This snapshot, taken on
, shows web content acquired for preservation by The National Archives. External links, forms and search may not work in archived websites and contact details are likely to be out of date.
The UK Government Web Archive does not use cookies but some may be left in your browser from archived websites.

Office of Science and Technology


Note by Sir Robert May FRS

This note sets out my personal view of a subject in which there remain significant uncertainties. The main source of assessment of the science is the UN Intergovernmental Panel on Climate Change (IPCC). This is supported by 150 nations, and the UK chairs its science working group. Its last scientific assessment drew on the work of some 3,000 of the world's leading scientists.

    The Greenhouse effect and concerns about it

  1. The physical principle of the greenhouse effect is well established. Put simply, the earth's surface temperature depends on the balance between incoming short-wave energy from the sun and outgoing long-wave energy emitted from the earth's surface and atmosphere. Some gases ("greenhouses gases") in the atmosphere allow short-wave solar radiation to pass through and warm the earth's surface, but at the same time these gases trap some of the long-wave infrared radiation emitted by the ground, and keep the earth warmer than it would otherwise be. Were it not for these natural "greenhouse gases", the most important of which is water vapour, the earth would be roughly 300C colder, and we would not be here.
  2. Concern arises because human activities are increasing the concentration of greenhouse gases, particularly carbon dioxide and methane. While these facts are certain, the implications for changes in average temperatures, both global and local, are less certain. The ultimate effects of such temperature change on rainfall and storm patterns - floods and droughts - and on other aspects of our environment are hard to predict in detail, although the broad outlines seem clear. In what follows I expand on these themes, sketch some likely consequences for different parts of Britain and other places, and outline some policy choices.

    Facts about greenhouse gases

  3. Concentration of carbon dioxide in the atmosphere has increased by about 25% over the past 100 years (Fig 1). If current trends in fossil fuel burning continue, carbon dioxide will be present in the atmosphere at twice pre-industrial levels by around the middle of the next century. Once atmospheric concentrations have been increased they take a long time, characteristically around 100 years, to decrease even if no more carbon dioxide is added. As with turning a large ship, there are long lags between actions aimed at levelling-off carbon dioxide levels, and the levels actually stabilising. This is a strong argument for early action.

  4. Other gases, including methane, nitrous oxide and chlorofluorocarbons (CFCs) also contribute to the greenhouse effect. They too have been increasing in the atmosphere. Methane levels have doubled over the last 100 years. Nitrous oxide levels are currently rising at around 0.25% each year. As for carbon dioxide, all these increases are clearly caused by human activities, largely connected with energy generation, transport and agriculture.

  5. Carbon dioxide contributes most to human-caused global warming, accounting for around 70% of the total. The other gases contribute the remaining 30%, with methane accounting for about 20%.

    Greenhouse gases and global temperature

  6. Over the past 130 years, global average temperature has risen about 0.6oC (Fig 1). This may sound trivial. But the temperature difference between today and the extreme of the last ice age, 20,000 years ago, is only about 5oC (although this was a decrease, rather than an increase, in temperature). The estimated range of variability in global temperature over the past 1,000 years is around 1oC.

  7. Direct attribution of these temperature changes to human activities is complicated by the fact that climate varies naturally from year to year, and from decade to decade. Long-term human-induced warming has to be distinguished against this natural background. Although we do not have data reaching back many hundreds of years, by comparing observations of global mean temperatures with natural variability estimated from climate models, we find the warming has, over the past couple of decades, extended beyond the bounds of our estimates of natural variability. This is why the IPCC considered it valid to conclude that the balance of evidence suggests a discernible human influence on global atmosphere.

    Figure 1

  8. In order to predict how increases in atmospheric concentrations of carbon dioxide and other greenhouse gases will affect global temperature and other climate variables in the future, complex mathematical models of the earth's climate system have been developed. There is no dispute that, all other things being equal, a doubling of atmospheric carbon dioxide concentration would, by itself, lead to an increase in average global temperature of around 1.2oC.

  9. The difficulty is that other things are not equal. A serious problem is that the models are highly non-linear. A doubled input does not necessarily lead to a doubled output; two and two do not always add up to four. These mathematical arcana are often manifested in feedback effects, which can amplify or ameliorate global warming. Important such feedbacks in global climate models arise from water vapour, cloud cover, ocean circulation, reflection from icecaps, and other things.

  10. With a warmer atmosphere, more evaporation occurs from the oceans and from wetland surfaces. On average, a warmer atmosphere will possess a higher water vapour content. Water vapour is a powerful greenhouse gas, so a positive feedback results, amplifying the warming effects.

