Alternative points of view and the problems of GCMs

The global warming scenario, in which a doubling of atmospheric CO2 would cause a temperature increase of 1.3-4.5°C, is widely accepted. This consensus has evolved from the results of many experiments with theoretical climate models which indicate the considerable potential for change in the earth/atmosphere system as concentrations of atmospheric greenhouse gases increase. Even the most sophisticated General Circulation Models (GCMs) cannot represent the working of the atmosphere exactly, however, and the limitations that this imposes on the results have long been a source of criticism. One of the earliest and most vociferous critics of the theoretical modelling approach was Sherwood Idso, who attacked the established view of future global warming through numerous publications (e.g. Idso 1980,

1981, 1982, 1987). He suggested that increasing CO2 levels would produce negligible warming, and might even cause global cooling. His conclusions were based on so-called natural experiments in which he monitored temperature change and radiative heat flow during natural events—such as dust storms, for example. From these he estimated the temperature change produced by a given change in radiation. Since the effects of increasing CO2 levels are felt through the disruption of the radiative heat flow in the atmosphere, it was therefore possible to estimate the temperature change that would be produced by a specific increase in CO2. Initially,

Idso (1980) suggested that the effect of a doubling of atmospheric CO2 would be less than half that estimated from the models. Later he concluded that increasing CO2 levels might actually cause cooling (Idso 1983). Idso received some support for his views (e.g. Gribbin 1982; Wittwer 1984), but for the most part, his ideas were soundly criticized by the modelling community, sometimes in damning terms (NRC

1982, Cess and Potter 1984). Although they initiated an intense—and sometimes acrimonious—debate on the methods of estimating climate change, in the long term Idso's natural experiments did little to slow the growing trend towards the use of GCMs. However, even with the growing sophistication of the models, those who used them were often the first to note their limitations (e.g. Rowntree 1990), and many of the criticisms of the estimated impact of elevated CO2 levels have arisen out of the perceived inadequacies of the models used (Kerr 1989).

Many of the problems associated with GCMs arise from their inability to deal adequately with elements that are integral to the functioning of the earth/atmosphere system. The roles of clouds and oceans in global warming are poorly understood, for example. The former are difficult to simulate in GCMs, in part because they develop at the regional level, whereas GCMs are global in scale. Parameterization provides only a partial solution (see Chapter 2), and the IPCC supplement has identified the problem of dealing with clouds, and other elements of the atmospheric water budget, as one of the main limitations to a better understanding of climate (Houghton et al. 1992).

Oceans create uncertainty in climate models mainly as a result of their thermal characteristics. They have a greater heat capacity than the atmosphere and a built-in thermal inertia which slows their rate of response to any change in the system. Thus, models which involve both atmosphere and ocean have to include some concessions to accommodate these different response rates. In representing the oceans it is not enough to deal only with surface conditions, yet incorporating other elements—such as the deep ocean circulation—is both complex and costly. As a compromise, many coupled ocean/ atmosphere climate models include only the upper, well-mixed layer of the oceans. These are the so-called 'slab' models in which the ocean is represented by a layer or slab of water about 70 m deep. While of limited utility in dealing with long-term change, this approach at least allows the seasonal variation in ocean surface temperatures to be represented (Gadd 1992). A more realistic representation of the system would require coupled, deep-ocean/atmosphere models, but their development is constrained by the limited observational data available from the world's oceans and the high demands that such models place on computer capacity and costs. Existing coupled models do provide results consistent with current knowledge of the circulation of the oceans, but they are simplified representations of reality, lacking the detail required to provide simulations that can be accepted with a high level of confidence (Bretherton et al. 1990).

GCMs also have difficulty dealing with feedback mechanisms which act to enhance or diminish the thermal response to increasing greenhouse gas levels (Rowntree 1990). Feedbacks are commonly classified as positive or negative, but in the earth/atmosphere system they may be so intimately interwoven that their ultimate climatological impact might be difficult to assess. For example, the higher temperatures associated with an intensified greenhouse effect would bring about more evaporation from the earth's surface. Since water vapour is a very effective greenhouse gas, this would create a positive feedback to augment the initial rise in temperature. With time, however, the rising water vapour would condense, leading to increased cloudiness. The clouds in turn would reduce the amount of solar radiation reaching the surface, and therefore cause a temperature reduction—a negative feedback—which might moderate the initial increase. Such complexities are difficult to unravel in the real world. Their incorporation in climate models is therefore not easy, but the importance of atmospheric water vapour feedback in climate change is well recognized by researchers (Ramanathan et al. 1983; Ramanathan 1988), and at least some of the mechanisms involved are represented in most current models.

