Developing a GIS-Based Weights of Evidence Predictive Model to Locate the Probability of Prehistoric Sites in Southwest Montana
To improve cultural resource management of the Bureau of Land Managements prehistoric sites, using GIS weights of evidence, this thesis has produced archaeological site predictive models for the Dillon Field Office in southwest Montana. Weights of evidence analysis of prehistoric archaeological sites and their association with evidential themes specifically distance to water, aspect, and slope form the basis for which the predictive models were generated. The study suggests that prehistoric archaeological sites in southwest Montana are likely to be located in areas near water, with minimal slope, facing in a general north direction.
Using a weights of evidence inductive model, it was shown that prehistoric sites in the Dillon Field office area were most likely located on a 0-10 % slope, 0-50 meters from water (specifically streams) and facing northeast. However, the contrast from analysis shows that distance to water (streams) and aspect have the most influence to predict for the occurrence of archaeological sites.
The archaeology of the Dillon Field Office area has been defined by a lack of information about inter-site spatial relationships as well as the relationships among sites and the environmental properties of the associated landscape. This GIS-based thesis provides predictive models for a better understanding of the spatial relationships between prehistoric sites and southwest Montanas versatile environment. The models in this thesis are intended to improve the cultural resource management of the Dillon Field Office by 1) reducing the monetary and field costs of fieldwork, 2) applying more effective land-use management, 3) devising more appropriate field survey strategies, 4) and assisting in updating the Dillon Field Offices cultural database.
Advisor:Dr. Paul B. Wilson; Dr. Kelly J. Dixon; Dr, Todd M. Ahlman
School:The University of Montana
School Location:USA - Montana
Source Type:Master's Thesis
Date of Publication:09/19/2007