The last decade has seen dramatic advances in the recognition that the paleoenvironmental record is a crucial component of efforts to reduce the uncertainty associated with assessing future climate variability. Major modes of past climate variability have now been explained, and other new ones have been discovered. The paleoclimatic community has drawn attention to many possible climatic ``surprises'' that would have never been recognized from the instrumental record alone. The integrated use of paleoenvironmental data and models has proven useful for testing hypothesized mechanisms of observed change and for investigating the ability of state-of-the-art predictive models to simulate observed climatic change. Surprisingly, much of the observed climatic variability of the past, particularly over decadal to millennial time scales, remains unexplained. This fact presents a major challenge for the future.
Important strides have been made toward understanding and modeling the equilibrium response of the climate system to altered forcing. Increasing attention has also been focused on the need to understand and model the time-dependent behavior of the climate system. The next five years will see increased use of coupled ocean-atmosphere general circulation models to investigate the transient patterns of climatic change [e.g., Delworth et al., 1993; Manabe and Stouffer, 1994]. New efforts will also focus on developing detailed time series of hypothesized forcing, and on using these time series to decompose the observed climate record into natural and non-natural components. A major challenge will be to develop the improved global array of paleoenvironmental data that are needed to unravel natural climate variability. A combination of innovative data analysis and modeling approaches should then yield additional advances in our understanding of climate dynamics.
Human societies will always be susceptible to climate variability, whether it is in the form of disasters such as floods and droughts, or in the form of economic concern over important water, forest, agricultural, or health-related resources that are influenced heavily by climate. A major goal of atmospheric and ocean science research is to provide decision-makers with reliable assessments of future climatic change. Significant progress toward meeting this goal requires that the climate system be studied on all relevant temporal scales. Only through the use of the paleoclimatic perspective can the full range of climatic variability can be identified, understood, and incorporated into the predictive capability required for policy decisions.
Acknowledgments. I thank David Anderson, Julie Cole, Robert Dunbar, Robert S. Webb and one anonymous reviewer for discussions and review of this paper. Tom Karl provided important insights into how the paleoclimate record could be tapped for detecting the cause of environmental trends. The writing of this paper was supported by the NOAA Paleoclimatology Program, and new insights into past climate dynamics were generated using NSF support, particularly to the ARRCC (Analysis of Rapid and Recent Climatic Change) Project.