  11. The effect of cloud cover seems to be very variable, depending on local conditions and on the kind of cloud. Clouds reflect some solar radiation back to space, so reducing the global warming effect. However, they counter this by acting as a blanket for thermal radiation from the earth's surface, thus increasing average temperatures. Which of the two effects dominates depends on cloud temperature, height and optical properties (whether it is ice or water, thick or thin). In general, low clouds cool global climate, whereas high clouds tend to increase temperature. Feedback can therefore be positive or negative, making the modelling difficult, with the effects varying from place to place.

  12. Ocean circulation is particularly important because the ocean acts as a big heat reservoir, redistributing heat globally via its circulation. The timescales involved in ocean circulation are much longer (typically decades) than those in the atmosphere and so couplings between oceans and atmosphere, and possible changes in ocean circulation, must be taken into account in predictions of climate change. Quite small changes in regional transportation by oceans can have a large, but difficult-to-predict, influence on local climate change. Conversely, it is possible that small changes in regional climate could result in large, and possibly abrupt, changes in ocean circulation patterns. All this introduces major uncertainties, particularly at the regional level.

  13. Another feature of non-linear systems is that, under certain circumstances, quite small changes in a "forcing" variable (for example atmospheric carbon dioxide) can lead to abrupt and large changes in a dependent variable (for example, ocean circulation). A possible example of this is the disruption we have seen to the "El Niño" system. This is a region of unusually warm water which appears every three to five years in the Equatorial Pacific and which strongly influences weather patterns, especially in tropical and sub-tropical areas. In recent years, intense El Niño phenomena have been recorded, which are thought to have led to extreme weather events in the Americas, Australia and Africa. If global warming continues, perturbation to weather systems like this are likely to become more common.

  14. The IPCC predicts that when all these feedback effects are taken into account, a doubling of atmospheric carbon dioxide would lead to an average global temperature increase of between 1.5 and 4.5oC, most probably 2.5oC.

  15. Any prediction will depend, of course, on the assumptions we make about future emissions of carbon dioxide and other greenhouse gases. These, in turn, depend on assumptions about future populations, economics and energy generation. The IPCC approaches these uncertainties by spelling out a range of possible scenarios, and then predicting the climate change for each.

  16. Figure 2 offers a summary - and a dramatic summary at that - of the IPCC findings. The left hand side of the figure shows the outcome of the IPCC's various scenarios for atmospheric carbon dioxide concentrations for the next two centuries. Each of these scenarios describes atmospheric carbon dioxide levels eventually stabilising at some steady level; in the case, for example, of S450 this happens around the year 2075, but in most of the scenarios it takes longer. The right hand side of the figure shows the predicted rise in average global temperature associated with each scenario, once carbon dioxide levels have reached their steady state; the horizontal line shows the range of predicted temperatures, and the dot the best guess (for example, for S450, the temperature increase is predicted to lie between 0.8 and 2.1oC, with a best guess around 1.3oC). The figure on the right also displays three vertical lines. The first (labelled a) represents the estimated range of variability in global temperature over the past 1,000 years (around 1oC), and the second (labelled b, at 2 oC) represents double this millennial variability and could be taken as a level at which manmade warming would be self-evident, beyond all dispute. The third (labelled c) shows the difference between the last ice age and the warmest time since (around 5oC).

    Figure 2

  17. These IPCC scenarios represent the levels at which atmospheric carbon dioxide will stabilise. To achieve any of these, carbon dioxide emissions (from transport, power generation, agriculture, etc.) will not just have to stop growing, but will have to be reduced below the present level. For example, S450 assumes that global carbon dioxide emissions will fall below current levels by about 2035, and will reduce below 40% of current levels after 2100. Presented this way, the assessments make stark viewing. If the "best guess" estimate for global warming associated with any of the IPCC scenarios is accepted, only scenarios below S550 - which I rate as rather optimistic, given current trends in emissions - keep temperature increases below the 2oC threshold.

    Consequences of global warming: general

  18. So far, this paper has focused on average global changes in temperature. Of great practical interest is how climate change will affect individual regions. Broadly, temperature increases will be greatest at high northern latitudes, in part because the melting of sea-ice will allow more solar radiation to be absorbed, thus amplifying warming in this region. Warming is also likely to be greater over land areas than over the oceans, due to the slow thermal response of the latter. The models predict extensive areas where rainfall will become greater, and others where the opposite is predicted; in general, places which already get heavy rainfall are likely to see it get heavier; conversely, where rainfall is now light it is likely to get lighter. But the geographical details of these findings remain, at present, uncertain.

