The debate about global climate change is as unusual as it is controversial. Scientific arguments usually concern interpretation of evidence gathered from the present or from the past, rather than the forecasting of complex future changes. For example, when epidemiologists argue about the effect of air pollution on asthma, they do so with reference to existing empirical research results. However, forecasting the health impacts of climate change requires us to undertake scenario-based risk assessment --- that is, to apply our knowledge of environment-health relationships gained from limited past experience to future environmental changes that are uncertain and that probably far exceed the range of past variation. For example, the rate of temperature increase that the IPCC has forecast for the next century is much faster than any that has occurred in past thousands of years. The resultant uncertainties in forecasting health impacts are compounded by uncertainties concerning social, demographic and technical changes that may influence human vulnerability or adaptive capacity.

This, then, is clearly not an "exact science". But the range and seriousness of the potential health impacts of climate change means that the risk assessments discussed in this volume constitute a very important scientific undertaking. A large part of the task entails considering the various indirect effects upon human health arises from climatic stresses upon the stability and productivity of ecological systems. with few exceptions, the causal relationships involved are complex and multifactorial. A premise underlying this volume, therefore, is that we must now think imaginatively, and within an ecological framework, about the longer-term implications for human health of disturbing or damaging components of the biosphere.

Chapter 1 describes the historical and economic context within which the climate change issue has arisen and discusses the scale, complexity and fundamental "newness" of the problem. An up-to-date review of the science of greenhouse gas accumulation and its effects upon the climate system is given in Chapter 2. Based on anticipated future trends in greenhouse gas emissions, the Intergovernmental Panel on Climate change estimates that global mean temperature will have risen by 2-3 degrees C by late next century. Chapter 2 also discusses the associated problem of stratospheric ozone depletion. Recent studies indicate that warming of the lower atmosphere may, via stratospheric cooling, exacerbate ozone depletion.

The various possible impacts of climate change and stratospheric ozone depletion upon human health are examined in Chapters 3 to 8. The impacts are referred to as "direct" or "indirect". Whereas some impacts can be foreseen easily --- such as an increase in deaths due to an increase in the frequency and severity of heatwaves --- others would depend on, for example, changes in patterns of mosquito populations and regional food production and are therefore less easy to predict. Other, more diverse and diffuse public health impacts would result from population displacement and conflict following sea level rise, and disruption of local economic activity and employment. In particular, these chapters examine:

• The potential adverse health impacts of summertime heat stress, the ameliorative effects of milder winters upon cold-related mortality, and the likely increased production of particular air pollutants and aeroallergens due to meteorological factors. Current models indicate that, by around 2050, many major cities around the world could be experiencing up to several thousand extra heat-related deaths annually, not taking into account the effects of population growth.
  
  
• The complex ways in which climatic changes would affect the potential transmission of vector-borne diseases (such as malaria, dengue, various other haemorrhagic viruses, and schistosomiasis), and the likely increased occurrence of waterborne and foodborne infections. Recent mathematical models suggest that by 2001 climate change could have increased substantially the proportion of the world's population living in potential malaria transmission zones.
  
  
• The potential impacts of climate change upon agricultural productivity. while temperature increases and soil moisture changes, and shifts in patterns of plant pest and diseases, could lead to decreases in agricultural productivity, carbon dioxide fertilization could lead to some increases in agricultural productivity. But regional variations in gains and losses would probably result in a slight overall decrease in world cereal productivity. Decreases would be most likely in regions that are already food-insecure. The potential effects of changes in ocean temperatures, ocean currents, nutrient flows and surface winds upon aquatic productivity are discussed with particular reference to the simultaneous impacts of human activities on the health of coast ecosystems.
  
  
• Possible changes in extreme weather events, such as heatwaves, floods, storms and droughts, and how these would affect human health. Although uncertain about the specific and local effects of climate change, climatologists anticipate regional increases in the frequency of droughts, and in heavy precipitation events (leading to increased flooding). Health effects could include not only greater risk of death, injury and starvation, particularly among vulnerable Third World populations, but also widespread incidence of psychological and social disorders.
  
  
• The impacts of sea level rise on the health of vulnerable populations. These impacts would include displacement, loss of agricultural land and some fisheries, freshwater salinization, and social disruption, all of which could affect health status adversely. Specific health hazards could arise from heightened storm surges and as a result of damage to coastal infrastructure (including waste-water, sanitation, housing and roads).
  
  
• Potential health effects of increased ground-level exposure to ultraviolet radiation (UVR). Increased UVR levels have become evident in recent years, particularly at mid and high latitudes. If sustained over several decades, these levels will lead to increased rates of skin cancer (especially the non-melanoma cancers) in light-skinned populations, and probably also to increased evidence of eye cataracts, and possibly to suppression of the bodyıs immune system. Additionally, increased UVR levels could have adverse effects on biomass production, and hence on human food production, although the magnitude of such effects is uncertain.

Chapters 9 and 10 address the implications of global climate change for research, monitoring and social-policy response. We must foster interdisciplinary research and techniques adapted for modeling complex processes an the reasonable handling of attendant uncertainties. Human health-related indices should be developed and incorporated into local, regional and global monitoring systems. These indices would embrace environmental signals such as weatherwatch indices, insect population densities and crop production), shifts in the ranges or densities of sensitive indicator species (such as rodents and phytoplankton) and explicit changes in human vulnerability to disease such as poorer nutritional status, UVR-induced tissue damage and altered infectious disease dynamics).

The sustainability of human population health is, of course, a fundamentally important criterion of successful social and economic policy. While there can as yet be no certainty in forecasts of the future health effects of climate change, the role of science in this context must be to promote and assist adoption of precautionary policies that balance current social needs against serious, perhaps unacceptable, future risks.