lack of understanding of the processes that control them. For example, mechanisms proposed to explain abrupt climate shifts do not fully describe the patterns of variability seen in either the paleoclimate or the historical records.
Long-term geological records show that in the past there were different stable states of the climate system from those of today. Differences in these climate states involved the coupled atmosphere, ocean, ice, and biological systems. Some of these old states, such as the proposed “snowball earth” (when most or all of the planet was frozen) (Harland, 1964; Caldeira and Kasting, 1992; Kirschvink, 1992; Hoffman et al., 1998), occurred long ago when geological and astronomic conditions were substantially different from today. However, others, such as the warm-polar “hothouse” pattern, were reached relatively recently, when geological conditions were similar to those of our modern earth (Barron, 1987). Despite the recognition that extreme shifts in the climate system can occur, little information is available on whether transitions between climate states are possible under modern or near-future conditions and whether such transitions would be abrupt.
Understanding how and why climate might change abruptly has important implications. The just-completed US National Assessment (National Assessment Synthesis Team, 2000) emphasizes the possible effects of gradual climate change on societies and ecosystems, and it concludes that effects will probably be larger in the case of faster changes that leave less time for adaptation or that give less warning that might make mitigation possible. Thus, when considering the possible causes and impacts of abrupt climate change, numerous important questions arise, including:
What caused the large, widespread, abrupt climate changes of the past? Could they recur? Might human activities affect the possibility of recurrence?
To what extent is abruptness a fundamental characteristic of regional and global climate changes? Might future regional changes be abrupt?
Could global or regional climates shift into modes different from those observed recently, such as the warm-period modes recorded in geological archives? Might such a shift be abrupt?
How do societies and ecosystems maintain resilience and adaptability? How might these be enhanced in the face of severe tests posed by abrupt climate changes?
Answers to these and related questions are best guided by a focused re-