QSR 1999 paper
Climatic Research Unit : Papers

Annual climate variability in the Holocene: interpreting the message of ancient trees

Keith R. Briffa
Quaternary Science Reviews 19 (2000) 87-105

See also

  1. Some of the chronologies presented in this paper have since been superseded by more recent work:


Over vast areas of the world's landmasses, where climate beats out a strong seasonal rhythm, tree growth keeps unerring time. In their rings, trees record many climate melodies, played in different places and different eras. Recent years have seen a consolidation and expansion of tree-ring sample collections across the traditional research areas of North America and Europe, and the start of major developments in many new areas of Eurasia, South America and Australasia. From such collections are produced networks of precisely dated chronologies; records of various aspects of tree growth, registered continuously, year by year across many centuries. Their sensitivities to different climate parameters are now translated into ever more detailed histories of temperature and moisture variability across expanding dimensions of time and space. With their extensive coverage, high temporal resolution and rigid dating control, dendroclimatic reconstructions contribute significantly to our knowledge of late Holocene climates, most importantly on timescales ranging from 1 to 100 years. In special areas of the world, where trees live for thousands of years or where subfossil remnants of long dead specimens are preserved, work building chronologies covering many millenia continues apace. Very recently, trees have provided important new information about major modes of general circulation dynamics linked to the El Niño/Southern Oscillation and the North Atlantic Oscillation, and about the effect of large volcanic eruptions. As for assessing the significance of 20th century global warming, the evidence from dendroclimatology in general, supports the notion that the last 100 years have been unusually warm, at least within a context of the last two millenia. However, this evidence should not be considered equivocal. The activities of humans may well be impacting on the 'natural' growth of trees in different ways, making the task of isolating a clear climate message subtly difficult.

Figure 1

Figure 1 Northern 'high-latitude' temperature changes over the last 2000 years. The curves show selected reconstructions of summer (annual series a + d) temperatures or temperature-sensitive tree-ring chronologies: (a) Mongolia (Jacoby et al., 1996a); (b) Eastern Siberia (Hughes et al., 1999); (c) eastern Canada (Luckman et al., 1997); (d) North American tree line (D'Arrigo et al., 1992); (e) Northern Sweden (Grudd et al., 1999) with a shorter density-based temperature series superimposed as thin line (Briffa et al., 1992); (f) western Siberia (Hantemirov, 1998 reprocessed) with a nearby density-based temperature series for the northern Urals (Briffa et al., 1995); (g) Central Siberia (Naurzbaev and Vaganov, 1999). All series are plotted as normalised values smoothed with a 50-year low-pass filter. The bottom curve is the average of the other data sets after rescaling to give equal mean and variance (over the common period 1601-1974), also plotted as 50-year smoothed values.

Graph available as GIF (48kb), PostScript (122kb) and PDF (66kb).

Data available as Excel-97 spreadsheet (708kb) and Comma-separated values (300kb)
 2003-10-28: corrected to swap chronology/reconstruction labels for E and F

Selected references

Full reference list appears in the published article.

Last updated: November 2000