  19. I note a few general conclusions, before turning in more detail to the UK. One class of consequences of climate change relates to sea level change. As the ocean warms, it will expand and sea level will rise. Some land ice will melt, and so changes to the large ice masses over Greenland and Antarctica will have additional effects. A rise of some 50cm in average sea levels may be expected over the next century, but there will be larger local effects. As heat diffuses slowly to the deeper ocean, it will cause further expansion; hence, at any given time, the observed sea level rise will only be a fraction of that which will inevitably follow. Even if there were to be no further change in climate (which would require, for example, a 60% decrease in carbon dioxide emissions), sea level rise will continue for hundreds of years.

  20. Temperature change also has effects upon the hydrological cycle, which effectively translates into changes in where rain falls, and where water ends up. As global warming increases, the world will see more and worse droughts and floods.

  21. By the year 2020, climate change in Britain is likely to correspond roughly to a northward shift in climate characteristics of some 100-200km. This and other changes will have major effects upon the habitats and ranges of many species of plants, animals, and micro-organisms. Many of the species, and indeed ecosystems, thus affected will not be able to respond fast enough to "move with the temperature change". The overall effects are extremely complicated, and vary from region to region, in all cases surrounded by a good deal of uncertainty. Rather than make any attempt to survey these questions, I observe that a major recent study has attempted to assess the economic value of the "ecosystem services" delivered by natural ecological processes: soil formation, water supplies, nutrient cycling, waste processing, pollination, and much else. The assessment, necessarily very rough, is around £10-34 trillion per year, with a best guess of around £21 trillion, most of it outside the market. This is roughly twice the conventional global GNP, at around £11 trillion per year. Large swathes of this £10-34 trillion are at risk from the possible environmental and ecological changes sketched by the IPCC.

    Consequences of global warming; Britain

  22. Climate models are not sufficiently accurate at present to give reliable predictions of local climate changes. In the UK, however, climate change may already be having an appreciable effect. Of the five warmest years in Central England's 337 year old temperature records, three (1989, 1990, 1995) have occurred in the past 10 years. The summer of 1976 was the warmest ever, and that of 1995 the second warmest; in summer 1995, temperatures in Central England were 3oC warmer than the average between 1961-1990. 1997 is challenging these records.

  23. In the summer of 1995, rainfall in Central England was about two-thirds of the normal amount. Overall, the most obvious impact was in the energy sector, with net savings to the consumer for the period November 1994 to October 1995 of about £335 million. There were negative impacts on agriculture (about £180 million), water supply (£96 million) and the building insurance sector.

  24. With global warming, we can generally expect the weather to become more extreme and more variable: more heat waves, more floods, more droughts. The deep depression, or "hurricane", which wrought such havoc in Southern Britain in late 1987 resulted in damage estimated at £1.9 billion. The indications from climate models are that the number of deep Atlantic depressions is expected to increase; by the middle of the next century, the incidence of gales across the country is predicted to increase on average by 30% if no major global actions are taken to reduce emissions of greenhouse gases. There may already be some evidence for this. Since the "hurricane" in 1987, there have been "billion dollar" storms around the world, each year. 1990 as well as 1987 was a particularly bad year for storms in Europe.

  25. Climate change scenarios suggest that, as well as becoming more windy, the south of the UK is likely to become hotter and drier, with very warm days becoming much more frequent and the demand for water increasing. In contrast, the north west is likely to become wetter. Drought in the south east and flooding in the north west are likely both to become more common. Storm damage will be more frequent, with effects on flooding and erosion of coastal areas and on the cost of flood defence. As temperature increases and precipitation patterns change, natural habitats, wildlife species and farming zones will steadily migrate northwards (insofar as they are able to) by around 50-80 km per decade.

  26. In the longer term, and more uncertain, are possible effects on ocean currents and productivity around the UK. As mentioned earlier (paragraph 13), there is an important link between deep ocean circulation and the hydrological cycle. Increased precipitation in the North Atlantic region, and increased fresh water run-off, will reduce the salinity of surface water. Water will therefore be less dense and will not sink so readily. Such changes in marine salt balance could modify the fluid dynamics which ultimately drive the Gulf Stream. I emphasise that the Gulf Stream, in effect, transports towards the British Isles "free" heat which amounts to 27,000 times the total power generation capacity of the UK! The possibility that this might be significantly reduced, much less turned off, is an awesome prospect.

