Large Stratospheric Cooling (virtually certain)

The combination of reduced ozone and increased carbon dioxide will lead to a

marked lowering of upper stratospheric temperatures, within the range of 15o. 30° fahrenheit.

Long Time to Draw Down Excess Carbon Dioxide (virtually certain)

We know that it takes decades to centuries to produce the buildup of greenhouse gases. Much less appreciated is that a "return to normal" from high carbon dioxide levels would require many additional centuries.

Global Surface Warming over the Past Century (virtually certain)

The observed warming in the surface temperature records of about one degree fahrenheit cannot yet be unambiguously ascribed to greenhouse warming. However, no other hypothesis is nearly as credible.

Natural Variability Adds Confusion (virtually certain)

Climate varies naturally on time scales from months to centuries. This effect makes a greenhouse warming interpretation of the climate record much more difficult.

Global-Mean Surface Warming (Very Probable)

For the middle of the next century, global-mean surface warming is estimated to be in the range of 2 to 6° fahrenheit. The largest uncertainty is due to the effects of clouds. My own opinion is the warming is more likely to be in the lower half of this range, mainly due to cooling offsets from sulfate aerosols and ozone depletion.

Global-Mean Precipitation Increase (Very Probable)

As the climate warms, the rate of evaporation should increase, leading to an increase in global-mean precipitation. Nevertheless, some local regions would experience decreases in precipitation.

Reduction of Northern Sea Ice (Very Probable)

As the climate warms, total sea ice should decrease in high latitudes of the Northern Hemisphere. However, recent atmosphere-ocean models calculate a resistance to climate change at high Southern Hemisphere latitudes. Thus, little change in sea-ice cover may occur there over the next century.

Rise in Global Mean Sea Level (Very Probable)

A further rise of 4-12 inches in mean sea level by the year 2050 is estimated

due to thermal expansion of warmer sea water. Continued sea level rise is expected for many centuries, probably to much higher values.

Summer Mid-Continental Dryness and Warming (Probable)

Model studies predict a marked decrease of soil moisture and elevated temperatures over summer mid-latitude continents. This result is sensitive to model assumptions about land-surface processes and projected levels of sulfate aerosol pollution.

Reduction of Ocean Overturning (Probable)

The predicted increase in precipitation in high northern latitudes would reduce the salinity, and hence the density in the oceans there. This effect slows down the tendency for water to sink, thus suppressing the overturning circulation.

Modeling of Cloud-Radiation Feedbacks (uncertain)

Some progress is being made in modeling the cloud-radiation response to increasing greenhouse gases. However, understanding the cloud response is still the most serious barrier to more confident predictions about climate warming.

Increased Tropical Storm Intensities (uncertain)

Some calculations suggest that a warmer, wetter atmosphere could lead to increased intensities of tropical storms, such as hurricanes. Presently, this has not been properly addressed due to inadequate computer power and to uncertainties in regional climate change predictions.

Regional and Temporal Details of the Next 25 Years (uncertain)

I have described calculated changes averaged over decades in the middle of the next century. The predicted warming up to now is not yet large compared to natural climate fluctuations. Such natural fluctuations will continue to occur. On these shorter time scales, the natural fluctuations can artificially reduce or enhance apparent measured greenhouse warming signals.

Even though these uncertainties are daunting, important advances have already been achieved in observing, understanding, and modeling the climate. Today's models can simulate many aspects of climate and its changes. Indeed, it was these advances that led to today's increased awareness of climate change and its implications. This awareness has, in turn, escalated the need for more reliable climate predictions.

Although major progress has been made, much more needs to be learned.

More focussed efforts are needed world-wide to provide a long-term climate measuring system. This must be backed by a commitment to continue such measurements over many decades.

Focussed research into climate processes must be continued. Theories must

be formulated and re-evaluated in the light of newer data. To reduce uncertainty, climate modeling efforts must receive resources that are in balance with the broader scientific programs.

The U.S. Global Change Research Program is making excellent progress on these fronts. However, sustained efforts will be required in the years ahead. This is particularly true for measuring climate change. This needs a long-term commitment that is not yet evident. Without a better climate-change measuring system, neither our research nor our predictions can be properly evaluated.

Personally, I believe that the state of knowledge of the wide range of possible impacts and costs of climate change is far less certain than are the predictions for the climate system. An investment for research in this climate impact arena will pay large dividends and greatly aid policy planning.