Over the line: a least-cost analysis of "community" in Mesa Verde National Park

Kelsey M. Reese

The spatial extent and social scope of archaeological communities in the Southwest has primarily been defined using arbitrary designations based on site observations, or by imposing a boundary derived from contemporary contexts to encompass groups of prehistoric sites. These approaches are either subjective, non-replicable, or presume that prehistoric peoples related to and interacted with one another in the same ways that people do today. This thesis aims to minimize such problems by using the spatial relationships among contemporaneous habitation sites in Mesa Verde National Park from A.D. 600 through abandonment in A.D. 1280 to find an emergent quantitative definition of community extent that can be used to understand changes in social organization through time in the Mesa Verde region. This analysis generates cost-distances between all pairs of contemporaneous habitation sites, produces a null model to determine the cost-distance distribution expected on a randomly settled landscape, and finds the difference between the two--referred to as the "Null Difference"--as a means to determine the point at which a household decided to move closer to or farther from other contemporaneous households. This process is repeated for each of 14 modeling periods, defined by the Village Ecodynamics Project, to determine the average vi landscape-wide cost-distance at which prehistoric households were settling closer to one another than expected by the null model. The Null Difference is then used to assign landscape cells to each habitation by calculating the cost-distance between every site and cell on the landscape. Affinity propagation cluster analysis then groups contemporaneous habitation sites based on the similarity of landscape cells that fall within the Null Difference extent for each modeling period. The cluster analysis also defines the optimal number of clusters for habitations in each period, producing areas of likely interaction where varying degrees of aggregation and social movement across the landscape can be observed. In general through time, the number of communities declines, the area of communities increases, the number of households in communities increases, and the average cost-distance among households within a communities increases--although there are interesting anomalies in each of these variables.

Over the line: a least-cost analysis of "community" in Mesa Verde National Park

Files and links (1)

Abstract

Metrics

Details