Identification of Geologic Variables in Development of An Archeological Predictive Model for Management of Military Lands in Desert Terrains

Military training and testing lands in the United States covering nearly 70,000 km2 contain thousands of archaeological sites reflecting >10,000 years of human land use. Long-term restricted public access has resulted in limited site exploitation, increased site integrity preservation, and enhanced significance. Planning for military maneuvers, expansion and development of new training areas and facilities, and other ground-disturbing activities requires identifying and evaluating cultural resources. Traditional, labor intensive archeological pedestrian survey methods are challenged with meeting expedited schedules required by the US Department of Defense. One solution to accelerating the process is linking key geologic variables (soil, landforms, landscape history) with known archaeological site characteristics to predict site location. We developed a GIS-based archeological predictive model using relations among key geologic variables at 140+ previously identified archaeological sites across diverse desert terrain at the US Army National Training Center, Fort Irwin, California. A favorability matrix was generated using statistical relations among the observed environmental variables to distinguish favorable and unfavorable places to find additional archeological sites. Maps depicting favorable surface and subsurface site locations were generated using the favorability matrix. Results indicate five geologic variables are required for accurate predictions: deposit age, surface age, lithology, landform morphology, and soil type. Principal observations include: surface cultural features are primarily associated with very stable soil features associated with desert pavements; habitation sites are commonly located near coarse-grained, highly-fractured plutonic rocks; intact, buried sites are unlikely in areas dominated by ephemeral fluvial processes. Because of the general low resolution of GIS-based geologic data and limited/incomplete soil databases for most military installations in the US desert southwest, we are developing rapid and cost-effective methods for mapping and delineating key geologic features and soil data for desert terrain characterization that can be used to enhance the model for wider application across the desert southwest.