Page images
PDF
EPUB

stronger and continues to increase and produces increasing warming.

Now, if we look at the same model calculations and look at their projection into the future we can-this is the tool that we use to predict how climate will change in the future. Now, I've plotted here the results of these model calculations up to the Year 2070. And you can see, the red line is the model with CO2 only, and the blue line is with CO2 and aerosols which does a better job of reproducing the current climate, and so we have more confidence in that prediction. And it indeed shows that there will be less warming in the future than we originally thought from our calculations with just CO2.

But, there will still be a tremendous amount of warming into the next century. It shows by the Year 2050, 4 or 5 degrees Fahrenheit. So there is less warming than we originally thought, but still quite a bit, to a climate much warmer than ever before experienced by our species and at a much more rapid rate.

These models include all the physics of the climate system as best as we understand them and as best we can put them into a model. They include the best representation that we know of how to do water vapor and clouds as Dr. Spencer mentioned, and indeed they're not perfect, but this is the best tool that we have. And we need to improve these tools. We also need to do calculations including other things like the indirect effect of aerosols which cause the change the reflectivity of clouds to change; or the effect of carbon aerosols which would produce warming. But, based on the latest result just published this year by Haywood et. al., this makes me think this is the best that we know now and it shows a good reproduction of the present climate and a prediction into the future.

What would be the now, it's also possible that there are surprises in the future. There have been rapid climate changes in the past as the ocean circulations shifts. Wally Broecker has suggested this might happen in the future, but surprises by definition are unknown, I mean, we're surprised, we don't know what would happen. The ozone hole is a good example of this. Nobody predicted that there would be this rapid decrease of ozone over the South Pole in October. Molina and Rowland talked about a small gradual change in ozone, but this effect, which includes reactions on clouds, was unpredicted. We were caught unaware, and now we understand it.

So, we don't know what surprises there might be in the future. There might be some rapid change; some rapid cooling. But, in the current climate regime that we're in, these things don't happen, but if we push the climate into a warmer regime, there might be some surprises.

As, you know there have been a number of possible impacts of climate change suggested by these same models, including: drought in the mid-latitude, stronger storms, sea-level rise, spreading of deserts, increase fire frequency and intensity.

The weather today was quite warm for October, but it's not evidence of global warming because it's just part of the weather patterns. But, perhaps it's something that might happen more often in a greenhouse-warm world. But, you can't say that the weather today proves global warming, it just shows that the weather patterns are what they are today. For example, the weather today is not a record high temperature, so it's been warmer than this in the past, but it's certainly an example of the things that might happen more often. And the probability of this happening more often is certainly supported by these model results.

I wrote down what we should do. I'll just mention briefly, I think, based on my experience, that climate's going to change no matter what the United States does about greenhouse gases. We can modify a little bit the future climate but we have to learn to adapt to climate change. And that requires more research into how the climate will change, what the patterns will be, which we don't understand very well. And if we think that rapid climate change is bad for society, then we should think about doing mitigation. But what we do to change the climate, to change our input of greenhouse gases, is a complex political decision which I'm not an expert in.

That concludes my testimony. [The prepared statement and attachments of Mr. Robock follow:)

Global Warming: State of the Science
Testimony before the House Committee on Science
Subcommittee on Energy and Environment

October 7, 1997

Professor Alan Robock

Department of Meteorology

University of Maryland
College Park, Maryland 20742
E-mail: alan@atmos.umd.edu

Phone: 301-405-5377

Scientific Consensus on Global Warming

I agree with the conclusions of the 1995 IPCC Working Group I report (Houghton et al., 1996) that "the balance of evidence suggests that there is a discernible human influence on global climate." Note that this is the balance of evidence, NOT unambiguous proof. The report points out that "our ability to quantify the human influence on global climate is currently limited because the expected signal is still emerging from the noise of natural variability, and because there are uncertainties in key factors. These include the magnitude and patterns of long term variability...." [Both these quotes are from p. 5 of the Summary for Policymakers.] I agree with this part of the assessment, too.

The evidence which supports a human influence on climate includes observations that the concentrations of "greenhouse gases" produced by human activity, especially carbon dioxide, are increasing and that these gases warm the surface by enhancing the natural greenhouse effect. These facts are undisputed. But these gases are not the only cause of climate change. When the most recent climate model experiments, done since the latest IPCC report, include the effects of greenhouse gases, aerosols (particles in the atmosphere), volcanic eruptions, ozone depletion, solar variations, and El Niño in their calculations, they produce simulations of climate change of the past 100 years that agree quite well with the past surface temperature record. For example, Haywood et al. (1997) describe calculations made with the climate model of NOAA's Geophysical Fluid Dynamics Laboratory (GFDL) at Princeton University. When they attempt to simulate the climate change of the past 130 years taking into consideration just the effects of CO2 increases, the model produces too much warming as compared to observations (Figure 1). When they now also include the direct effects of tropospheric sulfate aerosols (haze), their simulation almost exactly matches the past climate change (Figure 2). The enhanced industrialization following World War II rapidly increased aerosol and CO2 output. But the average aerosol lifetime is only 4 or 5 days as compared to the 100-year lifetime of CO2. Therefore the total aerosol effect is felt immediately, retarding for a couple decades the anthropogenic warming. This is an example of more realistic results from climate models produced by using more realistic scenarios of climate change.

