Executive Summary

The average surface temperature of the globe has warmed appreciably since the late 1800s, by about 0.6°C. Since this warming cannot be adequately explained by natural phenomena such as increased solar activity, human-induced increases in greenhouse-gas concentrations appear to be at least partly responsible. In addition to the warming effect of greenhouse-gas increases, however, changes in temperature over the past century are likely to have been significantly influenced by the cooling effect associated with changes in the sulfate aerosol loading of the atmosphere, arising from fossil-fuel-derived sulfur dioxide (SO₂) emissions. When greenhouse-gas, sulfate aerosol, and solar influences are considered together, observed climate changes are consistent with model predictions.

Projections of future global-mean temperature and sea level change made by the Intergovernmental Panel on Climate Change (IPCC) in its 1996 Second Assessment Report used emissions scenarios developed in 1992. Preliminary versions of new emissions scenarios produced by the writing team for the IPCC Special Report on Emissions Scenarios (SRES) are now available. The most important difference between the old (1992) and new (SRES) scenarios is that the new scenarios have much lower emissions of sulfur dioxide. The reduction in sulfur dioxide emissions (and their attendant cooling effects through the production of sulfate aerosols) results in a slight increase in temperature and sea level rise projections from those previously given by the IPCC. If central estimates of model parameters are used, global-mean warming from 1990 to 2100 ranges from 1.9°C to 2.9°C. Sea-level rise estimates over the same period range from 46 to 58 cm. For temperature and sea level changes over the next few decades, projections are virtually independent of the emissions scenario.

Based on results from a number of climate models, the rate of future warming over the United States is expected to be noticeably faster than the global-mean rate. Future regional-scale precipitation changes are highly uncertain. The only result that is common to all climate models is an increase in winter precipitation in northern latitudes, from the northern Great Plains to the northeastern states. Even in the absence of large precipitation changes, there could still be significant changes in the availability of water for agriculture, human consumption, and industry because of the increased evaporation that should accompany warming. This factor alone would lead to drier summer soil conditions and reduced runoff. The effects of increased evaporation, however, may be partly offset by the direct plant-physiological effect that carbon dioxide (CO₂) has in improving plant water-use efficiency and, hence, lowering evapotranspiration rates.

Changes in weather and climate extremes over the United States are certain to occur as the global climate changes. The frequency of extremely hot days is almost certain to increase, and the frequency of frosts should decrease. Changes in the frequency of daily precipitation extremes are highly uncertain, although there is evidence for an increase in the frequency of wet extremes. For hurricanes and tropical storms, the evidence suggests that there could be small increases in their windspeeds. It is also likely that future such storms will be accompanied by larger rainfall amounts. While there is no credible model-based information on changes in the number of hurricanes and tropical storms per year worldwide, there is empirical evidence that suggests that a small increase in frequency is possible in the North Atlantic region. For all extreme events, however, it is unlikely that the projected changes will become evident in a statistically convincing way for many decades, with the exception of temperature extremes, which should become evident sooner.