Oak Dendrochronology: Some Recent Archaeological Developments from an Irish Perspective. (Special Section)

Article excerpt

Introduction

The European oak chronologies were completed back to 5000 BC during the 1980s, with demonstrable replication between Ireland and Germany using stepwise correlation through long English bog-oak series (Pilcher et al. 1984; Baillie 1995). Longer German oak chronologies extend the annual record back to 10,430 BP (Friedrich et al. 1999). This suite of chronologies, and their constituent site chronologies, even individual trees, can be analysed for changes in common response to the environment through time and of course can be compared with other well-dated time series from around the world.

The single most important research application of the oak chronologies was calibration of the radiocarbon timescale, and the results of the various calibration exercises were initially published by Pearson & Stuiver (1986; 1993; Stuiver & Pearson 1986; 1993). When these calibration exercises were performed, it was believed by the main protagonists, Stuiver & Pearson, that radiocarbon was swiftly mixed in the atmosphere; the expectation being that organic samples of the same calendar age, from anywhere in the northern hemisphere, would yield the same radiocarbon age (it was always recognized that the radiocarbon dates from the southern hemisphere were systematically offset from the northern results, though it was not known whether the off-set was constant or time transgressive). Thus, it was assumed that once the two principal high precision calibration laboratories, in Seattle and Belfast, had each derived their calibration results the two curves would be identical within the limits of the measurement errors. Sure enough, when the definitive calibration curves were first published in 1986 it was observed that in comparing some 4500 years of calibration results the Belfast and Seattle results showed a mean difference of only 0.6 years (Pearson et al. 1986). As a result it was widely assumed that calibration was at an end, the definitive calibration curve was published and calibration was effectively over; under this paradigm dendrochronologists no longer had a role to play in radiocarbon issues. Nothing could have been further from the truth. Not only have revised calibrations been produced on a regular basis, but, following up hints in the original calibration results, new sets of measurements on known-age wood established that in the radiocarbon calibration curves there can be short-term regional off-sets; same age samples giving systematically different radiocarbon ages by up to a few decades (McCormac et al. 1995). While this has little effect in the understanding of overall archaeological chronology, it can have profound effects when it comes to detailed chronological questions involving so called `wiggle matching' calibration exercises, using multiple, stratified, radiocarbon determinations, and in comparisons with truly absolute chronologies. In addition, dendrochronologists have supplied known-age samples from the British Isles and New Zealand, for simultaneous radiocarbon measurement in two laboratories, in order to check the interhemispheric off-set through time. In this case the results have shown a time transgressive shift in off-set with potentially important implications for studies of past ocean circulation and carbon cycle studies (McCormac et al. 1998). Calibration is thus still an important application of dendrochronology, and it is worth mentioning here that the full implications of regional calibration off-sets have not yet been fully appreciated by the archaeological community.

Commercial dating and its results

The availability of long oak chronologies across northern Europe was instantly accepted by the archaeological community with the result that dendro dating rapidly became `routine'. In Ireland, right from the start, the distribution of real dates presented workers with a bit of a surprise. By the mid 1980s, when the Irish oak chronology was completed, only six prehistoric archaeological sites had contributed timbers to the chronology. …