soils warm up, then the rate at which microbes in the soil decompose that carbon will probably increase. In fact, George Woodwell was once asked how can you prove that, and he quipped, and I think appropriately, that if you don't believe that microbes work more effectively when it is warmer, just unplug your refrigerator before you go to bed, open it in the morning, and sniff.

[Laughter.]

Dr. SCHNEIDER. There is no question that that will happen

Exactly how the ecosystems and the soils will adjust is not clear at all. The point is you can name negative biological feedbacks where CO2 is taken up by living trees. You can name positive biological feedbacks where it is given back, and this is part of that litany to which I referred. Then you combine that with all the uncertainties over cloud feedbacks and ice feedbacks, and you end up with quite literally a factor of 10 projected into the future. Again, I think that is pretty standard consensus wisdom that you've got effects ranging from mild to potentially catastrophic.

Therefore, it is almost polemical for somebody to come along and grab the high scenario as if it were truth, and it is equally polemical to come along and grab the low scenario as if it is truth because I at least do not have a line from providence. I don't know how those uncertainties will fall out, and therefore the question of prudence becomes one about how to hedge relative to the time it will take for the community to really narrow that range, which in my own personal view is on the order of decades.

Now, why else would we have concern? In the Marshall report-I have prepared just quickly while we are here sort of a point-bypoint comparison against the IPCC, which I obviously don't have time to do now. But I will just give you the first point.

The first point is on page 1 of the Marshall report, which I only had yesterday and read on the plane flying in. There is a sentence which says, "One of the main reasons for concern over this aspect of climate change," by which they mean the greenhouse effect, "is the fact that the earth's temperature has risen by approximately half a degree Celsius in the last 100 years.'

If you go to IPCC on page 7, they talk about not the concern being motivated by what the earth did in the last 100 years. That is just simply consistent, but we don't know. It's consistent with or maybe not. It is because of the basic nature of heat trapping physics. What we don't know is how the additional details will add to that or subtract from it.

That is where the debate is. It isn't because of the performance of the earth in the last 100 years or what happened in the last decade because there are, as I said in my written testimony, so many other factors in addition to greenhouse gases that could be adding to or subtracting from, plus a very large natural variability in the system, that it is impossible to validate current climate models as to whether an equivalent doubling of CO2 would be a half degree warming or 4. That whole range is acceptable based

upon what assumption you make for other processes that could have caused climate to change. Everybody I know of agrees that it will be several decades before we can be certain.

Let me add a couple more points about why we have concern, and then I will stop.

First is validation. Dr. MacCracken mentioned that there were a number of validation exercises. Most climate modelers are, in spite of some popular press, fairly conservative scientists. They spend the bulk of their time trying to validate their models. I have a group, which is funded by the EPA, and we have spent years looking at what we call variability diagnosis, in other words, trying to find out if the difference in temperature from one day to the next is accurate in the model relative to nature. It is a very tough test of whether internal processes in the model are okay. Groups spend lots of time comparing to data, and we do that by looking at the temperature differences between winter and summer. We look at the geographic distribution of climate, at ice age interglacial cycles. We look at warm periods in the earth's history. We look at the effect of volcanic eruptions on cooling the earth. The community that does these things doesn't just simply take a theoretical model, spin out the results, tell society to spend a trillion dollars, and go away. There is a tremendous amount of effort at trying to validate these models. Again, all responsible modelers point out that these validations are circumstantial validations. They are not direct validations.

Why else is there concern? Well, ecologists have very recently in the last 10 years spent a lot of time on this problem, and what they have concluded is that species-and there is a vast number, obviously, on earth. We debate whether it is 5 million or 30. In that debate, we don't even know the absolute answer to that. But those species have ranges and abundances-ranges mean where they stop, and abundances, roughly the numbers-that depend on a complicated set of factors such as competition with each other.

