Detecting spatially local deviations in population change using summed probability distribution of radiocarbon dates

Summary

The increasing availability of large radiocarbon databases encompassing continental geographic scales (e.g. CARD, EUROEVOL, AustArch, etc.) is now opening new possibilities for evaluating spatial variation in prehistoric population. We have, for the first time, the opportunity to determine whether and when different geographic regions experienced distinct demographic patterns using an absolute chronological framework. This line of research is however hindered by spatially uneven sample sizes reflecting differences in regional archaeological practices. Furthermore, given that spatial resolution is intertwined with the sample representativeness, the choice of appropriate scale of analysis becomes a non-trivial issue. Existing solutions have overcome some of these problems (e.g. by using correction surfaces) but no methods have been devised to assess the statistical significance of the observed differences in the inferred population trajectories. Current cross-regional comparisons of SPDs are thus limited to qualitative accounts, with a considerable risk of failing to distinguish genuine differences from spurious ones arising from sampling error. Here we propose a new method that overcomes many of these problems. We test the robustness and applicability of our solution through two case studies (a simulated dataset and the EUROEVOL database) showing the limits and the potentials of our approach.

This Resource is Part of the Following Collections

• Society for American Archaeology 82nd Annual Meeting, Vancouver, BC (2017)   •
• Frison Institute Symposium: The Future of "Big Data" in Archaeology

Keywords

Spatial Coverage

Record Identifiers

Abstract Id(s): 16713