Rather than supporting the out-of-date concept of major centuries-long
globally-synchronous events, available data for the last millennium indicate
that multi-decadal, regional to continental-scale temperature
[4]
anomalies on
the order of 0.5 C were common [ Bradley and Jones, 1993; Hughes
and Diaz, 1994]. The data are also beginning to suggest that one of the
largest and most extensive temperature shifts of the past 500 to 1000 years
occurred between A.D. 1850 and today. Patterns of temperature variation over
the past 1000 years are generally compatible with one or a combination of
several hypothesized causes [ Rind and Overpeck, 1993;
Crowley and Kim,
1993]:
Hypothesized variations in solar output, in particular, are receiving renewed attention as a possible explanation for the decade- to century-scale climatic variations of the past millennium, but the warming during the last century seems more compatible with trace-gas forcing than with solar variations [ Hansen and Lacis, 1990; Stuiver and Braziunas, 1993; Eddy and Oeschger, 1993b; Rind and Overpeck, 1994; Lean and Rind, 1994].
Most attention has focused on temperature change of the last centuries, spurred on by the issue of whether human-induced warming can be identified above the background levels of natural variability. The susceptibility of society to droughts and floods has also led to investigations of interannual to century-scale hydrological variability. Recent paleohydrological investigations suggest that the variability witnessed over the recent period of instrumental observations is not unusual, and is often an underestimate of true natural levels of variability [ Hughes and Brown, 1992; Stahle and Cleaveland, 1992; Graumlich, 1993; Fritz et al., 1994; Stine, 1994]. Paleoclimatic studies of the Pacific and North Atlantic climate systems offer further evidence that the true range of interannual to decadal variability is more likely than not to be poorly represented in the instrumental record [ D'Arrigo et al., 1992, Cole et al., 1993; Dunbar et al., 1994; Quinn et al., 1994]. As with the crudely understood nature of century-scale climatic variability, a large amount of new data and improved climate theory will be necessary to determine the exact range and causes of interannual to decadal climatic variability.