But also many of them depend on climatic factors. Many North American birds, for example, as I said in my testimony, stop at a certain latitude in the winter because their bodies simply can't tolerate colder conditions. Well, those kinds of birds could migrate very rapidly where there is climate change. They would just simply respond quickly.

Yet, we know from looking at the last ice age, which began to break up about 15,000 years ago and was essentially gone by 8,000 years ago, that tree species take thousands of years to migrate 1,000 miles. Therefore, the rate at which trees migrate is very slow relative to birds, and insects and mammals are in between.

So, ecologists who are struggling trying to explain the current distribution of species, based on the fact that we have had 10,000 years of reasonably stable weather, are now confronted with the possibility of even at the low end, at the Marshall Institute end, of a half degree Celsius change in a century, or at the high end, something like 5 degrees Celsius change in a century, which compares something between 5 and 100 times faster than natural average rates of change looking at global average temperatures in an ice age to the global average temperatures in our current interglacial. They go up approximately 5 degrees Celsius.

Between 15,000 and 5,000 years ago, the earth warmed up 5 C. That is 9 Fahrenheit. Sea levels came up 300 feet. Species moved thousands of kilometers. Extinctions took place. Major revamping of the ecological face of the earth, all natural. We are not claiming we did it, and I hope our argument doesn't extend that far. But what we have learned from looking at that experiment is that natural rates of change are about a degree Celsius in a millennium on a global sustained basis.

So, the concern that many people have, myself included, is not the certainty in the answer, but that we run from being several times faster than natural rates of change up to 100 times faster. And most ecologists would argue that we haven't got a possible capacity to forecast in detail what kind of tearing apart of ecological communities would occur under scenarios of those kind.

So, therefore, the conclusion that people say when they talk about prudence-and that is the last thing I want to say-is is it prudent to take that experiment, to take that chance. Obviously, that is a value question for which reasonable people can differ.

But let me just add in one last minute what my own personal view of prudence is. My own personal view of prudence is that you slow down something when you do not know whether the outcome could be at the low or the high end, and the way you choose to slow that down is to do those things first that make the most sense whether or not the problem turns out to be serious, what we call a tie-in strategy. Do those things-and I understand you'll be hearing next week so I won't dwell on it, although I mentioned it in my testimony-which are either cost free or negative cost first. Do those things which help solve multiple problems. If you cut out CFC's, not only do you help with the greenhouse effect, but you help reduce ozone depletion. Try to look for high leverage, as any good business would do, strategies, and then as periodic reviews come in every several years from IPCC, NAS, and nongovernmental organizations like Marshall and others and those from environmental groups, we can continually reassess. But that to me is prudence to take those steps first that make the most sense even without global warming, and then because of the exigency of slowing it down, add a little bit more. At least that is my own personal view. Thank you.

[The prepared statement of Dr. Schneider follows:]

PREPARED STATEMENT OF DR. STEPHEN H. SCHNEIDER,1 SENIOR SCIENTIST, NATIONAL CENTER FOR ATMOSPHERIC RESEARCH,2 BOULDER, CO

THE MEDIA DEBATE

Several years after "global warming" hit the headlines in the wake of the heat wave and fires of 1988, so much misinformation about the "greenhouse effect" has been circulated that public understanding is confused and public policymaking paralyzed. The airwaves and printed pages have been clogged with assertions and counterassertions of opposing advocates with charges and countercharges over the alleged seriousness or triviality of global warming.

As a climatologist identified with this subject, I am constantly asked to explain what is actually happening and how important it is. It has accelerated since then to

1 Any opinions, findings, conclusions or recommendations expressed in this article are those of the author and do not necessarily reflect the views of the National Science Foundation.

2 The National Center for Atmospheric Research is sponsored by the National Science Foundation.

the accompaniment of a surge of criticism, some well-intended, some pure vitriol (for example, Detroit News, 1989, see Appendix A). The experience since those hotter than usual summers of '87 and '88 has confirmed for me at least two crucial points:

The extent of public concern ought to be shaped by the scientific and economic knowledge about possible long-term climate change and its effects on, for example, farms, floods, sea levels, forest fires, ecosystems and tropical diseases.

