Statistical estimation of population density maps for North America from radiocarbon data
A still frequently discussed topic in archaeology is the settlement history of North America. At the year 15,000 BP (before present), first tribes emigrated from Asia to North America, which marked the beginning of the settlement. In the centuries after Christopher Columbus discovered America, however, wars and diseases extremely decimated the native population. Therefore, no exact information about total number, temporal development, and spatial distribution of the native population between 15,000 BP and 1,500 BC is available (estimates rank from 2 to 18 million at 1,500 BC). The only approach to obtain population data is the elicitation and statistical evaluation of archaeological data.
In cooperation with the Departments of Mathematics and Statistics and Geography of the University of Ottawa, we investigate the temporal and spatial development of the population of the North American continent starting from 15,000 BP. We analyze data from the Canadian Archaeological Radiocarbon Database (CARD), which contains the locations and calibrated ages of more than 19,000 archaeological finds all of them indicating human activity. Using methods from stochastic geometry and spatial statistics (e.g. kernel density estimation) we try to reconstruct the settlement process in order to find out how the native population structure developed in space and time in pre-Columbian America.
The project is still in an early stage. Possible topics for theses include:
- Development, implementation, and evaluation of statistical methods for the estimation of population densities from archaeological radiocarbon data
- Development and implementation of methods for a temporal and spatial smoothing of radiocarbon data
- Implementation and evaluation of statistical tests for the analysis of the spatio-temporal development of population densities
This project is supported by the German Academic Exchange Service.
Back to projects.
Back to theses.