GSA Annual Meeting, November 5-8, 2001

Paper No. 0
Presentation Time: 9:45 AM

THE AGE AND DURATION OF THE LAST INTERGLACIAL HIGHSTAND FROM U-TH CORAL AGES: ACCOUNTING FOR ISOTOPIC ADDITION


THOMPSON, William G.1, FAIRBANKS, Richard G.1, RUBENSTONE, James L.1, MORTLOCK, Richard A.1 and SPEED, Robert C.2, (1)Lamont-Doherty Earth Observatory, Columbia Univ, Palisades, NY 10964-8000, (2)Northwestern Univ, Locy Hall 1847 Sheridan Rd, Evanston, IL 60208-2150, billt@ldeo.columbia.edu

The precise timing of past sea-level changes is critical to understanding Quaternary climate change. Mass-spectrometric U-Th isotope measurements of corals promise high precision age determinations, if closed system requirements are met. Despite excellent analytical precision, the age and duration of the last interglacial remains problematic.

Many published U-Th ages conflict with the orbitally-tuned marine record of d18O. Marine isotope substage 5e extends from 128 to 115 kyr BP, with a highstand maximum at 122.2 kyr (SPECMAP timescale). However, the range of 218 published last interglacial U-Th coral ages from seven localities is 104-152 kyr. While modern corals faithfully record seawater 234U/238U, 90% of published initial 234U/238U exceed this value, suggesting open system behavior during prolonged sub-aerial exposure of these reefs.

New U-Th data from the well-mapped coral terraces of Barbados, West Indies confirm a previously reported positive correlation between 234U/238U and 230Th/238U for coeval corals that may be explained by post-depositional isotope addition. We suggest that 234Th (which rapidly decays to 234U) and 230Th , decay products from an external uranium source, are continuously adsorbed onto fossil corals. This constitutes a substantial systematic bias to standard U-Th ages from uplifted reefs, producing older apparent ages and elevated initial 234U/238U.

A continuous addition age equation is solved by assuming constant 234U/238U for both seawater and the external uranium source. Measured isotopic compositions of coeval Barbados corals are reasonably well explained by this equation. Continuous addition ages calculated for corals with elevated 234U/238U agree well with standard U-Th ages of coeval corals with no evidence of open system behavior. The mean of 218 addition-corrected ages calculated from published measurements is 121.6 ± 5.6 kyr BP, in close agreement with the SPECMAP peak at 122.2 ± 2.3 kyr, and 74% fall between the boundaries of 115 and 128 kyr.