February 1, 2002

New Discovery Enhances Dating Accuracy

Print More

Researchers at Cornell and two universities in Europe have discovered that a 1,026 year chunk of Near Eastern tree ring chronology of the Bronze and Iron Ages had been missing 22 years. The researchers came to this conclusion, which was published in the Dec. 21 issue of Science, after carefully comparing data from tree-rings from Germany and Turkey.

The results have implications not only for the history and chronology of the Near East, but for climate models as well.

“Our interpretation of how cultures progress in the Bronze and Iron Ages has changed,” said Prof. Peter Kuniholm, art history and director of Cornell’s Lab of Dendrochronology.

“Any discovery that makes dating more accurate is important, and this is clearly one of these,” added David Owen, the Bernard and Jane Schapiro Professor of Ancient Near Eastern and Judaic Studies.

In their papers, Kuniholm, Maryanne Newton grad, and their collaborators at the University of Reading in England, and the Heidelberger Akademie der Wissenschaften in Germany, detailed discrepancies in the radiocarbon data which prompted them to revise the dates.

Two sets of Turkish juniper tree rings covering 500 years from the first and second millenniums B.C. were analyzed by radiocarbon dating. These sets were compared with a German oak chronology of a comparable time period and known date.

The researchers found that while for the most part the dates agreed, in the younger Turkish data set, there was a consistent carbon-14 difference that made the Turkish wood seem 20 years older than it was.

“Over 500 years there are hardly any differences, but in selected decades, there are weird things going on,” said Kuniholm. These inconsistencies, along with other noise in the younger data set, led the researchers to choose the older of the two Turkish sets to anchor the chronology. This led to the upward revision of 20 years. However, the data could not be made to fit well unless the five data points comprising the offset — which took place from 850-750 B.C.E. — were ignored.

“For direct chronological implications, this [shift] has the most impact for Assyrian chronology,” said Newton, but it also has broad implications for the history of the entire Near East.

Ancient Near Eastern chronology has been difficult because scribes recorded dates in terms of the current ruler or magistrate, whose names and positions in the overall chronology are often not known.

“The whole thing is fraught with uncertainty and aggravation,” Kuniholm said. “What we’re trying to do is make a tree ring chronology that improves on this.”

Sealings and documents who mention rulers or magistrates may now be more definitively dated if they are associated with wood-containing structures.

The dates of several documents, buildings, and artifacts have been directly affected by this change. These include the dates of a bowl with an early alphabetic inscription and of the eruption of the volcano Thera, a pivotal event of the time, whose date has been hotly contested.

“Some people are very happy [with the change] — delighted. Other people are irritated because it messes up their [dating] systems,” said Kuniholm of the reaction to the findings.

“Most of what passes for chronology [in this period] is stylistic dating of pottery. This is a corrective to that kind of thing. The rings don’t lie.”

The discovery that a difference between carbon-14 tree ring dates in different regions of the Northern Hemisphere exists was a major finding itself. It has always been assumed that carbon-14 levels are uniform over the Northern Hemisphere. Trees in different areas should therefore give the same radiocarbon date as long as equivalent rings are being compared.

“It’s weird that there’s a consistent difference. This is the first time that someone has been able to demonstrate that there is such a thing as a regional offset,” said Kuniholm.

“We wondered often if the difference was a mistake in our lab procedures or the procedures of our colleagues,” said Newton, but multiple comparisons convinced them they were right.

To explain the difference, researchers proposed a model based on the timing of growth of trees in the different regions.

Carbon-14 is incorporated into trees in the form of carbon dioxide, but this only happens during periods of growth. German oak grows primarily in mid to late summer, while Turkish pine does so mostly in the spring.

Thus, German oak incorporates more radioactivity and Turkish juniper less. Given two tree rings of the same age, the one with less radioactivity will appear older, which explains the Turkish trees apparent age difference. A similar phenomenon was observed during the “Little Ice Age” of the 16th Century A.D.

“There may be other periods where this is true, but we happened to find this in the period we sampled,” said Kuniholm.

Archived article by Jennifer Frazer