Greenhouse: Planning for Climate Change

By CSIRO PUBLISHING

It is important for the reader to understand clearly the objectives of these papers. They are not an attempt to provide accurate predictions of what is going to happen in Australia over the next few decades. Rather they represent sensitivity studies, designed to illustrate to what extent we as a nation are dependent on the climate and likely to be affected by climatic change, and attempts to develop the techniques for such sensitivity analyses. For this, the climate scenario (reproduced in the Appendix to this volume), was a key.

• Language: English
• Publisher: CSIRO PUBLISHING
• Release date: Jan 1, 1988
• ISBN: 9780643105744

Book preview

Background to the greenhouse effect

Greenhouse gases: evidence for atmospheric changes and anthropogenic causes

G.I. Pearman

Abstract - The last decade has seen an enormous growth in our knowledge of the trace-gas composition of the global atmosphere. Now, after extensive national and international research on this topic, it is becoming clear that the atmosphere demonstrates a chemical weather and climate as complicated as the physical weather and climate. We now understand that the chemical climate is changing, particularly with respect to those gases which we call greenhouse gases.

This paper highlights some of the evidence that greenhouse gases have increased in concentration and are likely to continue to do so in the coming decades. This evidence forms the basis upon which estimates of the climatic response to these changes can be made.

The key greenhouse gas species which are identified are carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), tropospheric ozone (O3) and several halocarbon species. Since preindustrial times, CO2, CH4 and N2O concentrations have risen by 23,110 and about 8% respectively. Current rates of increase are 0.4, 1.0 and approximately 0.3% y-1 with estimates of increases of 45-115, 200-500 and 25-60%, respectively, above preindustrial levels over the next 50 years.

The evidence that tropospheric ozone is increasing is not convincing even though there are reasons to believe that it may increase by up to 50% in the future. The key chlorofluorocarbon gases CCl3F (CFC-11) and CCl2F2 (CFC-12) are both increasing at rates of about 5% y-1 having been none existant in the atmosphere prior to the 1930s. They are expected to increase by about 300% in the next 50 years. Several other such man-made gases are also expected to make a contribution on these time-scales.

Key word index: greenhouse gases, carbon dioxide, methane, nitrous oxide, ozone, CFCs, atmospheric trends.

1. INTRODUCTION

Over 100 years ago it was proposed that changes in the level of carbon dioxide (CO2) in the atmosphere might have been the cause of geological variations of climate. CO2 is a naturally occurring atmospheric gas which has the properties of being transparent to solar radiation, but somewhat opaque to terrestrial infrared radiation (heat). This gas and other gases with similar properties, often referred to as greenhouse gases, tend to cause the surface of the earth to be warmer than it otherwise would be. The argument was that any event that tended to change the levels of CO2 in the atmosphere might change the global climate.

Not until the middle of this century did scientists start to suggest that CO2 released into the atmosphere during the combustion of fossil fuels might induce a man-made global warming. At that time insufficient was known about the global carbon cycle (the exchange of carbon between the atmosphere, oceans and biosphere), to estimate if the CO2 released would remain in the atmosphere. Nor were there observations to tell if CO2 was accumulating in the atmosphere. Late in the 1950s precise measurements were commenced, so that during the 1960s, it became clear that the level of CO2 in the atmosphere was increasing, and that it was likely to continue to do so. By the end of the 1970s, an international research effort was underway to address the many uncertainties in the various assumptions linking the release of this gas into the atmosphere and its impact on global climate.

Throughout the early 1980s this effort was intensified and has led to a better understanding of the carbon cycle and the likely future levels of CO2 in the atmosphere. In addition, it was discovered that other radiatively important gases, including methane (CH4), nitrous oxide (N2O), and the chlorofluorocarbons (CFCs) were increasing as well. Estimates of the combined effects of these additional gases on the temperatures of the lower atmosphere suggested that they might contribute as much as CO2 alone, thus leading to a greater sense of urgency about the potential for major climatic change on relatively short time-scales.

These studies are continuing, but key reviews were completed in late 1985 (DOE, 1985 a,b,c,d; Bolin et al., 1986). This paper presents some of the current evidence for atmospheric changes and attempts to outline the kinds of changes which are expected over the next 50 years.

2. CARBON DIOXIDE

Trends

The fact that CO2 is increasing in concentration in the global atmosphere is now well established. Indeed, late last year and early this year marked the 30th anniversary of high precision measurements of this gas in Antarctica and Hawaii, respectively. Over Australia we have been observing the increase (Figure 1), which is currently about 1.5 ppmv y-1 (0.4% y-1), for 15 years.

Only recently have we been able to extend this record backwards in time through the analysis of air trapped in the Antarctic ice sheet (Pearman et al., 1986; Etheridge et al., 1987). These studies and similar studies in Switzerland (eg. Friedli et al., 1984, 1986) show that preindustrial and pre-agricultural levels were about 285 ppmv, or about 23% lower than 1987 concentrations (Figure 2).

Indeed, the most recent ice-core results show that the major temperature variations of the last 160000 years were accompanied by significant changes in CO2 concentrations (Figure 3; Barnola et al., 1987). While such correlations do not prove the past occurrence of CO2-induced climatic change, it appears likely that CO2 variations formed an integral link with climate changes of the last ice-age (Genthon et al., 1987). This recent evidence leads us back to the earlier suggestion of a controlling influence of CO2 on the global climate.

Figure 1. Atmospheric CO2 measured in the mid troposphere over southestern Australia. Updated from Pearman and Beardsmore (1984).

Figure 2. The changes of atmospheric CO2 concentration for the past few centuries as revealed in air extracted from Antarctic ice. Data are from the following sources: and , Frledli et al.(1984) and (1986) respectively; , Neftel et al.(1985); , Etheridge et al.(1987), horizontal bars represent the range of gas ages of the samples, vertical bars are the 1 sigma limits of the data represented by the mean; o, unpublished data from G I Pearman and D Etheridge (Australian Antarctic Division). Full line represents modern Antarctic measurements (Bacastow and Keeling, 1981; Komhyr et al., 1985).

Figure 3. Temperature and CO2 variations for the past 160 000 years. a, surface ice temperature variations in Antarctica (Barnola et al., 1987); b, global annual mean surface temperature (Jones et al., 1986); c, CO2 variations from ice-core analysis (Barnola et al., 1987); d, CO2 variations from ice-core analysis (Pearman et al., 1986), and, e, modern Antarctic CO2 observations (Bacastow and Keeling, 1981; Komhyr et al., 1985).