In addition to the above results, there are many other examples of phenomena explained and predicted by climate models that we can observe, demonstrating the fingerprint of human causation of climate change. Observations show that stratospheric temperatures are decreasing, sea level is rising, sea ice is retreating, snow cover is decreasing, and glaciers are melting, all in agreement with these calculations. Stratospheric cooling can be well simulated by observed CO2 increase and ozone depletion. In the stratosphere the human impact is much easier to detect, as it has much less weather variability to disguise the human signal. Near the surface, weather noise is much larger and includes the effects of El Niños, so the greenhouse warming is harder to detect. Nevertheless, it is impossible to explain the warming of this century in any other way. Solar variations are speculative and much too small. Even the temporary cooling after the 1991 Pinatubo volcanic eruption, the largest of the past 250 years, was not enough to cool the climate to 19th century levels.

The disagreement between satellite (MSU - Microwave Sounding Unit; Spencer et al., 1990) and surface temperature records (Jones et al., 1986a, b, updated) for the past 20 years are well understood and do not argue against global warming. In fact, if the climate changes caused by the 1982-83, 86-87, and recent El Niños, ozone depletion, and the 1982 El Chichón and 1991 Pinatubo volcanic eruptions are accounted for, the MSU record shows a warming during this period, in agreement with climate model simulations.

It is the same models described above that we use for projections of future climate. Figure 3 shows the GFDL model predictions until near the end of the next century. While the predicted warming is less when including the more realistic effects of aerosols, significant warming is still predicted from now on. The additional CO2 being put in the atmosphere is already overwhelming the aerosol effect. These calculations show global average warming of more than 7°F by the end of the next century with "Business as Usual." Other model experiments described in the 1995 IPCC report forecast that the global average temperature will rise by 2 to 6°F and sea level will rise by 1 to 3 feet by the end of the next century. Even for the smallest increase projected, "the average rate of warming would probably be greater than any seen in the last 10,000 years." (p. 6 of the IPCC Summary for Policymakers.] The IPCC goes on to say, "actual annual to decadal changes would include considerable natural variability. Regional temperature changes could differ substantially from the global mean value." This means that certain regions of the globe are apt to warm much more than the global average. Some areas would not warm much at all or could even cool slightly. We do not now understand where each of these regions will be.

Surprises are also possible, including rapid warming once the climate has passed a certain threshold, by a mechanism previously unknown. This is exactly what happened when the ozone hole suddenly appeared over the South Pole during the previous decade. It was not predicted and came as a complete surprise. Broecker et al. (1985) have suggested possible rapid shifts in ocean circulation with resulting rapid climate change on land in certain regions, but surprises, by definition, cannot now be described or quantified.

It is important to point out that many human impacts of climate change depend on climate elements other than temperature. For example, "ocean currents, frequency and strength of oceanic storms, winds, frequency of fog, sea-ice distribution and thickness, and sea level will all be important for impacts on ocean transportation. For agriculture, temperature, precipitation, cloudiness, wind, carbon dioxide concentration, intensity of ultraviolet light, frequency of severe storms, and soil moisture" (Robock, 1993, p. 301) will all be important.

Table 1. Areas of Human

Endeavor That Could be
Affected by Global Warming

Agriculture

Forestry

Fisheries
Water Resources
Human Health

Insurance

Wetlands
Rivers and Lakes

Estuaries

Air Pollution
Land Transportation

Air Transportation
Ocean and River Transportation

Recreation

Tourism

Biodiversity
Electricity Demand
Wind Energy Generation

Solar Energy Generation
Hydroelectricity Generation

Societal Systems
Political Systems

In order to determine the net impact of greenhouse warming on humans, we must first examine each potential area of impact (Table 1). Then for each activity, we must determine the future changes of the mean, variability, and extreme values of each important element for that activity, for every region of the earth where it would have an impact, for all times of the year. If this cannot be done, then we must at least determine the sensitivity to each element by varying it over a range of possible values. Next, we must evaluate the direct impact of climate change on the activity, taking into consideration future technological, sociological, economic, political, and military responses to each impact, singly, and in all combinations. Finally we assign probabilities to each choice and result, and determine the net human impact (Robock, 1993). It is unlikely that this will be done with any confidence for the foreseeable future, say the next 10 or 20 years. In light of this, what can we say about impacts of global warming on society? What does the latest research tell us?

46-495 - 2

« PreviousContinue »