North-Central Section - 39th Annual Meeting (May 19–20, 2005)

Paper No. 11
Presentation Time: 11:40 AM

A CONTINUATION OF THE COLD WATER CAVE SPELEOTHEM RECORD BACK TO 11,000 YEARS BP


KNIGHT, Charles, Geoscience, Univ of Iowa, Iowa City, IA 52242, DORALE, J.A., Geoscience, Univ of Iowa, 121 Trowbridge Hall, Iowa City, IA 52242, BAKER, Richard G., Department of Geoscience, Univ of Iowa, Iowa City, IA 52242, GONZÁLEZ, Luis, Geology, Univ of Kansas, Lawrence, KS 66045 and EDWARDS, R. Lawrence, Geology and Geophysics, Univ of Minnesota, 310 Pillsbury Dr SE, Minneapolis, MN 55455, charles-knight@uiowa.edu

Previous work at Cold Water Cave, IA (Dorale et al., 1992: Science 258, 1626) and Roberts Creek, IA (Chumbley et al., 1990: Science 249, 272) and a subsequent comparison of the two sites (Baker et al., 1998: Geology 26, no. 12, 1131) documented the arrival of prairie into Northeast Iowa around 6,000 years b.p. (ka). At Robert's Creek the vegetation sequence as documented by pollen and macrofossils begins around 14 ka and consists of a spruce/sedge assemblage. Beginning around 12 ka, spruce and elm are the dominant taxa. This non-analogue spruce/elm assemblage is replaced by a wholly deciduous forest at ~ 10.7 ka, which persists with minor to moderate perturbations until the replacement by prairie at 6 ka.

At Cold Water Cave the replacement of forest elements by prairie was captured by the δ13C values of stalagmite calcite, which shifted from early Holocene values around -9 per mil to values around -5 per mil after 6 ka. However, the published Cold Water Cave work extends back to only ~ 8.5 ka, and thus fails to capture conditions of the Late Glacial and earliest Holocene. Here we report an extension of the Cold Water Cave speleothem record back to approximately 11 ka based on new, improved dating of one of the original three stalagmites collected in the early 1990's. These new high-precision U-Th dates are 7,960 ± 70 yrs b.p., 9,680 ± 90 yrs b.p., and 11,420 ± 500 yrs b.p. from the top, middle, and bottom, respectively of the ~ 20 cm stalagmite. The error on the bottom age is large because of moderately high detrital thorium content, and is sensitive to the initial 230Th correction.

The extended δ13C record displays a pattern consistent with the previously published work from Cold Water Cave and Robert's Creek. At ~ 7.9 ka, δ13C values are -9 per mil, which seamlessly links the young part of the new Cold Water data with the old part of the published Cold Water work. The lightest δ13C values of -10.2 per mil occur around 9 ka, and are consistent with a conspicuous peak in the tree/herb ratio at the exact same time at Roberts Creek. From 9 ka back to 11 ka, δ13C values increase from -10 per mil to -8 per mil, possibly reflecting the general tree to herb ratio characteristic of the forest composition during the transition from mixed conifer/hardwood forest to wholly deciduous forest.