But the extent of public concern is being shaped by the blurring of scientific and economic facts under the impact of political opinion, media miscommunication, and a debate among battling scientists themselves.

Scientists too often share responsibility with the media for not communicating complex science issues clearly to the public. Most members of the general public, as well as many officials in government, do not recognize that most scientists spend the bulk of their time arguing about what they don't know. Most scientists consider discussions of well-accepted, proven ideas as "old hat" and not worth our time. That attitude is not without merit, however, for the scientific method operates on the basis of constant questioning, particularly for those issues that are not yet well validated. But if the public and its representatives do not understand our process and its focus on not-yet-resolved issues, they will not easily be able to interpret what has been called the "dueling scientists" debate over global warming, regardless of whether the debating scientists are ideologically driven or not. We simply have to spend more time making clear the distinctions among (1) what is well known and accepted by most knowledgeable scientists, (2) what is known with some degree of reliability, and (3) what is highly speculative.

The public debate on global warming rarely separates those components, thereby leaving the false impression that somehow the scientific community is in overall intellectual disarray. In fact, the 15-year-old often-reaffirmed US National Academy of Sciences consensus estimate of 1.5 degrees C to 4.5 degrees C global average warming in the if CO2 were to double still reflects the best estimate from a wide range of current climate models (IPCC 1992) and ancient climatic eras (Lorius et al 1990). The earth has not been more than 1 to 2 degrees C warmer than now during the 10,000-year era of human civilization. The previous ice age, in which mile-high ice sheets stretched from Hew York to Chicago to the Arctic, was "only" 5 degrees C colder than the current 10,000 year old interglacial epoch we now enjoy. This 1.5-4.5 °C warming range still includes those studies that recently halved the "best guess" on warming from over 4 degrees C to 2.5 degrees C. Perhaps some new discovery next week will push it back up again, but even if not, that enduring 1.5 to 4.5 degrees C warming consensus still remains.

Changes of this magnitude could dramatically alter accustomed climatic patterns, affecting agriculture, water supplies, disease patterns, ecosystems, endangered species, severe storms, sea level, and coastal flooding.

Unless scientists communicate what they know along with what they don't know, the public policy process is subverted in an endlessly confusing debate that inadequately represents the actual nature of informed opinion. It is difficult for the media to do what sometimes I wish they would: back off of their concept of "balance" in favor of the concept of "perspective". If an issue is complicated, it simply is not enough to give equal inches or minutes to "all sides"-a practice which often leaves the public more confused that before, particularly if the "sides" that are left out are the middle: the bulk of experts, i.e., the people who created the established

consensus.

Moreover, that established consensus must be stated in terms of probabilities, as very few scientists, myself included, would say they believe the future climate "will" be in or out of the 1.5 to 4.5 degrees C warming range for certain. Rather, most believe this range to be reasonably probable. Therefore, if scientific opinion is to be communicated accurately, it must be by conveying issues in probabilistic terms and providing perspective on the range of views rather than by conducting an entertaining but misleading debate among the most extreme of the dueling scientists-or occasionally stretched beyond caricature in editorials or articles by polemicists and ideologues.

It is sad that the climate change debate recently has taken on a decidedly ad hominum, personal character, perhaps mirroring the lowering of the intellectual level of debate prevalent in so many political contests. The policy focus should not be, as the Detroit News once tried to argue, (Detroit News, 1989, Appendix A), on whether an individual scientist purportedly represents some complex issue fairly any more than whether some scientist believes the Detroit News to be biased (e.g., Schneider, 1989, Appendix A). Such personal attacks and defensive responses actually cloud, rather than clarify technically complex, controversial issues. What counts, then, is the nature of the evidence and the spectrum of opinions of a broadly representative