Climate change is a threat to mankind. But no one is certain about its future effects or their severity. Responding to the threat is expected to be complicated and difficult. There is even some remaining disagreement over whether any problem exists at all: while many people worry that the effects will be extremely serious, others still argue that scientists cannot prove that what they suspect will happen will actually happen. In addition, it is not clear who (in the various regions of the world) will suffer most. Yet if the nations of the world wait until the consequences and victims are clear, it will probably be too late to act. What should we do?

The truth is that in most scientific circles the issue is no longer whether or not climate change is a potentially serious problem. Rather, it is how the problem will develop, what its effects will be, and how these effects can best be detected. Computer models of something as complicated as the planet’s climate system are not far enough advanced yet to give clear and unambiguous answers. Nevertheless, while the when, where, and how remain uncertain, the big picture painted by these climate models cries out for attention.

For example:

• Regional rain patterns may change. At the global level, the evapotranspiration cycle is expected to speed up. This means that it would rain more, but the rain would evaporate faster, leaving soils drier during critical parts of the growing season. New or worsening droughts, especially in poorer countries, could reduce supplies of clean, fresh water to the point where there are major threats to public health. Because they still lack confidence in regional scenarios, scientists are uncertain about which areas of the world risk becoming wetter and which drier. But with global water resources already under severe strain from rapid population growth and expanding economic activity, the danger is clear.

• Climate and agricultural zones may shift towards the poles. In the mid-latitude regions the shift is expected to be 150 to 550 kilometres for a warming of 1-3.5ºC. Increased summer dryness may reduce mid-latitude crop yields, and it is possible that today’s leading grain-producing areas (such as the Great Plains of the United States) would experience more frequent droughts and heat waves. The poleward edges of the mid-latitude agricultural zones - northern Canada, Scandinavia, Russia, and Japan in the northern hemisphere, and southern Chile and Argentina in the southern hemisphere - might benefit from higher temperatures. However, in some areas rugged terrain and poor soil would prevent these countries from compensating for reduced yields in today’s more productive areas.

• Melting glaciers and the thermal expansion of sea water may raise sea levels, threatening low-lying coastal areas and small islands. The global mean sea level has already risen by around 10 to 15 centimetres during the past century, and global warming is expected to cause a further rise of 15 to 95 cm by the year 2100 (with a "best estimate" of 50 cm). The most vulnerable land would be the unprotected, densely populated coastal regions of some of the world’s poorest countries. Bangladesh, whose coast is already prone to devastating floods, would be a likely victim, as would many small island states such as the Maldives.

These scenarios are alarming enough to raise concern, but too uncertain for easy decisions by governments. The picture is fuzzy. Some governments, beleaguered by other problems and responsibilities and bills to pay, have understandably been tempted to do nothing at all. Maybe the threat will go away. Or someone else will deal with it. Maybe another giant asteroid will hit the earth. Who knows?

• It establishes a framework and a process for agreeing to specific actions later. The diplomats who wrote the Framework Convention on Climate Change saw it as a launching pad for potential further action in the future. They recognized that it would not be possible in the year 1992 for the world’s governments to agree on a detailed blueprint for tackling climate change. But by establishing a framework of general principles and institutions, and by setting up a process through which governments meet regularly, they got things started.

A key benefit of this approach is that it allowed countries to begin discussing the issue even before they all fully agreed that it is, in fact, a problem. Even skeptical countries have felt it is worthwhile participating. (Or, to put it another way, they would have felt uneasy about being left out.) This created legitimacy for the issue, and a sort of international peer pressure to take the subject seriously.

The Convention is designed to allow countries to weaken or strengthen the treaty in response to new scientific developments. For example, they can agree to take more specific actions (such as reducing emissions of greenhouse gases by a certain amount) by adopting "amendments" or "protocols" to the Convention. This is what happened in 1997 with the adoption of the Kyoto Protocol.

The treaty promotes action in spite of uncertainty on the basis of a recent development in international law and diplomacy called the "precautionary principle." Under traditional international law, an activity generally has not been restricted or prohibited unless a direct causal link between the activity and a particular damage can be shown. But many environmental problems, such as damage to the ozone layer and pollution of the oceans, cannot be confronted if final proof of cause and effect is required. In response, the international community has gradually come to accept the precautionary principle, under which activities that threaten serious or irreversible damage can be restricted or even prohibited before there is absolute scientific certainty about their effects.

• The Convention takes preliminary steps that clearly make sense for the time being. Countries ratifying the Convention - called "Parties to the Convention" in diplomatic jargon - agree to take climate change into account in such matters as agriculture, energy, natural resources, and activities involving sea-coasts. They agree to develop national programmes to slow climate change. The Convention encourages them to share technology and to cooperate in other ways to reduce greenhouse gas emissions, especially from energy, transport, industry, agriculture, forestry, and waste management, which together produce nearly all greenhouse gas emissions attributable to human activity.Countries ratifying the Convention - called "Parties to the Convention" in diplomatic jargon - agree to take climate change into account in such matters as agriculture, energy, natural resources, and activities involving sea-coasts. They agree to develop national programmes to slow climate change. The Convention encourages them to share technology and to cooperate in other ways to reduce greenhouse gas emissions, especially from energy, transport, industry, agriculture, forestry, and waste management, which together produce nearly all greenhouse gas emissions attributable to human activity.

• The Convention encourages scientific research on climate change. It calls for data gathering, research, and climate observation, and it creates a "subsidiary body" for "scientific and technological advice" to help governments decide what to do next. Each country that is a Party to the Convention must also develop a greenhouse gas "inventory" listing its national sources (such as factories and transport) and "sinks" (forests and other natural ecosystems that absorb greenhouse gases from the atmosphere). These inventories must be updated regularly and made public. The information they provide on which activities emit how much of each gas is essential for monitoring changes in emissions and determining the effects of measures taken to control emissions.