Carbon budget

Each year about 5 Gt of carbon is released into the atmosphere due to the combustion of fossil fuels and, to a small extent, to the manufacturing of cement. This carbon, as CO2, is transported throughout the global atmosphere on time-scales of about 1 year, is exchanged with the biosphere through the photosynthetic and respiratory processes, and is exchanged with the oceans. The rate of increase of CO2 in the atmosphere, therefore, depends on the rates of exchange of CO2 with these other reservoirs and their capacity to accumulate carbon. Table 1 shows the sizes of each of these reservoirs. Clearly if each reservoir was to take up the CO2 released in proportion to its size, then the increase in the atmosphere due to fossil fuel combustion would be very small. However, in the case of the oceans, two factors markedly reduce their capacity to do so. The first is the chemical buffering factor of the carbonate system which means that the rise in concentration of total carbon in the water at equilibrium will be about one order of magnitude less than the rise in atmospheric CO2 concentration. The second is the fact that the oceans are poorly mixed. The time-scale for the complete mixing of the global oceans is of the order of 1000 years. It is only the upper layers of the oceans that can participate in the sequestering of the fossil fuel CO2 on time scales of decades. The upper 50 - 100 m of the ocean are well mixed, while below that layer, the thermocline region is less well mixed. In fact, the oceanic reservoir available for taking up CO2 on these times-scales, is approximately the same effective size as that of the atmosphere. It is for this reason that the current rate of increase of atmospheric CO2 is approximately half (in fact 57%) of that which would occur if all the CO2 released were to remain airborne.

TABLE 1. A summary of the size of the global carbon reservoirs and the rates of exchange of carbon between them. The sources for these data are numerous, but access to the relevant literature may be obtained through Bolin et al. (1986) and Pearman and Hyson (1986).

This statement presupposes that the biosphere plays no part in either contributing to the increase in atmospheric CO2, or removing any of the fossil fuel contribution. This fact has been debated at length over the past decade. In the mid and late 1970s, biologists claimed that global deforestation, particularly in the tropical regions, was a net source of CO2 to the atmosphere at least as large as the fossil source. Recent research has tended to indicate a much smaller contribution from this source. This work includes:

Refinements to the estimates of the rate of deforestation have led to a generally lower estimate of the rate of CO2 release from that source (eg. < about 2 Gt y-1; Bolin et al., 1986; Houghton et al., 1983, 1985, 1987).

Ice-core observations of the ¹³C/¹²C ratio and concentration of atmospheric CO2 in the past when compared with modern observations do not support the concept of any major additional source of CO2 of biological origin (eg. Freidli et al., 1986; Siegenthaler and Oeschger, 1987; Enting and Pearman, 1987; Oeschger and Siegenthaler, 1988).

Transport models of the meridional distribution of CO2 do not support the presence of a major (> 2 Gt y-1) equatorial source, and indeed may even suggest the presence of a net biospheric sink in the northern mid to high latitudes (eg. Pearman and Hyson, 1986).

Enting and Mansbridge (1987a) proved that there was no possible linear steady-state ocean model that would reconcile the ice-core CO2 data of Neftel et al. (1985) and the biotic source estimates of Houghton et al. (1983).

Inspection of Table 1 will show that while the global increase in atmospheric CO2 concentrations relates to an accumulation of about 2.9 Gt y-1, this is a very small percentage of the total amount of carbon annually circulated both between the atmosphere and the biosphere and the atmosphere and the oceans. It remains a feature of the global carbon cycle that these exchange processes are nearly in balance and appear to have been reasonably so over the past two centuries at least.

Future CO2 levels

Predictions of future CO2 concentrations are usually performed using global carbon cycle models (see eg. Siegenthaler and Oeschger, 1987; Enting and Pearman, 1987). These can include extra detail about the distribution and isotopic composition of the excess carbon which is useful in calibrating and validating such models. It is however, possible to give a much simpler description of the predictions of such models (Enting and Mansbridge, 1987b). Given that the greatest uncertainties for the next 30-50 years probably lie in what will be the economic factors determining the growth of energy usage, the sophistication of such models is not that important. Carbon dioxide concentrations, C(t) in ppmv, are related to fossil fuel plus biotic sources, S(t), in Gt y-1 by the relation,

The response function R(t) decribes the oceanic uptake of CO2 and can be approximated as

This approximation describes the box-diffusion model as used by Siegenthaler and Oeschger (1987). An even simpler and more commonly used approximation is to assume a constant airborne fraction. This is valid for a range of continually growing source scenarios but tends to break down in other cases.

It is beyond the scope of this paper to attempt to discuss the considerable uncertainties and difficulties in making estimates of the source function S(t). This is discussed at length by Keepin et al. (1986) who attempt to account for such things as the future global demands for energy and the extent to which fossil fuels will be utilized. Examples of other considerations of future energy demands are found in Edmonds and Reilly (1983a,b), Perry et al. (1982), Reister and Rotty (1983) and Rotty and Reister (1986). For our present purpose, I have chosen to illustrate the impact of particular scenarios on the CO2 concentration by assuming that fossil fuel use grows,

At the rate of 2.3% y-1 until 2050, regarded as an upper limit scenario by Keepin et al. (1986).

At the current rate of 1.5% y-1 until 2050.

At the current rate of 1.5% y-1 until 2000 and then remains static until 2050.

It should be remembered that CO2 release rates grew quite steadily at about 4.3% y-1 following the last world war until the energy crisis of 1974. Using Equation 1 and these scenarios one obtains the growth of CO2 in the atmosphere as shown in Figure 4.

These scenarios suggest that the CO2 concentration is likely to double sometime in the next 50-75 years. At least during the next few decades, the uncertainties of these predictions are influenced most by economic factors: exactly how energy consumption will grow? Thus increases of 50% are most likely in the next 30-50 years. Beyond that time, other factors may become important. These include the fact that higher CO2 levels may lead to a net growth of the biota, and climatic change due to the greenhouse effect means that it is most unlikely that the world and thus carbon storage will remain unchanged. The possibility cannot be excluded that on these time scales technological breakthroughs in the use of alternative fuels and reviewable energy sources may reduce the demands for fossil fuels.

Further, the capacity of the oceans to sequester fossil fuel CO2 depends on their current dynamical state. With significant changes in the climatic conditions, this will not necessarily be the same (Broecker, 1987). It is in these two areas that research of the next decade must be concentrated.

Figure 4. Predicted growth of atmospheric CO2 concentration given the following assumptions concerning the annual growth in the rate of net release of CO2: 1, 2.3% y-1 until 2050; 2, 1.5% y-1 (current growth of fossil fuel CO2 release) until 2050; 3, 1.5% y-1 until 2000, and thereafter no further growth in the rate of release.

3. METHANE

Trends

Unlike CO2, high precision observations of CH4 were not commenced on a regular and inter-calibrated basis until about 10 years ago. But within a few years it became clear that CH4 was increasing in the atmosphere at about 1% per year (Figure 5; Khalil and Rasmussen, 1983a; Blake and Rowland, 1986; Fraser et al., 1986; Steele et al., 1987). Recently, ice-core studies in several laboratories have shown that CH4 has increased since pre-agricultural times from about 775 ppbv to a 1987 level in Antarctica of about 1614 ppbv, or 110% (Figure 6; Rasmussen and Khalil, 1984; Stauffer et al., 1985; Pearman et al., 1986; Pearman and Fraser, 1988; P. Fraser, pers.comm.).

Methane budget

The reason for the past and current trend in CH4 concentrations is less well understood than the CO2 increase. There are a number of sources of atmospheric CH4 (Table 2), many of which are related to human activities, and might explain the increase. The CH4 increase might also possibly relate to decreasing atmospheric levels of the hydroxyl radical (OH; caused itself by the increase of CH4), and increasing concentrations of carbon monoxide (Khalil and Rasmussen, 1985; Thompson and Cicerone, 1986).

