W. Jack Rink, Associate Editor
Cautions for Use of Thermoluminescence Dating of Sediment
The use of thermoluminescence (TL) dating of sediments at the sites of Jinmium (Fullagar et al. 1996) and Diring Yuriakh (Waters 1997) have yielded shockingly old ages that have created a storm in Palaeolithic archaeology (Gibbons 1997). At Jinmium, the results suggest that the peopling of Australia occurred about 110,000 years ago, more than twice as old as previously believed (Roberts et al. 1990). Similarly, the site of Diring Yuriakh in Siberia, a site regarded by many to not be any older than about 30,000 years has been dated to greater than about 250,000 years. In contrast to dating of sediments where optical exposure resets the clock, the more conventional applications of thermoluminescence dating of burned flint and pottery operate on the basis of resetting by heating. The great advantage of the latter methods is that heat is a potent agent to completely remove the pre-existing (geological) TL signals, whereas solar resetting can only partially deplete those TL signals in quartz and feldspar grains in sediments. Most workers agree that the TL results for Jinmium and Diring Yuriakh need confirmation using modern optical luminescence dating methods.
Recent improvements in luminescence dating technology have greatly enhanced our ability to date sedimentation events in this time range through the application of optical luminescence (OSL) dating, which is rapidly replacing the older technology of TL dating. Natural light exposure in air will only reduce the TL signal to a residual non-zero value. The size of the residual TL signal is dependent upon the type and duration of light exposure. Solar resetting of the TL signal is most efficient when the proportion of ultraviolet radiation is large, whereas low levels of visible light and ultraviolet light will be the least efficient. Full sunlight has the highest UV levels, while daylight under cloud cover has less UV and underwater light has the least UV and reduced levels of visible light. Turbidity also reduces visible light in air and water by scattering processes. Solar resetting of OSL signals in quartz and infrared-stimulated luminescence (IRSL) in feldspar and quartz also obey the same general rules, but relative to TL signals, the rate of depletion is much more rapid for OSL and IRSL signals. In full sunlight this can occur in seconds to minutes. Furthermore, OSL and IRSL signals reach a true zero value rather than the non-zero value for TL in quartz and feldspar.
What does all this mean for the archaeologist who is trying to interpret the significance of dating results? It means that one can be more certain that OSL or IRSL signals were depleted to a known (zero) level than for TL signals under the same set of ancient conditions. Even underwater conditions are known to yield complete zeroing of OSL signals. Aeolian sediments are the best, while those deposited in water are less preferable. Moreover, there are techniques available in IRSL and OSL to test whether partial resetting of sediments occurred (e.g. Clarke 1996).
So, it seems that any TL date on sediment should at least be verified by IRSL or OSL. But there are also other problematic aspects to the use of these methods. As with radiocarbon dating, the question arises: What is the event that is being dated? Some have argued that artifacts can move down through sandy sediments, and therefore the age of the sediment may be older than that of the artifacts. Another problem is that burrowing rodents can move sediment and artifacts through a vertical dimension, and then the effects of time can obscure the evidence of the burrows. Certainly though, evidence of burrows in the youngest part of a site are a clue that past burrowing might have created disturbances in the archaeological and sedimentary record.
How can controversial dating results be best debated among dating experts and users? I believe that no matter how modern the dating technology, the ultimate test of a date is whether it can be reproduced by an independent lab with access to the original site. Reassessment of the original context is essential, because without additional dating and reconsideration of the depositional context, informed scientific debate cannot easily develop. After all, if a date cannot be reproduced by an independent group of researchers, then there is a basis for debate. But if a date stands up to critical appraisal, then there can be greater confidence that new ground has been broken, and Science can move forward. Are the dates for Jinmium and Diring Yuriakh right? Only time will tell, but see Roberts (1998) for an appraisal.
Fullagar, R.L.K., Price, D.M. and Head, L.M. 1996. Early human occupation of northern Australia: archaeology and thermoluminescence dating of Jinmium rock-shelter, Northern Territory. Antiquity
Clarke, M.L. 1996. IRSL dating of sands: bleaching characteristics at deposition using single aliquots. Radiation Measurements 26: 611-620.
Gibbons, A. 1997. Doubts over spectacular dates. Science 278: 220-222.
Roberts, R.G., Jones, R. and Smith, M.A. 1990. Thermoluminescence dating of a 50,000 year-old human occupation site in northern Australia. Nature 345: 153-156.
Roberts, R.G. (1998) Luminescence dating in archaeology: from origins to optical. Radiation Measurements (in press).
Waters, M.R., Forman, S.L. and Pierson, J.M. 1997. Diring Yuriakh: a Lower Paleolithic site in central Siberia. Science 275: 1281-1284.