Future Climates

Estimates of climates in the future depend on the relative significance of the factors causing past changes (Section 15.3), including the random element. It appears that random events such as

Figure 15.15 'Relative trends' in precipitation between 1950 and 1993, i.e. the change expressed as a percentage of the average precipitation during that period. For instance, if the average rainfall between 1950 and 1993 was 1,000 mm/a and a gradual decrease of 100 mm/a occurred during that period (e.g. from 1,050 to 950 mm/a), then the relative trend is about 10 per cent. The largest black circles mean a relative trend of 200 per cent, the next largest 100 per cent, the next 50 per cent, the next 25 per cent and the smallest 10 per cent. Solid rings imply increases and hollow rings decreases.

Figure 15.15 'Relative trends' in precipitation between 1950 and 1993, i.e. the change expressed as a percentage of the average precipitation during that period. For instance, if the average rainfall between 1950 and 1993 was 1,000 mm/a and a gradual decrease of 100 mm/a occurred during that period (e.g. from 1,050 to 950 mm/a), then the relative trend is about 10 per cent. The largest black circles mean a relative trend of 200 per cent, the next largest 100 per cent, the next 50 per cent, the next 25 per cent and the smallest 10 per cent. Solid rings imply increases and hollow rings decreases.

Table 15.4 Changes of the annual number of raindays at places on the southern tablelands of New South Wales

Intensity 1880-1900 1901-20 1921-40 1941-60 1961-80 1981-90 (mm/aay)

volcanic eruptions have an appreciable effect on the atmosphere (Note 2.G), but only for a year or two. Cyclic processes such as the Milankovic variations operate slowly. The impacts of human society are of increasing importance.

Effects of Human Society

People are likely to determine future climates in various ways. An extreme instance is nuclear warfare, which would entail the horrors of airborne radioactivity, possibly damage the stratospheric ozone layer, and maybe force enough aerosols into the stratosphere to exclude the sunshine to the extent that temperatures fall and crops are seriously depleted (Note 2.G). More certain will be the impacts of population growth and industrialisation, producing greater urban heating in the swollen cities and more atmospheric pollution by chemicals and dust. Sulphate pollution from the burning of coal is particularly threatening, because of the impact on cloud formation and hence global temperatures (Sections 8.9 and 15.4). It also leads to acid rain (Section 10.1). So it is a matter of concern that increasing amounts of coal are being burnt, especially in developing countries.

More profoundly, there is a fear that the carbon dioxide and similar gases (Section 2.7) that we produce in increasing amounts (Figure 1.2) will eventually create a dangerous degree of enhanced greenhouse heating (Note 2.L). Emissions are likely to be three times the 1990 rate by the year 2100, if the world population doubles, as seems probable, and if there is moderate economic growth without strong pressure to reduce the emissions. In the meantime, the effect on the atmospheric concentration of carbon dioxide seems likely to be a doubling (compared with 1900) by the middle of the next century. To prevent such an increase, it was resolved at an international conference in Rio de Janeiro in 1992 that emissions should be cut back to 1990 levels by the year 2000. The chances of that being achieved seem slender, especially in developing nations, though the increase of emissions had been cut back from 1.5 ppm/year in the 1980s to just below 1 ppm/year in 1995, and a similar slowdown has been measured for methane.

The effects of the expected increase of carbon dioxide can be estimated in various ways. Firstly, consider the remarkable dependence of global temperature on CO2 concentration during the last 160,000 years (Figure 15.8), implying future warming. Secondly, extrapolate twentieth-century changes, which have shown warming recently (Figure 15.10). Thirdly, calculate the change of temperature from the expected effect on the loss of longwave radiation to space (Note 15.K), which again suggests global warming. Fourthly, note that computer simulations of future climates by means of GCMs (Section 12.6) all indicate a rise of global temperatures.


A general circulation model (or climate model) is basically the same as a NWP model except that the GCM is set to follow developments over many years, not a few days, at the expense of spatial and temporal detail. (A wide spacing of the points which sample the climate, and a long time step between recalculation of changes of temperature, etc., are both needed to reduce the number of calculations, to permit working out climates decades hence, in a reasonable time.) A climate model uses the same equations as a weather model, starting from present January-average conditions, for instance, and particular attention is paid to processes which are too slow to be important in weather forecasting, such as changes of land or ocean-surface conditions. Modern models (there are about a score of different GCMs in the USA, Australia, Europe and elsewhere) are much improved on those used by pioneers in the 1970s. They now allow for a gradual increase of CO2 concentration, for the other greenhouse gases, for sulphate aerosols, for the daily cycle of solar radiation, for the interaction between atmosphere and ocean, etc.

The models are used for 'what-if?' questions, e.g. what would the general circulation be like if the carbon-dioxide concentration were doubled? The answers achieved are made credible by the success of GCMs in the following ways:

(a) in representing current patterns of pressure, temperature and rainfall (Figure 15.16), and their variability,

(b) in simulating the climates of about 9,000 BP, when the seasonal variation of radiation was different because of the altered orbit around the Sun (Section 2.2), and

(c) in reproducing the global response of the atmosphere to perturbations such as El Niño and volcanic eruptions.