The major (∼85%) sink for CH4 is reaction with OH. The global level of OH can not be measured directly. However, it can be determined indirectly by knowing the rate of release of the man-made gas, methylchloroform (CH3CCI3), which is also destroyed in the atmosphere by reaction with OH, and for which the atmospheric abundance is known. Such determinations indicate a global sink strength for CH4 of 400-600 Tg y-1 (Fraser et al., 1986; Prinn et al., 1987). The current global concentration of CH4 of about 1600 ppbv implies a global abundance of about 4300 Tg and thus an average atmospheric life-time of 7-11 years.

Figure 5. The observed increase of atmospheric methane at the Cape Grim Observatory, Tasmania, updated from Fraser et al.(1986, 1987a).

Figure 6. Methane variations in the past few centuries From; ▲, Etheridge et al. (1987);●, higher quality unpublished data of D Etheridge (Australian Antarctic Division) and G I Pearman and P J Fraser (CSIRO); , modern measurements from the Cape Grim Observatory, Tasmania, Fraser et al. (1987a). Vertical and horizontal bars as for Figure 2.

TABLE 2. A summary of the major global atmospheric sources of CH4 (based on Bingemer and Crutzen, 1987; Crutzen et al., 1986; Ehhalt, 1987; Mathews and Fung, 1987; Sebacherer et al., 1986; Seiler and Conrad, 1987; Seiler et al., 1984).

The range of total sources of CH4 shown in Table 2 covers that of the estimated range of uncertainty for the sink strength, although clearly the lower limit of the total source estimates is insufficient to balance the lower limit of the estimated sink strength. Table 2 also shows that the release of CH4 from fossil sources (coal and gas fields) amounts to to 5-23% of the sink strength. Recently, Lowe et al.(1988) have shown that the radiocarbon content of atmospheric CH4 suggests that about 32% comes from fossil sources. These results may indicate that the fossil sources are somewhat larger than indicated here, or that some proportion of the Tundra source is CH4 which is from old carbon (Pearman and Fraser, 1988).

Future methane levels

With these uncertainties, it is not yet possible to predict, with due consideration of the causes of the past and present perturbations, what future levels of CH4 may be in the global atmosphere. However, given that for more than a century increases have paralleled population growth, it is reasonable to assume that this may continue for at least a few more decades. Table 3 gives the predictions of Dickinson and Cicerone (1986) showing that by the middle of the next century CH4 may have increased by between 25 and 135%. These authors have taken the upper limit to be based on the extrapolation of the current growth slightly increased because of the likelihood of some reduction of OH levels. The lower limit assumes that the current growth continues for 20 y, but thereafter there is no further growth.

TABLE 3. Current and estimated future levels (2050) of greenhouse gases in the global atmosphere together with estimates of the additional energy flux into the troposphere ( Q). Based on the data from Dickinson and Cicerone (1986). The predictions (to the year 2030) of surface temperature increases are from Ramanathan et al. (1985) who calculated similar future changes for each of the greenhouse gases as Dickinson and Cicerone (1986).

4. OZONE

Depending on the latitude, the ozone (O3) in a vertical column of the atmosphere shows a peak of concentration in the middle and lower stratosphere (15-35 km). This region of maximum concentration is referred to as the ozone layer. Recently it has become clear that the O3 in this layer in the region over Antarctica has shown a marked decrease in concentration particularly in the spring months (NASA, 1985; Farman et al.,1985; Cicerone, 1987). Unpublished data show that 1987 spring levels over Antarctica were about 50% less than they were 10 years ago. Unusually high levels of stratospheric chlorine and the unique Antarctic environment seem to be the cause of this dramatic loss of O3 (NASA, 1988).

The evidence for a decrease in O3 concentrations throughout the global O3 layer is less convincing, but appears to show a 2-3% loss over the last decade, more than half of which is due to changes in solar activity (NASA, 1988). O3 in this layer is the cause of the thermal stability of the upper atmosphere and thus has an effect on the dynamics of the atmosphere as a whole.

Ozone in the troposphere, while at much lower concentrations than in the stratosphere, is particularly effective as a greenhouse gas because of the pressure broadening of the infrared absorption lines at the higher atmospheric pressures. There are good reasons to suspect that concentrations of tropospheric O3 might be increasing. Tropospheric O3 originates from the mixing downwards of stratospheric air and from local photochemical production. The precusors for this production include CO and CH4 which are being released into the atmosphere in increasing amounts, particularly in the northern hemisphere. However, the lifetime of tropospheric O3 is short (several weeks) and therefore there is a great deal of spatial variability in the concentration, making the detection of trends much more difficult than for the longer-lived species such as CO2 and CH4. The evidence for tropospheric increases in the southern hemisphere is effectively non-existant. Single records such as the one at the Cape Grim Observatory can not be interpreted as being globally or even hemispherically representative. The situation in the northern hemisphere is clearer. Studies of observations made in the latter part of the last century (Bojkov, 1986; Volz and Kley, 1988) and of modern records (Logan, 1985) suggest that, at least for certain regions of the northern hemsphere, some increases have occurred.

Future tropospheric ozone levels

It appears likely that there will be some increase in tropospheric O3 concentration during the next few decades, although exactly how much is very uncertain. Dickinson and Cicerone (1986) have argued that with the upper estimates of fossil fuel combustion, one might expect a 60% increase of tropospheric O3. Given that upper tropospheric and stratospheric O3 is expected to decrease (because of chlorofluorocarbons, CFCs), an increase averaged through the troposphere of up to 50% of current levels might be expected, although a much smaller change is possible.

5. NITROUS OXIDE

Trends

As with CH4, precise atmospheric measurements have identified a trend in N2O concentration only in the last decade (Weiss, 1981; Khalil and Rasmussen, 1983b; Fraser and Derek, 1987; Prinn et al., 1988), although the trend is less precisely defined than for the other greenhouse gases (Figure 7). These studies, together with the 5-station records of the US NOAA-GMCC baseline observatories (Schnell and Rosson, 1987) suggest a current rate of increase of between 0.2 and 0.3% y-1.

Ice-core measurements show a preindustrial level about 8% lower than at present (Pearman et al., 1986; Etheridge et al., 1987; Khalil and Rasmussen, 1987; Zardini et al., 1987). Although the historical changes are not very precisely determined, the indication is that much of the change has occurred in the last four decades.

Budget

The current average atmospheric content of N2O (307 ppbv) is about 1.5 Gt of nitrogen, with the major sink being photolysis in the stratosphere estimated at 6-10 Tg y-1 (Crutzen and Schmailzl, 1983) and implying a residence time of about 170 years. To balance this sink and the observed increase (3-5 Tg y-1), the global sources must total 9-15 Tg y-1.

The preindustrial level of N2O was about 280 ppbv, so that at that time the natural source of N2O, in balance with the stratospheric sink, must have been 5-9 Tg y-1. The emission ratios of N2O/CO2 from coal, and oil combustion are about 2 × 10-4 (Weiss and Craig, 1976; Hao et al., 1987), so that the N2O source from the combustion of fossil fuels must be about 2 Tg y-1(20-40% of the natural source). To this should be added an additional source of perhaps as much as 1-2 Tg y-1 due to biomass burning. A further additional source might be the loss of N2O from nitrogen fertilizers, estimated to be of the order of 1-2 Tg y-1 (Bolle et al., 1986). Thus the current increases in atmospheric N2O are possibly due to releases during the combustion of fossil fuels, the use of nitrogen fertilizers, with some contribution from biomass burning.