  27. More generally, the world may also be affected by migration of populations from areas severely affected by changes in sea level. For example, over 6 million in Bangladesh will be displaced, and 7 million in Egypt severely affected, by a 1m rise in sea level (assuming these populations do not increase, which is unrealistic).

    What to do

  28. I believe the world must aim to reduce the emission of greenhouse gases, especially carbon dioxide. The Prime Minister's personal appearance at the Special Session at the UN General Assembly in June, where he affirmed the UK's target of a 20% reduction in CO2, was a valuable demonstration that the UK takes these issues very seriously. Kyoto in December will present another opportunity for the Deputy Prime Minister to work with our partners, moving further toward concrete international actions to combat global warming.

  29. The UK has already taken exemplary steps to reduce carbon dioxide emissions, mainly as a result of the moves from coal to gas power generation in the late 1980s and early 1990s (which will result in our carbon dioxide emissions in 2000 being 4 to 8% below 1990 levels). Although painful to the coal industry in the UK, these changes entailed relatively little cost in overall measures of environmental protection or changes in national lifestyle. The next steps in the UK will be a lot harder.

  30. The UK's atmospheric input of carbon dioxide can be broken down, by end user, into road transport (around 22%) domestic uses (about 27%) and industry (about 28%); a miscellany of other categories (including other transport, and heating and other energy use in shops and offices) make up the remaining 23%. Thus the task of reducing such inputs potentially falls very broadly.

  31. And we need to think long. Whatever target for 2010 is agreed at Kyoto can only be a first step for the international community. A long term view of policy options is necessary; effective decisions in the short term must not hinder our options for taking measures to meet future reductions targets.

  32. In the short term, significant reductions in greenhouse gas emissions are technically possible, and can be economically feasible. Policy measures to accelerate technology development, and to encourage diffusion and transfer of new technologies to all centres, will help. An integrated transport policy, plus strict air quality standards, should help reduce vehicle emissions, but will not of itself be sufficient without development of cleaner fuels, more efficient engines, and alternative energy sources. The same is true for energy supply for industrial and domestic uses. New technologies will need to include reductions in emissions of greenhouse gases in the "harvesting" and subsequent use of fossil fuels, switching to non-fossil fuel sources of energy and a better efficiency of energy use where possible. Better management of the natural environment can also help. Improvements in how we develop and sustainably use forests, agricultural lands, and soils in general could play an important part in reducing emissions and in enhancing the rate at which carbon is biologically fixed.

  33. UK industry should see climate change as opportunity not as threat. Industry should be constructively addressing climate change in their forward planning. Mitigation of climate change will require big changes in the energy, transport and construction industries, in terms of much greater efficiency in production and use (which in turn will bring greater industrial efficiency and competitiveness) and the development of appropriate, known, technologies. Many aspects of the Foresight initiative are addressing these questions.

  34. Most of these actions will be impossible unless the public in general are persuaded of the need for them. For example, if we are to cut transport emissions not just per vehicle but in aggregate, then some tough choices may have to be made about private car usage. I believe there is a crucial role for ministers collectively to play, leading the British people to appreciate the need to take firm and early action.

  35. The quality of the UK's contribution to research on climate change, in the broadest sense, is strong out of all proportion to our relative size or research spending. We should aim to maintain this strength, so as to have international policy underpinned by fundamental understanding, to continue our position of international policy and scientific leadership in this arena, and to help persuade the world's sceptics that climate change is a real and serious problem.

  36. Ultimately, the problem of climate change demands international co-operation and co-ordination. No matter how good a job we do in the UK, its global effects will be marginal compared with what happens over the next few decades in, say, China. Developed countries, with the highest level of emissions, need to take the lead. We need to promote among developed countries an understanding that serious and urgent action to limit emissions is needed, and to press for a solid result in Kyoto. I also believe the Government should use the moral authority conferred by the UK's conscientious efforts to meet its own targets, combined with our disproportionate contributions to basic understanding of the underpinning science, to help developing countries reconcile sustainable development with amelioration of atmospheric greenhouse gas emissions as far as possible. By this, of course, I do not mean patronising moral exhortations nor throwing money at problems, but rather partnerships in which we explore appropriate forms of help. The Global Environment Facility and, on a small scale (around £3m pa), the post-Rio Darwin Initiative for helping developing countries record and conserve their biological diversity, are models.

Back to contents

Back to Office of Science and Technology Home Page