Modeling Early Human Migration Patterns in South America: A Preliminary Spatial Analysis on the Peruvian Coastline Using Machine Learning and Bayesian Statistics

Summary

This is an abstract from the "SAA 2021: General Sessions" session, at the 86th annual meeting of the Society for American Archaeology.

The first South Americans' coastal migration routes remain a central question to studying the settlement patterns of human colonizations worldwide. However, these early migrations likely occurred along a coastline that today is mostly submerged. Consequently, in countries like Peru, there is currently a shortage of coastal archaeological sites that date to this time. This study presents a preliminary spatial analysis focused on evidence for the early human migration patterns along the Peruvian coastline. The main objective of this research is to increase knowledge on the probability of finding early archaeological remains in this area of the Andes, especially in the Cañete region, located on the central coast of Peru. To achieve this objective, we conducted three primary activities: 1) archival research of Peruvian archaeological literature detailing previous surveys and excavations of early archaeological sites on the coast; 2) generated a database of locations and descriptions of these sites, and 3) designed models to predict new site locations using machine learning and Bayesian statistical techniques in the R programming environment. Results of cross-validation tests show that models make successful predictions of known sites using independent datasets. Pedestrian surveys will verify new potential site locations once field activity can resume safely.

This Resource is Part of the Following Collections

• Society for American Archaeology 86th Annual Meeting, Online (2021)   •
• SAA 2021: General Sessions

Keywords

Spatial Coverage

Record Identifiers

Abstract Id(s): 33442