Future nitrous oxide levels

Clearly, without a better understanding of the relative contribution of these various sources, the reliable prediction of future levels of N2O is difficult, if not impossible. However, it does appear that fossil fuel combustion is a significant component of the current increase, and that contribution is likely to continue to grow at least in the next few decades. Because of the long lifetime of this gas, even if the growth of the anthropogenic releases was curtailed, the atmospheric concentration would continue to grow into the next century. Accepting these uncertainties, Ramanathan et al. (1985) anticipate a level of 350-450 ppbv by the year 2050, a rate of growth accepted by Dickinson and Cicerone (1986) although the upper limit is regarded as un-realistically high by Bolle et al. (1986).

Figure 7. Atmospheric N2O measured in surface air at the Cape Grim Observatory, Tasmania. Updated from Fraser and Derek (1987) and in a preliminary calibration scale (P J Fraser, pers. comm.).

6. CHLOROFLUOROCARBONS

Trends

In the contemporary atmosphere, the two most significant CFCs with respect to the greenhouse effect, are trichlorofluoromethane, CCl3F (CFC-11) and dichlorodifluoromethane, CCl2F2 (CFC-12). Both of these gases were not present in the atmosphere prior to the 1930’s when their first anthropogenic use commenced. They are used as refrigerants, aerosol propellants and in the foaming of plastics. One of their commercially desirable qualities is their chemical inertness, a property which leads to their rapid accumulation in the atmosphere and the prospect of their presence in the atmosphere decades to centuries after their release. International measurement programs over the past decade (see for example Prinn et al.,1983; Cunnold et al., 1986) have established the global levels and rates of change of these and several other significant or-ganochlorine compounds such as CHClF2 and CCl2FCClF2(Fraser et al., 1987b), CH3CCl3 (Prinn et al., 1987) and CCl4 (Simmonds et al., 1983, 1988). Figures 8 and 9 show the changes in CCl3F and CCl2F2 as observed at the Cape Grim Observatory.

Relationship to the ozone issue

It is these and some other halogenated gases that are strongly implicated in the seasonal depletion of O3 over Antarctica and potentially for the weakening of the O3 layer in general (Cicerone, 1987). As pointed out earlier, O3 is significant in the dynamics of the atmosphere. Thus these gases may have an important impact both through their greenhouse warming effect in the troposphere, where they strongly absorb in the infrared atmospheric window, and in the cooling of the stratosphere due to the depletion of O3.

Wide publicity given to the O3 layer depletion has impacted on the growth of some of these gases in the atmosphere. For example, prior to 1974, when the first alarms were raised about the potential for O3 depletion by CFCs (Molina and Roland, 1974), the rate of production of CCl3F and CCl2F2 was increasing by about 15% y-1. Since that time there has been little change in the release rate, leading to a more constant growth rate in the atmosphere of about 5% y-1. During 1987 an international agreement (the Montreal Protocol; Johnson, 1987) was introduced because of concern about global O3 depletion. It aims at reducing the usage of these gases by industrialized nations by 50% by the year 2000.

There are some feedbacks between greenhouse and ozone problems and there are some common features of both, eg. the CFCs and climate. But, in spite of common misconceptions in the community at large, the issues are largely distinct.

Figure 8. Trichlorofluoromethane (CFC-11) concentrations as observed at the Cape Grim Observatory (data from Fraser and Derek, 1987; updated by P J Fraser, pers.comm.).

Figure 9. Dichlorodifluoromethane (CFC-12) concentrations as observed at the Cape Grim Observatory (data from Fraser and Derek, 1987; updated by P J Fraser, pers.comm.).

Future CFC levels

The prediction of future levels of the several significant halogenated gas species in the atmosphere depends on understanding their lifetimes (now reasonably well established), the likely acceptance, development and impacts of the Montreal Protocol, the successful discovery and introduction of alternative products and the extent of recycling of existing products. Relatively complex models have been used to make such predictions, but, as for CO2, the behaviour of such models can be described by simple approximations. Enting (1985) expressed the behaviour of the model used by Fraser et al. (1983) using an analytic approximation for the response of the model to inputs of CCl3F. This makes it possible to calculate the stratospheric and tropospheric inventories as a function of time for any proposed scenario such as a continuation of past growth or full or partial compliance with the Montreal Protocol (see above). For example, if the growth in CCl3F release is finally halted so that releases continue at some constant rate, then Enting’s approximation indicates that the steady-state inventory of atmospheric CCl3F will correspond to 65.8 years of release (9.1 years of release in the stratosphere and 56.8 years release in the troposphere). This means that a freeze at say 1986 levels of release will lead to concentrations late next century of approximately three times present levels. Note that such estimates are considerably lower than some of those made prior to the introduction of the Montreal Protocol (see the more extreme upper predictions in Table 3).

7. CONCLUSIONS

A little more than a decade ago, our knowledge of the chemical composition of the atmosphere was quite poor. Many questions remain unanswered, but following a decade of active growth in the field, we can now see that the atmosphere features chemical weather and climate as much as physical weather and climate. What is more, we know that the chemical climate is changing, that mankind is strongly implicated in these changes, and that the composition of the atmosphere is strongly coupled to the state of the global biosphere.

There are a number of consequences of the changing composition of the atmosphere. Amongst the better known of these are acid rain, ozone depletion and the Antarctic ozone hole and the changing climate due to the greenhouse effect. It is the implications of the last of these that are the subject of the papers in this volume.

It is now clear that the concentrations of a number of greenhouse gases are increasing in the atmosphere at various rates and for a number of reasons. Carbon dioxide, CH4 and N2O are currently increasing at 0.4, 1.0 and 0.2-0.3% y-1, respectively, and have increased over the past century or so, by 23, 160 and 8% respectively. The key CFCs, CCl3F and CCl2F2, are both increasing at about 5% y-1. There is no reason to expect that these increases will not continue at least into the next few decades, even though the rates of increase may be affected by such factors as a slowing of the growth of the human population, agricultural expansion, and energy usage, or the imposition of control strategies as is the case for the CFCs.

Ramanathan et al. (1985) have used a one dimensional radiative-convection model to assess the relative contribution of each of the greenhouse gases. While the one dimensional model ignores many of the important features incorporated in the general circulation models, it provides a convenient framework for the comparison of the effects of all gases. The radiative flux divergence at the earth’s surface due to the increase of CO2 in the next 50 years is expected to be 0.9-3.2 W m-2 with a best estimate of the warming due to CO2 by the year 2030 being 0.7 °C (Table 2). The combined contributions of the other greenhouse gases mentioned above will be 1.3-3.9 W m-2 or a best estimate of warming of 0.7 °C, that is as much as for CO2 alone. The realization of this fact has meant that estimates of the time required for significant greenhouse warming of the planet are now about half what it would be due to CO2 increases alone. This is one reason for the greater sense of urgency expessed by the atmospheric research community in recent years concerning the need for the wider community to consider these changes and their potential effect on the planetary climate (Bolin et al., 1986).