It is notable that every GCM shows the inevitability of global warming from a doubling of carbon dioxide. Though estimates of the amount of warming vary within the range 1.55 K, the range is being narrowed, converging towards 3 K or so. This is approximately equivalent to the difference between the annual

Figure 15.16 The degree to which the output of a modern GCM reproduces the actual variation of zonal-mean rainfall with latitude.

means of Wellington and Auckland, or Canberra and Adelaide, for example.

The various models differ chiefly in their estimates of future rainfalls in particular regions, because of uncertainty about the effects of changes of cloudiness at either low or high levels. Some models, but not all, indicate that twice the carbon dioxide implies fewer raindays around 30°S in Australia, but more rain on each rainday, leading to a 10 per cent increase overall. Indeed, one would expect an increase of precipitation as a consequence of more evaporation from warmer oceans.


Higher average temperatures are likely to increase the frequency of days hot enough to affect mortality (Note 3.C). The increase in the case of Alice Springs, for instance, can be estimated by comparing the number of hot days in past years of different average temperature (Figure 15.17). The increase would be less at places by the sea, where daily maxima are limited by sea breezes (Section 14.2). So a general warming by 3 K at Brisbane would increase the number of hot days by only 5.4 annually, and at Melbourne would increase the annual number from the present eight, to fifteen.

Higher sea-surface temperatures might reduce the subsidence in the north Atlantic which drives the global oceanic conveyor belt (Section 11.5). In that case, there would be considerable alterations to the distribution of heat around the world, with consequences hard to foresee, but possibly comparable to the switching of climate at the end of past Ice Ages.

Evidence in Sections 15.3 and 15.4 indicates that warming in the past has gone hand in hand with increased rainfall at many places. Other clues to future rainfalls can be obtained by comparing rainfalls in the five warmest years during the period 1925-74, for instance, with the pattern in the five coldest; it appears that some places get

Figure 15.17 The connection between annual average temperature and the number of days when the daily maximum exceeds 35°C, at Alice Springs (24°S).

less rain, notably in summer in important grain areas in central USA and much of Europe and Russia. Another method of assessment is to consider rainfalls during the Altithermal period, some 6,000 years ago, when temperatures in Europe were 2-3 K warmer than now (Figure 15.9). This, too, indicates relatively dry conditions in North America but wetter conditions in some other areas, including western Australia. It is clearly unsafe to plan future water supplies, irrigation schemes, hydro-power engineering, etc. on recent climate data alone.


Sea-levels were about 5 m higher than now during the previous interglacial period 120,000 years ago, and over 100 m lower at the depth of the Ice Age at 18 kBP (Section 15.3). So future global warming must lead to the sea rising, chiefly as the result of the increased melting and calving of icebergs from Greenland and the West Antarctic icecap, and, more immediately, thermal expansion of the sea above the thermocline. Sea water expands by roughly 0.02 per cent per degree (Note 11.B), so a layer of 500 metres warming by 1 K on average would rise by 100 mm, perhaps at a rate of about 1 mm/year. This approximates the rate measured at Wellington over the last century. However, any rise may be offset initially by the accumulation of snow in Antarctica due to increased precipitation following more evaporation from warmer oceans. The exact rate of rise is still uncertain, though it is important to people living on coral islands and in Bangladesh and Holland, and eventually to the vast populations living in coastal cities.

Effects on Agriculture

The patchy warming and either increase or decrease of precipitation that seem to lie ahead will affect agriculture, again in an irregular way. Some high-latitude areas, where greenhouse warming is expected to be most dramatic, will have enhanced yields because of greater warmth and fewer frosts, and some semi-arid areas because of additional rain. C3 crops (e.g. wheat and rice) will be stimulated by the higher carbon-dioxide concentration (Note 1.B). But the enhanced evaporation that results from warmth removes soil moisture, tending to reduce crops and extend the deserts. Also, higher temperatures promote insect attack, weeds and disease, and accelerated crop development in warm conditions allows less time before maturity for collecting the radiation which determines the harvest (Note 2.I).

Comparisons of the crops of warm and cool years in the last few decades, calculations based on computer models of crop growth and experiments on plants grown in controlled conditions, all indicate that global warming due to twice the carbon dioxide might reduce yields of wheat and maize by 5-10 per cent in northern America, Europe and New Zealand, for example. It has to be remembered though that such figures ignore the impacts on crops of possible changes in the frequency of extreme events, like droughts, floods and strong winds, and also the mitigation achieved by crop adaptation, selection and breeding.

In short, global warming has to be taken seriously. There are still gaps in our understanding of the process, and uncertainties about its consequences, though rapid progress is being made. Even with the caution traditional in scientific investigation, we can say now that the evidence of the last few decades is 'not inconsistent' with the prospect of considerable changes due to the enhanced greenhouse effect.

In the next chapter we turn from speculations about the future to consideration of the present climates at places in the southern hemisphere.


15.A Weather radar

15.B Places of weather measurement

15.C Frost forecasting

15.D Feedback, chaos and unpredictability

15.E Sunspots and forecasting

15.F Effects of the Moon's phase

15.G Primitive equations for weather forecasting

15.H Numerical Weather Prediction

15.I Forecasting skill

15.J The 'tropics'

15.K Radiative equilibrium and global warming

15.L Radiometric methods of dating past climates

Was this article helpful?

0 0
Solar Power

Solar Power

Start Saving On Your Electricity Bills Using The Power of the Sun And Other Natural Resources!

Get My Free Ebook

Post a comment