Future greenhouse gas emissions will depend on global population, economic, technological, and social trends. The link to population is clearest: the more people there are, the higher emissions are likely to be. The link to economic development is less clear. Rich countries generally emit more per person than do poor countries. However, countries of similar wealth can have very different emission rates depending on their geographical circumstances, their sources of energy, and the efficiency with which they use energy and other natural resources.

As a guide to policymakers, economists produce "scenarios" of future emissions. A scenario is not a prediction. Rather it is a way of investigating the implications of particular assumptions about future trends, including policies on greenhouse gases. Depending on the assumptions (which may be quite wrong), a scenario can project growing, stable, or declining emissions.

Four storylines have recently been developed as a basis for producing scenarios. The resulting four scenario "families" contain a total of 40 individual scenarios. One storyline describes a future world marked by very rapid economic growth, a population that peaks in mid-century and declines thereafter, and the rapid introduction of new and more efficient technologies. A second storyline is similar but assumes a rapid transition towards a cleaner economy based on services and information. A third describes a world where population continues to increase, economic development trends are regional rather than global, and per-capita economic growth and technological change are slower and more fragmented. A fourth emphasizes local and regional solutions to sustainability, with a slowly but steadily growing population and medium economic development. None of these scenarios explicitly assumes that the Climate Change Convention is implemented or that policies are adopted to achieve the Kyoto Protocol’s emissions targets. Nevertheless, they do include scenarios where there is less emphasis on fossil fuels than at present.

The future concentrations of greenhouse gases and aerosols resulting from these storylines vary widely. climate model results Depending on the assumptions used, "non-intervention" models predict that – in the absence of new climate change policies to reduce emissions – For example, carbon cycle models project atmospheric concentrations of carbon dioxide for the year 2100 will range of from 490 to 1,260 parts per million. This represents anywhere from a 75 to 350% increase over pre-industrial levels. Projected changes in methane range from –10% to +120%, and increases in nitrous oxide range from 13 to 47%.

"Intervention" scenarios are designed to examine the impact of efforts to reduce greenhouse gas emissions. They depend not only on assumptions about population and economic growth, but also about how future societies will would respond to the introduction of climate change policies such as taxes on carbon-rich fossil fuels.

Existing international commitments could slightly reduce the rate of growth in emissions. Under the Climate Change Convention and its Kyoto Protocol, developed countries are to reduce their greenhouse gas emissions to 1990 levels and to 5% below these levels, respectively. Such commitments are important first steps, but they will make only a small contribution towards the ultimate goal of stabilizing greenhouse gas concentrations in the atmosphere.A goal of making more substantial reductions in atmospheric concentrations would clearly require all countries to make dramatically larger cuts in emissions.

Stabilizing greenhouse gas concentrations will require a major effort. Stabilizing carbon dioxide concentrations at 450 ppm (some 23% above current levels) would require global emissions to drop below 1990 levels within a few short decades. Stabilizing CO₂ at 650 ppm or 1,000 ppm would require the same emissions decline within about one century or two centuries, respectively, with continued steady declines thereafter. Eventually CO₂ emissions would need to decline to a very small fraction of current levels – despite growing populations and an expanding world economy.

Reducing uncertainties about climate change impacts and the costs of various response options is vital for policymakers. Stabilizing or reducing emissions world-wide would have consequences for almost every human activity. To decide if it is worthwhile, we need to know how much it would cost, and how bad things will get if we let emissions grow. There are tough moral questions too: how much are we prepared to pay for the climate of the 22nd century, which only our children's children will see?