Feedbacks associated with global warming are present in all sectors of the earth/atmosphere system, and some have the potential to cause major change. The colder northern waters of the world's oceans, for example, act as an important sink for CO2, but their ability to absorb the gas decreases as temperatures rise (Bolin 1986). With global warming expected to be significant in high latitudes there would be a reduction in the ability of the oceans to act as a sink. Instead of being absorbed by the oceans, CO2 would remain in the atmosphere, thereby adding to the greenhouse effect. On land the feedbacks often work through soil and vegetation. Increased organic decay in soils at higher temperatures would release additional greenhouse gases—such as CO2 and CH4—into the atmosphere, producing a positive feedback. This may be particularly effective in higher latitudes where the tundra, currently a sink for CO2, would begin to release the gas into the atmosphere in response to rising temperatures

Figure 7.13 289

Changing global annual surface temperature

Source: After Schneider and Mass (1975)

Note: The lower line between 1900 and 288

1990 indicates actual change, whereas the O upper line indicates the estimated temperature if the enhanced greenhouse effect is included. The temperature expressed in K is equivalent to temperature in degrees Celsius plus 273.15°C. It is possible that the difference has been caused by the cooling effects of increased atmospheric turbidity, which have 286 prevented the full nnn impact of the 1800

enhanced greenhouse effect from being realized including C02 effect






(Webb and Overpeck 1993). The additional flux of carbon from terrestrial storage might add as much as 200 billion tonnes of carbon to the atmosphere in the next century (Smith and Shugart 1993). In contrast, higher temperatures and more efficient photosynthesis in low and middle latitudes would initiate a negative feedback in which increased vegetation growth would cause more CO2 to be recycled and stored (Webb and Overpeck 1993).

Feedback mechanisms are incorporated in some form in most GCMs. However, the number and complexity of the feedbacks included, varies from model to model, and current modelling techniques continue to have difficulty dealing with them. Such constraints in the existing models must be recognized, and appropriate allowances made when predictions of global warming are used.

A basic concern among some researchers is the concentration on one variable—greenhouse gas levels—which has allowed the role of other elements in the system to be ignored. It is well-known that the earth's climate is not static, but has varied over the years (see e.g. Lamb 1977). Some of the variations have been major, such as the Ice Ages, whereas others have only been detectable through detailed instrumental analysis. Some have lasted for centuries, some only for a few years. While it is relatively easy to establish that climatic change has taken place, it is quite another matter to identify the causes. There are a number of elements considered likely to contribute to climatic change, however.

Since the earth/atmosphere system receives the bulk of its energy from the sun, any variation in the output of solar radiation has the potential to cause the climatic change. The links between sunspot cycles and changes in weather and climate have long been explored by climatologists (see e.g. Lamb 1977), and there are researchers who claim that solar variability has a greater impact on global climate than the greenhouse effect. For example, a report prepared for the Marshall Institute—a think-tank in Washington DC—suggested that reduced solar output in the near future might offset current global warming sufficiently to initiate a new Ice Age. The IPCC assessment considered this unlikely, however, pointing out that the estimated solar changes are so small that they would be overwhelmed by the enhanced greenhouse effect (Shine et al. 1990). Even if the solar energy output remains the same, changes in earth/sun relationships may alter the amount of radiation intercepted by the earth. Variations in the shape of the earth's orbit, or the tilt of its axis, for example, have been implicated in the development of the Quaternary glaciations (Pisias and Imbrie 1986).

The present intensification of the greenhouse effect is directly linked to the anthropogenic production of CO2; in the past, however CO2 levels have increased with no human contribution whatsoever. During the Cretaceous period, millions of years before the Industrial Revolution, CO2 concentrations were much higher than they are today (Schneider 1987). Other changes in the composition and circulation of the atmosphere have to be considered also. The impact of increased atmospheric turbidity is not clear. It may add to the general warming of the atmosphere (Bach 1976) but it has also been used to explain global cooling between 1940 and

1960, at a time when CO2 levels continued to rise (see Figure 7.13). Although this cooling is usually acknowledged as a problem by researchers, it has not yet been adequately explained (Wigley et al. 1986). More recently, the eruption of Mount Pinatubo in mid-1991 caused cooling which appears to have been sufficient to reverse the global warming of the 1980s and early 1990s. There is also some evidence that the turbidity increase caused by the eruption may have contributed to regional warming (see Chapter 5).

Thus there are many elements in the earth/ atmosphere system capable of producing measurable climate change. Given their past contribution to change, it seems unlikely that they will now remain quiescent while anthropogenic CO2 provides its input. Despite this, they have been ignored by most researchers or are considered of minor importance compared to the potential impact of the greenhouse gases (Roberts 1989).

Research into global warming is continuing at a high level, and it is possible that a better understanding of its interaction with other elements in the earth/atmosphere system will emerge to resolve some of the existing uncertainties. If not, society will have to deal with the future environmental changes in much the same way as it has done in the past—by reacting to them after they have happened.

The Basic Survival Guide

The Basic Survival Guide

Disasters: Why No ones Really 100 Safe. This is common knowledgethat disaster is everywhere. Its in the streets, its inside your campuses, and it can even be found inside your home. The question is not whether we are safe because no one is really THAT secure anymore but whether we can do something to lessen the odds of ever becoming a victim.

Get My Free Ebook

Post a comment