Radiocarbon calibration curve variations and their implications for the interpretation of New Zealand prehistory.

Abstract

The shape of a distribution of calibrated (super 14) C dates displays spurious peaks and troughs, brought about by changes in the slope of the calibration curve interacting with the spreading effect of the stochastic distribution of counting errors. The distortion results in a positive correlation between the numbers of dates per calendar year and the slopes of the calibration curves, for assemblages of archaeological dates from such widely separated areas as British Columbia, South Australia and New Zealand. The distortion also increases the possibility of date reversals, and increases the overall spread of calibrated (super 14) C dates. After taking into account this systematic distortion and inbuilt age of charcoal and wood samples, we estimate dates for the initial settlement and first appearance of fortifications, and infer a likely trend of population growth for prehistoric New Zealand.