A Predictive Model for Lithic Resources in Iowa

Bibliographic Data

ID9218309
AuthorsChad A Goings (Iowa Lakes Community College, 300 South 18th Street, Estherville, IA 51334, corresponding author)
Year2003
Volume48
Issue184
Pages53-67
Publication date2003-02-01
Open AccessNo
TypeARTICLE
VenuePlains Anthropologist (JOURNAL)
Journal identifiersISSN: 0032-0447 • E-ISSN: 2052-546X
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1080/2052546.2003.11949281
OpenAlexW2462989587
LanguageEN
Citations received2
References cited8

Geographic Information Systems (GIS) provide a powerful tool for modeling the natural distribution of Iowa lithic resources now found in archaeological sites. I use two methods to create a predictive GIS model of lithic exposures available for prehistoric use in Iowa: (1) comparing interpolated bedrock surfaces with ground elevations to create outcrop zones, and (2) using four predictors to statistically model the likelihood of rock outcrops within zones. The first method predicted twenty-four of twentyseven known outcrops (88.8% success rate) in the study area. The second method employs DempsterShafer Weight-of-Evidence modeling and logistic regression methods. Kolomogrov-Smirnov tests show the four predictors to be significantly different from background values and model assignments show high probabilities at known outcrops while eliminating vast areas as low probabilities. It is agreed that such models could be used to further study lithic procurement and use. A comparison of predicted outcrops to utilized lithic resources at a multi-component Woodland site demonstrates the use of high quality, non-local material over poor quality local material at this site

Archaeology · Bedrock · Geography · Geomorphology · Logistic regression · Outcrop · Prehistory · Statistics · Woodland · Ecology · Geochemistry and Geologic Mapping · Geology · Mathematics · Pleistocene-Era Hominins and Archaeology · Wildlife Ecology and Conservation

Tools

Open DOISci-Hub