Assessing Use of Non-Cultural Charcoal for Age Control in Archaeological Studies

No Thumbnail Available
Authors
Roberts, Sarah
Advisor
Editor
Date of Issue
2018-04-20
Subject Keywords
Publisher
Citation
Series/Report No.
item.page.identifier
Title
Assessing Use of Non-Cultural Charcoal for Age Control in Archaeological Studies
Other Titles
Type
Description
Abstract Only
Abstract
Archeological studies at the Sundog site in the Big Belt mountains, suggest that lithic artifacts are non-randomly distributed and in stratochronologic order (older artifacts on bottom, younger on top) reaching bac to the McKean Complex ~8,000 BP. In contrast, distinct sediment layers at Sundog were absent or potentially disturbed. Non-cultural charcoal (sampled every 10 cm in one 150 cm pit) gave thirteen radiocarbon ages that did not come out entirely as hoped. The upper 100 cm gave a jumble of dates ranging from 3,800-400 years ago, many were reversed suggesting mixing. More promising, the lower 50 cm of the site had four dates ranging from 10,100-9,300 yrs BP. Despite some reversals, these dates suggest the lower strata are consistently and significantly older and have a higher sedimentation rate than the upper unit. The significant age gap at 100 cm suggests a depositional break or erosion between 9000 and 4000 BP. Grainsize analysis shows a general fining upward trend, with little stratigraphic variation in the upper 60 cm, and weak variation in the small pebble fraction of the bottom 80 centimeters. This study explores several potential hypotheses explaining the mixed radiocarbon ages and lack of stratigraphy. Detailed examination of charcoal fragment characteristics (size, preservation, clay coatings) along with sediments may help determine if materials have been mixed or disturbed, transported from nearby slopes, or perhaps created in situ through root burn after fire. This study may help inform future collection methods of non-cultural charcoal used as age control in archaeological sites.
Sponsors
Degree Awarded
Semester
Department