Although the dramatic climate disruptions of the last glacial period have received considerable attention, relatively little has been directed toward climate variability in the Holocene (11,500 cal yr B.P. to the present). Examination of ?50 globally distributed paleoclimate records reveals as many as six periods of significant rapid climate change during the time periods 9000"8000, 6000"5000, 4200"3800, 3500"2500, 1200"1000, and 600"150 cal yr B.P. Most of the climate change events in these globally distributed records are characterized by polar cooling, tropical aridity, and major atmospheric circulation changes, although in the most recent interval (600"150 cal yr B.P.), polar cooling was accompanied by increased moisture in some parts of the tropics. Several intervals coincide with major disruptions of civilization, illustrating the human significance of Holocene climate variability.

A number of reconstructions of millennial-scale climate variability have been carried out in order to understand patterns of natural climate variability, on decade to century timescales, and the role of anthropogenic forcing. These reconstructions have mainly used tree-ring data and other data sets of annual to decadal resolution. Lake and ocean sediments have a lower time resolution, but provide climate information at multicentennial timescales that may not be captured by tree-ring data. Here we reconstruct Northern Hemisphere temperatures for the past 2,000 years by combining low-resolution proxies with tree-ring data, using a wavelet transform technique to achieve timescale-dependent processing of the data. Our reconstruction shows larger multicentennial variability than most previous multi-proxy reconstructions, but agrees well with temperatures reconstructed from borehole measurements and with temperatures obtained with a general circulation model. According to our reconstruction, high temperatures--similar to those observed in the twentieth century before 1990--occurred around ad 1000 to 1100, and minimum temperatures that are about 0.7 K below the average of 1961-90 occurred around ad 1600. This large natural variability in the past suggests an important role of natural multicentennial variability that is likely to continue.

No abstract

In the northeastern St. Elias Mountains in southern Yukon Territory and Alaska, C14-dated fluctuations of 14 glacier termini show two major intervals of Holocene glacier expansion, the older dating from 3300-2400 calendar yr BP and the younger corresponding to the Little Ice Age of the last several centuries. Both were about equivalent in magnitude. In addition, a less-extensive and short-lived advance occurred about 1250-1050 calendar yr BP (A.D. 700–900). Conversely, glacier recession, commonly accompanied by rise in altitude of spruce tree line, occurred 5975–6175, 4030-3300, 2400-1250, and 1050-460 calendar yr BP, and from A.D. 1920 to the present. Examination of worldwide Holocene glacier fluctuations reinforces this scheme and points to a third major interval of glacier advances about 5800-4900 calendar yrs BP; this interval generally was less intense than the two younger major intervals. Finally, detailed mapping and dating of Holocene moraines fronting 40 glaciers in the Kebnekaise and Sarek Mountains in Swedish Lapland reveals again that the Holocene was punctuated by repeated intervals of glacier expansion that correspond to those found in the St. Elias Mountains and elsewhere. The two youngest intervals, which occurred during the Little Ice Age and again about 2300–3000 calendar yrs BP, were approximately equal in intensity. Advances of the two older intervals, which occurred approximately 5000 and 8000 calendar yr BP, were generally less extensive. Minor glacier fluctuations were superimposed on all four broad expansion intervals; those of the Little Ice Age culminated about A.D. 1500–1640, 1710, 1780, 1850, 1890, and 1916. In the mountains of Swedish Lapland, Holocene mean summer temperature rarely, if ever, was lower than 1°C below the 1931–1960 summer mean and varied by less than 3.5°C over the last two broad intervals of Holocene glacial expansion and contraction.Viewed as a whole, therefore, the Holocene experienced alternating intervals of glacier expansion and contraction that probably were superimposed on the broad climatic trends recognized in pollen profiles and deep-sea cores. Expansion intervals lasted up to 900 yr and contraction intervals up to 1750 yr. Dates of glacial maxima indicate that the major Holocene intervals of expansion peaked at about 200–330, 2800, and 5300 calendar yr BP, suggesting a recurrence of major glacier activity about each 2500 yr. If projected further into the past, this Holocene pattern predicts that alternating glacier expansion-contraction intervals should have been superimposed on the Late-Wisconsin glaciation, with glacier readvances peaking about 7800, 10,300, 12,800, and 15,300 calendar yr BP. These major readvances should have been separated by intervals of general recession, some of which might have been punctuated by short-lived advances. Furthermore, the time scales of Holocene events and their Late-Wisconsin analogues should be comparable. Considering possible errors in C14 dating, this extended Holocene scheme agrees reasonably well with the chronology and magnitude of such Late-Wisconsin events as the Cochrane-Cockburn readvance (8000–8200 C14 yr BP), the Pre-Boreal interstadial, the Fennoscandian readvances during the Younger Dryas stadial (10,850-10,050 varve yr BP), the Alleröd interstadial (11,800-10,900 C14 yr BP), the Port Huron readvance (12,700–13,000 C14 yr BP), the Cary/Port Huron interstadial (centered about 13,300 C14 yr BP), and the Cary stadial (14,000–15,000 C14 yr BP). Moreover, comparison of presumed analogues such as the Little Ice Age and the Younger Dryas, or the Alleröd and the Roman Empire-Middle Ages warm interval, show marked similarities. These results suggest that a recurring pattern of minor climatic variations, with a dominant overprint of cold intervals peaking about each 2500 yr, was superimposed on long-term Holocene and Late-Wisconsin climatic trends. Should this pattern continue to repeat itself, the Little Ice Age will be succeeded within the next few centuries by a long interval of milder climates similar to those of the Roman Empire and Middle Ages.Short-term atmospheric C14 variations measured from tree rings correlate closely with Holocene glacier and tree-line fluctuations during the last 7000 yr. Such a correspondence, firstly, suggests that the record of short-term C14 variations may be an empirical indicator of paleoclimates and, secondly, points to a possible cause of Holocene climatic variations. The most prominent explanation of short-term C14 variations involves modulation of the galactic cosmic-ray flux by varying solar corpuscular activity. If this explanation proves valid and if the solar constant can be shown to vary with corpuscular output, it would suggest that Holocene glacier and climatic fluctuations, because of their close correlation with short-term C14 variations, were caused by varying solar activity. By extension, this would imply a similar cause for Late-Wisconsin climatic fluctuations such as the Alleröd and Younger Dryas.

Tree-ring widths from 880 living, dry dead, and subfossil northern Swedish pines (Pinus syl vestris L.) have been assembled into a continuous and precisely dated chronology (the Torneträsk chronology) covering the period 5407 BC to ad 1997. Biological trends in the data were removed with autoregressive standardization (ARS) to emphasize year-to-year variability, and with regional curve stan dardization (RCS) to emphasize variability on timescales from decades to centuries. The strong association with summer mean temperature (June–August) has enabled the production of a temperature reconstruction for the last 7400 years, providing information on natural summer-temperature variability on timescales from years to centuries. Numerous cold episodes, comparable in severity and duration to the severe summers of the seventeenth century, are shown throughout the last seven millennia. Particularly severe conditions suggested between 600 and 1 BC correspond to a known period of glacier expansion. The relatively warm conditions of the late twentieth century do not exceed those reconstructed for several earlier time intervals, although replication is relatively poor and confidence in the reconstructions is correspondingly reduced in the pre-Christian period, particularly around 3000, 1600 and 330 bc. Despite the use of the RCS approach in chronology construction, the 7400-year chronology does not express the full range of millennial-timescale temperature change in northern Sweden.