Interpreting these divergences in terms of actual surface temperature change is not simple. The one dimensional model used to compare the relative radiative importance of the various gases in Table 3 is not the best way to take into account the full complexity of the climate system. The most reliable way of making such estimates is by the use of general circulation models. However, because of the expense of running these models, so far they have been used only for studying the effects of fixed increases (usually 100%) in concentration of CO2. Tucker (this volume) shows that these models indicate that a doubling of CO2 concentration might be expected to bring about a global surface warming of 1.3-4.0 °C. While such a doubling is not expected until the later half of the next century, an effective doubling, with the consequential climatic effects, is expected within the next 50 years, because of the combined effect of all of the greenhouse gases.

8. ACKNOWLEDGEMENTS

Much of the data presented in this paper to illustrate our growing knowledge of the changing chemical atmosphere have been obtained at the Cape Grim Observatory (Baseline Air Pollution Monitoring Station, Tasmania), through the efforts of a number of scientific and technical colleagues on the staff of the station and at the CSIRO Division of Atmospheric Research. This station has set remarkable standards and goals for its endeavours and I acknowledge the personal committments of these persons to the project, their release of data and assistance in the preparation of this publication. I acknowledge also the permission of my colleague Mr D M Etheridge (Australian Antarctic Division) to use some of our unpublished ice-core data.

9. REFERENCES

Bacastow, R.B. and Keeling, C.D. (1981): Atmospheric carbon dioxide concentration and the observed airborne fraction. In: Carbon Cycle Modelling. Scope, 16. B.Bolin (ed.), John Wiley, Chichester, 390pp.

Barnola, J.M., Raynaud, D., Korotkevich, Y.S. and Lorius, C. (1987): Vostock ice core provides 160,000-year record of atmospheric CO2. Nature, 329, 408-414.

Bingemer, H.G. and Crutzen, P.J. (1987): The production of methane from solid wastes. J.Geophys.Res., 92, 2181-2187.

Blake, D.R. and Rowland, F.S. (1986): World-wide increase in tropospheric methane, 1978-1983. J.Atmos.Chem., 4, 43-62.

Bojkov, R.D. (1986): Surface ozone during the second half of the nineteenth century. J.Climate and Applied Meteorol., 25, 343-352.

Bolin, B., Doos, B.R., Jager, J. and Warrick, R.A. (eds.) (1986): The Greenhouse Effect, Climatic Change and Ecosystems. Scope 29. John Wiley, Chichester, 541pp.

Bolle, H.-J., Seiler, W. and Bolin, B. (1986): Other greenhouse gases and aerosols. In: The Greenhouse Effect, Climatic Change and Ecosystems. Scope 29. Bolin, B., Doos, B.R., Jager, J. and Warrick, R.A. (eds.). John Wiley, Chichester, 541pp.

Broecker, W.S. (1987): Unpleasant surprises in the greenhouse effect. Nature, 328, 123-126.

Cicerone, R.J. (1987): Changes in stratospheric ozone. Science, 237, 35-42.

Crutzen, P.J. and Schmailzl, U. (1983): Chemical budgets of the stratosphere. Planet. Space Sci., 32, 1009-1032.

Crutzen, P.J., Aselmann, I. and Seiler, W. (1986): Methane production by domestic animals, wild ruminants, other herbivorous fauna, and humans. Tellus, 38B, 271-284.

Cunnold, D.M., Prinn, R.G., Rasmussen, R.A., Simmonds, P.G. Alyea, F.N. Cardelino, C.A., Crawford, A.J., Fraser, P.J. and Rosen, R.D. (1986): Atmospheric lifetime and annual release estimates for CFCl3 and CF2Cl2 from 5 years of ALE data. J.GeophysRes., 91, 10 797-10 817.

Dickinson, R.E. and Cicerone, R.J. (1986): Future global warming from atmospheric trace gases. Nature, 319, 109-115.

DOE (1985a): Detecting the climatic effects of increasing carbon dioxide. US Department of Energy, Washington D.C., DOE/ER-0235,198pp.

DOE (1985b): Glaciers, ice sheets and sea level: effect of a CO2-induced climatic change. US Department of Energy, Washington D.C., DOE/ER/60235-1, 330pp.

DOE (1985c): Projecting the climatic effects of increasing carbon dioxide. US Department of Energy, Washington D.C., DOE/ER-0237, 381pp.

DOE (1985d): Direct effects of increasing carbon dioxide on vegetation. US Department of Energy, Washington D.C., DOE/ER-0238, 286pp.

Edmonds, J.A. and Reilly, J.M. (1983a): Global energy production and use to the year 2050. Energy, 8, 419-432.

Edmonds, J.A. and Reilly, J.M. (1983b): A long-term global energy-economic model of carbon dioxide release from fossil fuel use. Energy Econ., 5, 74-88.

Ehhalt, D.H. (1987): How has the atmospheric concentration of CH4 changed? In The Changing Atmosphere. Dahlem Conference, Berlin, 1987. Rowland, F.S. and Isaksen, I.S.A. (eds.). In Press.

Enting, I.G. (1985): Green’s functions and response functions in geochemical modelling. Pageoph., 123, 328-343.

Enting, I.G. and Mansbridge, J.V. (1987a): The incompatibility of ice-core CO2 data with reconstructions of biotic CO2 sources. Tellus, 39B, 318-325.

Enting, I.G. and Mansbridge, J.V. (1987b): Inversion relations for the deconvolution of CO2 data from ice-cores. Inverse Problems, 3, 63-69.

Enting, I.G. and Pearman, G.I. (1987): Description of a one-dimensional carbon cycle model calibrated using techniques of constrained inversion. Tellus, 39B, 459-476.

Etheridge, D.M., Pearman, G.I. and de Silva, F. (1987): Atmospheric trace gas variations as revealed by air trapped in an ice core from Law Dome, Antarctica. Annals of Glaciol, 10, 28-33.

Farman, J.C., Gardiner, B.G. and Shanklin, J.D. (1985): Large losses of total ozone in Antarctica reveal seasonal ClOx/NOx interaction. Nature, 315, 207-210.

Fraser, P.J. and Derek, N. (1987): Atmospheric halocarbons and nitrous oxide, 1982-1985. In: Baseline Atmospheric Program (Australia), 1985. Forgan, B.W. and Fraser, PJ. (eds.). Australian Bureau of Meteorology/CSIRO Division of Atmospheric Science, Canberra, 78pp.

Fraser, P.J., Hyson, P., Enting, I.G. and Pearman, G.I. (1983): Global distribution and southern hemisphere trends of CCI3F. Nature, 302, 692-695.

Fraser, P.J., Hyson, P., Rasmussen, R.A., Crawford, A.J. and Khalil, M.A.K. (1986): Methane, carbon monoxide and methyl chloroform in the southern hemisphere. J.Atmos.Chem., 4, 3-42.

Fraser, P.J., Coram, S and Derek, N. (1987a): Atmospheric methane, carbon monoxide and carbon dioxide by gas chromatography, 1978-1985. In Baseline Atmospheric Program (Australia) 1985. Forgan, B.W. and Fraser, P.J. (eds.), Department of Science/CSIRO, Canberra, 78pp.

