Counting years in wood
In many climates a tree adds a layer of new wood each growing season, visible in a cross-section as a ring. Counting the rings gives the age of the tree. Dendrochronology, the scientific study of tree rings, goes much further, because the rings are not all the same. Their width varies with weather, particularly rainfall and temperature, and trees of the same species growing in the same region respond to the same good and bad years.
The result is a pattern of wide and narrow rings that is shared by many trees, like a barcode. By matching the barcodes of living trees with those of older timbers, and then older ones still, researchers can build a continuous record that extends far beyond the life of any single tree. A piece of wood whose pattern fits a well-established record can be placed at an exact calendar year.
Origins of the method
The method was developed in the early twentieth century by the American astronomer Andrew Ellicott Douglass, who was interested in whether tree rings reflected cycles of sunspots and climate. At the University of Arizona he built sequences from living pines in the American Southwest and then turned to the beams in ancient Pueblo buildings, such as the great houses of Chaco Canyon. For years he had a sequence from living trees and a separate “floating” sequence from old beams, with an unknown gap between them. In 1929 a charred beam found at Show Low, Arizona, bridged the gap and fixed the ancient sequence to the calendar. Tree-ring dates are now an essential part of the study of the Ancestral Puebloan migration, helping to show when the major pueblos were built and when areas such as Mesa Verde were left, in the late thirteenth century CE.
Later work extended chronologies in other regions. Oak and pine sequences in central Europe, built from living trees, timbers in buildings and wood preserved in river gravels and bogs, reach back well over ten thousand years. Irish and British oak chronologies reach back several thousand years, and long records also exist for parts of Scandinavia. Researchers working in the Aegean and Anatolia have built sequences from timbers at archaeological sites, though the Mediterranean is a more challenging region because its trees respond less strongly to a single climate signal.
How cross-dating works
The central technique is cross-dating, matching the ring patterns of different samples. First, a sample is measured, often to a fraction of a millimetre, and its pattern of widths is recorded. Computers compare the pattern statistically with existing records, and the best match is checked by eye and tested against other samples.
A single tree can be misleading, because a missing or double ring happens occasionally. Using many trees and requiring that patterns agree across them guards against such errors. Species and region matter, since a master chronology for oak in one area cannot date pine from another. New regions need new reference sequences built from the ground up.
What a date actually means
A tree-ring date applies to the ring that was measured, so the details matter. If the outermost ring just under the bark is preserved, the date indicates the year, and often the season, when the tree was felled. If the outer rings were trimmed away, the date gives only a point before which the tree could not have been cut, a “terminus post quem”, and the felling date is later by an unknown number of years.
Even a precisely dated beam does not give the date of the building by itself. Timbers were stockpiled, reused from earlier structures, or replaced in repairs. Archaeologists therefore look for groups of timbers that were felled within a few years of each other and interpret the result in context.
Calibrating radiocarbon
Perhaps the most important role of dendrochronology is to calibrate radiocarbon dating. Because each ring is dated by counting, wood of known calendar age can be measured for carbon-14. Plotting radiocarbon ages against known ages produced the calibration curves that convert radiocarbon measurements into calendar dates. For the period covered by the long European and American chronologies, the curve rests largely on tree rings, and for earlier periods it draws on other archives such as corals and lake sediments. It is why a radiocarbon age is no longer taken at face value.
Climate and history
Tree rings also record environmental conditions. Wide rings tend to follow favourable years, and narrow rings point to stress such as drought or cold. Historical climate reconstructions use this, and some tree-ring records have been linked to major volcanic eruptions that cooled summers.
Such links have sometimes been used in debates about ancient events. Researchers have noted growth anomalies in some tree-ring records of the seventeenth century BCE and suggested a connection to a large eruption, which has been discussed alongside the date of the eruption at Akrotiri on Thera. The case is debated, because the dating of the eruption itself is contested and tree-ring anomalies can have several possible causes.
Limits of the method
Not every place can be dated this way. Tropical trees may not form clear annual rings, wood must survive, and timber must be large enough to show a long sequence, usually at least several decades of rings. In arid regions, little wood survives. Ancient Mesopotamia, built largely of mudbrick and in a region with few suitable trees, offers little for dendrochronology compared with the forested north.
Where it works, however, tree-ring dating is among the most precise methods available. It gives exact years rather than ranges, tests radiocarbon results, and ties archaeological sites to the calendar. In combination with radiocarbon measurements and stratigraphy, it helps hold together the chronologies on which much of archaeology depends.