The abrupt circum-North Atlantic climate oscillation associated with the Younger Dryas event of Europe has been the most intensely studied century-scale event, in part because of its large amplitude in a well-studied region, but also because the event has much to reveal about the dynamics of the coupled atmosphere-ocean-ice system. Ice core, marine sediment, and terrestrial proxy data suggest that the Younger Dryas cold event abruptly terminated the relative warmth of the Bø lling/Allerø d period, lasted about 1300 years, and changed from near glacial to interglacial conditions in less than a century at approximately 10 ka (ca. 11.5 ka calendar years; Dansgaard et al., 1989; Alley et al, 1993]. Data from high accumulation-rate sediment and ice cores also indicate that the atmospheric cold event was associated with abrupt shifts in atmospheric ocean circulation [ Broecker et al., 1989; Broecker and Denton, 1989; Keigwin and Jones, 1994; Lehman and Keigwin, 1992; Mayewski et al, 1993; Taylor et al., 1993; Charles et al., 1994]. Whereas it appears that the circum-North Atlantic Younger Dryas event resulted from the effects of glacial meltwater on ocean thermohaline circulation and dramatic decrease of northward heat transports [ Fairbanks, 1989; Broecker and Denton, 1989; Lehman, in press], there remains significant debate on how this event may have been expressed at lower latitudes and outside the circum-North Atlantic region [ Overpeck et al., 1989; Oglesby et al., 1989; Engstrom et al., 1990; Markgraf, 1993; Heusser, 1993; White et al., 1994].
The Younger Dryas event was one of the many major abrupt century-scale warm-cold oscillations that took place in the circum-North Atlantic region during the last glacial period. These so-called ``Dansgaard-Oeschger events'' have been clearly recognized in both the new Greenland ice cores and in the sediments of the North Atlantic, and they represent a climate system oscillating between significantly different states [ Broecker and Denton, 1989; Bond et al., 1993; Taylor et al., 1993; Grootes, et al., 1993; Mayewski et al., 1994]. These events, apparently related to reorganization of the ocean-atmosphere system, also seem to group into longer oscillations, each terminating in massive discharges of icebergs into the North Atlantic. These discharge events, known now as ``Heinrich events,'' appear to be related to instabilities in the glacial ice sheets, particularly the North American Laurentide Ice Sheet, and result in abrupt changes in oceanic circulation and circum-North Atlantic climate [ Miller and Kaufman, 1990; Bond et al., 1992; 1993; Grimm et al., 1993; MacAyeal, 1993; Keigwin and Lehman, 1994; Andrews et al., 1994; Lehman, in press].
Most global change scientists were surprised with the paleoclimatic observations that the climate system, including the deep circulation of the oceans, can switch abruptly from one state to a dramatically different one. At first, these observations were restricted to cold glacial periods, but recent results from the GRIP (Greenland Ice Core Project) ice core further surprised scientists with the possibility that similarly dramatic (5-10 C) abrupt oscillations might have characterized the last interglacial [ Dansgaard et al., 1993]. Since the last interglacial was somewhat warmer than the present Holocene interglacial, these ice-core observations led to the speculation that a climate warmer than today's (i.e., a greenhouse-gas warmed Earth) might be unstable. However, the pattern of abrupt events observed in the GRIP core was not observed in the GISP2 (Greenland Ice Sheet Project 2) core only 30km away, suggesting that the events were an artifact of glacial ice dynamics, and not indicative of true climatic change [ Grootes et al., 1993; Taylor et al., 1993].
The interglacial ice core debate contributed to the misconception that interglacials are climatically stable, in sharp contrast to the surprisingly unstable glacial periods. However, as highlighted by Broecker [1987], we don't have an observational basis for anticipating how stable a greenhouse-gas warmed climate system will be. Investigation of the abrupt oscillations of the last glacial, even if they are ice-sheet dependent, reveals much about how the coupled atmosphere-ocean-ice system operates. A growing body of data from low to middle latitudes suggests that the present interglacial (Holocene) has also been far from stable. An increasing number of paleohydrologic records from lower latitudes indicate that abrupt shifts in drought and flood frequency are characteristic of warm (interglacial) climates [e.g., Hodell et al., 1991; Ely et al., 1993; Knox, 1993; White et al., 1994]. It also appears that the interglacial tropical atmosphere-ocean system can vary in more ways than indicated by the short instrumental time series that are available [ Overpeck et al., 1989; Peterson et al., 1991; Cole et al., 1993; Dunbar et al., 1994; Quinn et al., 1993]. Perhaps most surprising, however, is how little we understand the patterns or causes of the natural interglacial variability on which future change will be superimposed.