Fraser, P.J., Rasmussen, R.A. and Khalil, M.A.K. (1987b): Atmospheric observations of chlorofluorocarbons, nitrous oxide, methane and carbon monoxide from the Oregon Graduate Center (OGC) flask sampling program, 1984-1985. In Baseline Atmospheric Program (Australia) 1985. Forgan, B.W. and Fraser, P.J. (eds.), Department of Science/CSIRO, Canberra, 78pp.

Friedli, H., Moor, E., Oeschger, H., Siegenthaler, U. and Stauffer, B. (1984): ¹³C/¹²C ratios in CO2 extracted from Antarctic ice. Geophys. Res. Letters, 11, 1145-1148.

Friedli, H., Lotscher H., Oeschger H., Siegenthaler U. and Stauffer B. (1986): Ice core record of the ¹³C/¹²C ratio of atmospheric CO2 in the past two centuries. Nature, 324, 237-238.

Genthon, C., Barnola, J.M., Raynaud, D., Lorius, C., Jouzel, J., Barkov, N.I., Korotkevich, Y.S. and Kotlyakov, V.M. (1987): Vostok ice core: climatic response to CO2 and orbital forcing changes over the last climatic cycle. Nature, 329, 414-418.

Hao, W., Wofsy, S., McElroy, M., Beer, J. and Toqan, M. (1987): Sources of atmospheric nitrous oxide from combustion. J.Geophys.Res., 92, 3098-3104.

Houghton, R.A., Hobbie, J.E., Melillo, J.M. Moore, B., Peterson, B.J., Shaver, G.R. and Woodwell, G.M. (1983): Changes in the carbon content of terrestrial biota and soils between 1860 and 1980: A net release of CO2 to the atmosphere. Ecol.Monog., 53, 235-252.

Houghton, R.A., Boone, R.D., Melillo, J.M., Palm, C.A., Woodwell, G.M., Myers, N., Moore, B. and Skole, D.L. (1985): Net flux of carbon dioxide from tropical forests in 1980. Nature, 316, 617-620.

Houghton, R.A., Boone, R.D., Fruci, J.R., Hobbie, J.E., Melillo, J.M., Palm, C.A., Peterson, B.J., Shaver, G.R., Woodwell, G.M., Moore, B., Skole, D.L. and Myers, N. (1987): The flux of carbon from terrestrial ecosystems to the atmosphere in 1980 due to changes in land use: geographic distribution of the global flux. Tellus, 39B, 122-139.

Johnston, K. (1987): First steps in ozone protection agreed. Nature, 329, 189.

Jones, P.D., Wigley, T.M. and Wright, P.B. (1986): Global temperature variations between 1861 and 1984. Nature, 322, 430-434.

Keepin, W., Mintzer, I. and Kristoferson, L. (1986): Emisison of CO2 into the atmosphere. In: The Greehouse Effect, Climatic Change and Ecosystems, SCOPE 29. Bolin, B., Doos, B.R., Jager, J. and Warrick, R.A. (eds.), pp.35-91. John Wiley, Chichester, 541pp.

Khalil, M.A.K. and Rasmussen, R.A. (1983a): Sources, sinks and seasonal cycles of atmospheric methane. J.Geophys.Res., 88, 5131-5144.

Khalil, M.A.K. and Rasmussen, R.A. (1983b): Increase and seasonal cycles of nitrous oxides in the earth’s atmosphere. Tellus, 35B, 161-169.

Khalil, M.A.K. and Rasmussen, R.A. (1985): Causes of increasing methane: depletion of hydroxly radicals and the rise of emissions. Atmos.Environ., 19, 397-407.

Khalil, M.A.K. and Rasmussen, R.A. (1987): Nitrous oxide: Trends and global mass balance over the last 3 000 years. Annals of Glaciol., 10, 73-79.

Komhyr, W.D., Gammon, R.H., Harris, T.B., Waterman, L.S., Conway, T.J., Taylor, W.R. and Thoning, K.W. (1985): Global atmospheric CO2 distribution and variations from 1968-1982 NOAA/GMCC CO2 flask sampling data. J.Geophys.Res., 90, 5567-5596.

Logan, J.A. (1985): Tropospheric ozone: Seasonal behavior, trends and anthropogenic influence. J.GeophsRes., 90, 10 463-10 482.

Lowe, D.C., Brenninkmeijer, C.A.M., Manning, M.R., Sparkes, R. and Wallace, G. (1988): Radiocarbon determination of atmospheric methane at Baring Head, New Zealand. Nature, 332, 522-525.

Matthews, E. and Fung, I. (1987): Methane emission from natural wetlands: Global distribution, area, and environmental characteristics of sources. Global Biogeochemical Cycles, 1, 61-86.

Molina, MJ. and Rowland, F.S. (1974): M.J. and Rowland, F.S. (1974): Stratospheric sink for chlorofluorocarbons: climate atom catalysed destruction of ozone. Nature, 249, 810-812.

NASA (1985): Atmospheric ozone: Assessment of our understanding of the processes controlling its present distribution and change. World Meteorological Organization, Global Ozone Research and Monitoring Project, Report No. 16. WMO, Geneva. Vol.I, II and III, total 1095pp, 86pp references.

NASA (1988): Executive summary of the Ozone Trends Panel, NASA, Washington DC, 31pp. (preprint).

Neftel, A., Moor, E., Oeschger, H. and Stauffer, B. (1985): Evidence from polar ice cores for the increase in atmospheric CO2 in the past two centuries. Nature, 315, 45-47.

Oeschger, H. and Siegenthaler, U. (1988): How did atmospheric CO2 change? In The Changing Atmosphere. Dahlem Conference, Berlin, 1987. Rowland, F.S. and Isaksen, I.S.A.(Edts.). John Wiley, Chichester, In Press.

Pearman, G.I. and Beardsmore, D.J. (1984): Atmospheric carbon dioxide measurements in the Australian region; ten years of aircraft data. Tellus, 36B, 1-24.

Pearman, G.I., Etheridge, D.M., de Silva, F. and Fraser, PJ. (1986): Evidence of changing concentrations of atmospheric CO2, N2O and CH4 from air bubbles in Antarctic ice. Nature, 320, 248-250.

Pearman, G.I. and Hyson, P. (1986): Global transport and inter-reservoir exchange of carbon dioxide with particular reference to stable isotopic distributions. J.Atmos.Chem., 4, 81-124.

Pearman, G.I. and Fraser, P.J. (1988): Sources of increased methane. Nature, 332, 489-490.

Perry, A.M., Araj, K.J., Fulkerson, W., Rose, D.J., Miller, M.M. and Rotty, R.M. (1982): Energy supply and demand implications of CO2. Energy, 7, 991-1004.

Prinn, R.G., Simmonds, P.L., Rassmussen, R.A., Rosen, R.O., Alyea, F.N., Cardelino, C.A., Crawford, A.J., Cunnold, D.M., Fraser, PJ. and Lovelock, J.E. (1983) The Atmospheric Lifetime Experiment 1. Introduction, instumentation and overview. J.Geophys.Rev., 88, 8353-8367.

Prinn, R.G., Cunnold, D.M., Rasmussen, R.A., Simmonds, P.L., Alyea, F.N., Crawford, A.J. Fraser, P.J. and Rosen, R.O. (1987): Atmospheric trends in methylchloroform and the global average for the hydroxyl radical. Science, 238, 945-950.

Prinn, R.G., Cunnold, D.M., Rasmussen, R.A., Simmonds, P.L., Alyea, F.N., Crawford, A.J. Fraser, P.J. and Rosen, R.O. (1988): Atmospheric trends and emissions of nitrous oxide deduced from nine years of ALE-GAGE data. Manuscript in preparation.

Ramanathan, V., Cicerone, R.J., Singh, H.B. and Kiehl, J.T. (1985): Trace gas trends and their potential role in climatic change. J.Geophys.Res., 90, 5547-5566.

Rasmussen, R.A. and Khalil, M.A.K. (1984): Atmospheric methane in the recent and ancient atmospheres. J.Geophys.Res., 89, 11 599-11 605.

Reister, D.B. and Rotty, R.M. (1983): Simple models for the consumption rate of fossil fuel. Energy, 8, 283-289.

Rotty, R.M. and Reister, D.B. (1986): Use of energy scenarios in addressing the CO2 question. J. Air Pollution Control Assoc., 36, 1111-1115.

Schnell, R.C. and Rosson, R.M. (eds.) (1987): Geophysical Monitoring for Climatic Change No. 15. Summary Report 1986. US Dept. Commerce, NOAA/ERL/ARL/GMCC, Boulder, 155pp.

Sebacher, D.L, Harriss, R.C, Bartlett, K.B., Sebacher, S.M. and Grice, S.S. (1986): Atmospheric methane sources: Alaskan tundra bogs, an alpine fen, and a subarctic boreal marsh. Tellus, 38B, 1-10.

Seiler, W., Holzapfel-Pschorn, A., Conrad, R and Scharffe, D. (1984): Methane emission from rice paddies. J.Atmos.Chem., 1, 241-268.

Seiler, W. and Conrad, R. (1987): Contribution of tropical ecosystems to the global budgets of trace gases, especially CH4, H2, CO and N2O. In: The Geophysiology of Amazonia, Vegetation and Climate Interactions. Dickinson, R.E.(Edt.), pp. 133-162. John Wiley, 526pp.

Siegenthaler, U. and Oeschger, H. (1987): Biospheric CO2 emissions during the past 200 years reconstructed by deconvolution of ice core data. Tellus, 39B, 140-154.

Simmonds, P.G., Alyea, F.N., Cunnold, D.M., Lovelock, J.E., Prinn, R.A., Rasmussen, R.A., Lane, B.C., Cardelino, C.A. and Crawford, A.J. (1983): The Atmospheric Lifetime Experiment, 6, Results for CCl4 based on 3 years of data. J.Geophys.Res., 88, 8427-8441.

Simmonds, P.G., Cunnold, D.M., Alyea, F.N., Cardelino, C.A., Crawford, A.J., Prinn, R.A., Fraser, P.J., Rasmussen, R.A., and Rosen, R.O. (1988): Carbon tetrachloride lifetimes and emissions determined from daily global measurements during 1978-1985. J.Atmos.Chem. (In press).

Stauffer, B., Fischer, G., Neftel, A. and Oeschger, H. (1985): Increase of atmospheric methane recorded in Antarctic ice core. Science, 229, 1386-1388.

Steele, L.P., Fraser, P.J., Rasmussen, R.A., Khalil, M.A.K., Conway, T.J., Crawford, A.J., Gammon, R.H., Masarie, K.A. and Thoning, K.W. (1987): The global distribution of methane in the troposphere. J.Atmos.Chem., 5, 125-171.

Thompson, A.M. and Cicerone, R.J. (1986): Possible perturbations to CO, CH4 and OH. J.Geophys.Res., 91, 10 853-10 864.

Volz, A. and Kley, D. (1988): Evaluation of the Montsouris series of ozone measurements made in the nineteenth century. Nature, 332, 240-242.

Weiss, R.J. (1981): The temporal and spatial distribution of tropospheric nitrous oxide. J.Geophys.Res., 86, 7185-7195.

Weiss, R.J. and Craig, H. (1976): Production of atmospheric nitrous oxide by combustion. Geophys.Res.Letts., 3, 751-753.

Zardini, D., Raynaud, D., Scharffe, D. and Seiler, W. (1987): N2O measurements of air extracted from Antarctic ice cores. Annals of Glaciol., 10, 222.

Climate modelling: how does it work?

G. B. Tucker

Abstract - This review for non-specialists argues that a numerical modelling approach is the only reliable way to obtain estimates of the effect on climate of increasing levels of trace gases in the atmosphere. Three types of numerical model have been used for this purpose; in increasing order of complexity they are: energy balance models, radiative-convective models and general circulation models (GCMs). But GCMs are the best means of quantitatively synthesising the wide range of processes at work controlling climate. An outline of some of the difficulties associated with GCMs is followed by extracts from some recent results, including one which shows limitations in the ability of modern GCMs to simulate the present southern hemisphere climate. Confidence in GCM estimates of future climate change will depend on:

• Their ability to simulate current climate.

• The similarity between results from different models.

• The emergence of a climate change signal from the real atmosphere which conforms to model predictions.

Key word index: general circulation modelling, climate simulation.

1. INTRODUCTION

The atmosphere-ocean system is extremely complex and highly non-linear. Interactive processes abound and there are many feed-back loops. Even if a cause-effect relation between an input variable and an output variable can be recognized, the consequences of halving or doubling the input cannot be inferred from an assumption of linearity. Accordingly, it is dangerous to use simplistic qualitative arguments to predict future behaviour.

Assessing the greenhouse effect involves estimating the atmosphere’s response to gradual changes in composition that are happening now, and will continue to happen for as long as the world population continues to increase. These progressive developments are occurring over a time which is very long compared with the life-time of individual weather systems. This means that small but systematic effects which can be ignored for day to day weather forecasting require careful consideration and precise calculation.

There is another fundamental problem too. In most branches of science it is possible to carry out direct experiments on the subjects of interest. They can be pushed or pulled, expanded or contracted, magnetized or demagnetized, treated with chemicals or bombarded with energetic particles, etc. - to see what happens. Then the whole experiment can be repeated, changing the variables. But in some other branches of science this is not possible. Experiments cannot be carried out on other galaxies, on our own sun or our sister planets. Neither is it possible to experiment with the circulation of the Pacific Ocean nor the effect of tectonic plate movements on the rate of growth of the Himalayan mountains. Similarly, experiments cannot be carried out on the global atmosphere. The only techniques of study available are:

To observe its behaviour and to attempt to understand the processes at work by deduction from past events and from established scientific laws.

To use some form of surrogate for experimentation.

In the present case, the need is to determine atmospheric behaviour for trace gas levels that have never existed in the atmosphere before (chlorofluorocarbons), or have never existed in similar concentrations (methane), or were last at such concentrations 100 million years ago (carbon dioxide, CO2). Observing present behaviour is therefore of limited use, and little is known about the climate of the distant past. Climate fluctuations in the more recent past can give some idea of the type of change possible. But because it is debatable whether any portion of these fluctuations were caused by atmospheric composition changes, past climatic anomalies - even if they are adequately known - may not be good analogues for the future. Surrogates therefore assume an important role in climate change studies. There are three types of surrogate:

Analytical mathematical theory.

Physical (laboratory-based) modelling.

Numerical (computer-based) modelling.

For studies of global scale atmospheric circulation, analytical mathematical theory is of limited use because the forces at work and the resulting behaviour are much too complicated to be handled analytically in the necessary detail. Similarly, physical (laboratory-based) modelling is relevant only if the scale of behaviour of the medium in the laboratory (water) is a good enough simulation of the real system. This is true in a very general sense, but only some broad principles can be investigated in this way.

The third surrogate is by far the most promising and the remainder of this paper is concerned with the use of such models to assess the likely climatic response to increasing levels of radiatively important gases. In the following Sections an outline of numerical (computer-based) modelling includes a brief description of a hierarchy of climate models and a comparison of the results of their application to the problem of doubling (from 300 ppmv to 600 ppmv) the concentration of CO2 in the atmosphere. General circulation models (GCMs) provide the most complete treatment of the problem. A discussion of some of the outstanding difficulties associated with this type of model is followed by a comparison of selected results. Finally, examples of some recent analyses of detailed results are presented - with a recommendation that similar studies need to be undertaken for regions of Australia, as a prerequisite for realistic future assessments of the socio-economic impact of the greenhouse effect.

2. THE MODELLING CONCEPT

Let’s first be clear about what is meant by a mathematical or numerical model in the context of this topic. It means a set of one or more mathematical equations that describes the behaviour of the atmosphere. Usually this involves two components: a physical characterisation of the problem, and the derivation and solution of the relevant equation(s). In the case of the earth’s climate, the practical concern is usually what is happening close to the surface. But to understand this it is necessary to consider events in the entire system.

However, the simplest models consider only the average surface air temperature, Ts, and assume that Ts responds only to variations in the radiation (energy) budget at the surface or at the top of the atmosphere. Such models are called ’energy balance models’ (EBMs); they involve the crudest parameterization. Schlesinger (1986) has selected some of the published results of the application of these models to a doubling of CO2 (generally from 300 ppmv to 600 ppmv). These changes in global average surface air temperature, TS, are given in Table 1 for EBMs that consider the energy balance at the surface (SEBMs) and at the top of the planetary atmosphere (PEBMs).

TABLE 1. Global average surface temperature change, TS, of selected energy balance models (EBMs) due to a doubling of atmospheric CO2 concentration.

* See Schlesinger (1986) for the above references.

As might be expected the models have an inherent difficulty in specifying the behaviour of the atmosphere away from the energy balance level. This, together with the range of options available for this crudest set of parameterizations, accounts for the wide range of results. But all the predicted temperature changes are positive, and the more recent PEBM results are grouped around +2°C.

A second type of model goes one step further than considering merely the radiation balance at one level; it considers also vertical adjustments due to convective processes - vertical movements of air due to density changes. These ’radiative-convective models’ (RCMs) give scope for considerable ingenuity in representing not only solar and terrestrial radiation but also the vertical re-distribution of heat due to turbulent transfer across the earth’s surface, dry and moist convection, atmospheric water vapour and clouds. Schlesinger’s (1986) selection of RCM results for a doubling of CO2 is given in Table 2.

TABLE 2. Global average surface temperature change, TS, of selected radiative-convective models (RCMs) due to a doubling of atmospheric CO2 concentration

These results, like those from EBMs, are grouped around a temperature change of +2°C for a CO2 doubling. However, such models are useful also to test parameterization schemes to be applied to the more comprehensive GCMs and also to undertake feedback analyses of GCM results.

One RCM (Ramanathan et al., 1985) has been used to study the relative contribution of different trace gases using their predicted concentrations for the year 2050 (Figure 1). It shows that the combined effect of all the so-called ’greenhouse gases’ on surface temperature change is more than double the effect of CO2 alone.

However, an accurate understanding and reliable prediction of the effect of a change in trace gas concentrations requires a consideration of the 3-dimensional structure and behaviour of the atmosphere and its variation in time. This is necessary not only to take into account all the interactive atmospheric processes influencing the re-distribution of heat but also the effect of this re-distribution on the re-distributive processes themselves. Further, because the atmosphere interacts with what is underneath it, processes affecting the earth’s surface and the ocean’s surface have to be included. In the case of the ocean, because, like the atmosphere it is a fluid, the best way of handling sea-air interaction is to use a joint ocean-atmosphere model.

Changes in heating (and cooling) patterns influence all the other elements of climate: rainfall, cloudiness, frequency of tropical cyclones, mid-latitude storm tracks, first and last frosts, snowfall, sea ice, etc. For such a complete treatment it is necessary to consider the whole globe from the surface to well into the stratosphere, and all the major processes influencing the complete earth-ocean-atmosphere system (Figure 2). GCMs are designed to satisfy these requirements.

For comparison with SEM and RCM results, Schlesinger (1986) has listed the global average surface air temperature results of selected GCMs for a CO2 doubling (Table 3).

Figure 1. Cumulative near-surface air warming for an adopted trace gas scenario. Period: 50 years from 1980 levels. These results are based on a one-dimensional radiative-convective model. See Ramanathan et al. (1985).

TABLE 3. Global average surface temperature change, TS, of selected general circulation models (GCMs) due to a doubling of atmospheric CO2 concentration.

* See Schlesinger (1986) for the above references.

Figure 2. Schematic illustration of the components of the coupled atmosphere-ocean-ice-land surface-biomass climatic system. The full arrows are examples of external processes and the open arrows are examples of internal processes in climatic change. After GARP (1975).

These values are grouped around +3°C. They are somewhat larger than the EBM and RCM results. These models also depict other climatic elements. They describe detailed patterns of distribution around the globe and the day-to-day variation of these elements. But before looking at these more detailed results from GCMs, it is useful to note some of the problems they encounter.

3. SOME MAJOR PROBLEMS

Atmospheric mechanisms exist on all time scales. In formulating the equations of GCMs, it is necessary to select the time scale of interest (in this case 50-100 years) and somehow to represent the way in which smaller scale processes and changes in overall conditions influence this scale. This is done in three ways:

By parameterizing the smaller scale processes; that is, by developing mathematical expressions that represent these processes in terms of larger scale variables that the model will handle explicitly.

By integrating the equations in time; that is, using equations that represent the average values of the variables over a small time interval, say, 10 minutes, and also the rate of change of the critical variables. The equations are then solved for this change which is subsequently applied to advance the model forward one time-step. Proceeding onwards in time to, say, one month or one year, the average behaviour of the model during this time is an analogue of the climate during that month or that year.

By using these techniques to carry out sensitivity studies; that is, running the model under present-day conditions to obtain a measure of its representativeness and then re-running it under the predicted conditions to obtain its changed climate.

Many problems exist in the planning, formulation and execution of such a complex numerical experiment. Some of the most important can be grouped under three headings:

Physical parameterization of processes.

Mathematical solution of equations.

Interaction between the atmosphere and the ocean.

The physical parameterization problems are associated with the need to represent a wide variety of interacting processes over a range of scales. Because even a very small systematic change can become substantial over a period of 50-100 years, it is dangerous to dismiss any process as insignificant. Accordingly, parameterization schemes must be capable of providing approximate statistical steady state (balanced) solutions but also of accurately representing apparent