GSA Connects 2021 in Portland, Oregon

Paper No. 139-6
Presentation Time: 9:35 AM


DUMITRU, Oana-Alexandra1, DYER, Blake C.2, AUSTERMANN, Jacqueline1, SANDSTROM, Michael1, CASHMAN, Miranda1, GOLDSTEIN, Steven L.1, D'ANDREA, William1, CREEL, Roger1 and RAYMO, Maureen E.1, (1)Lamont-Doherty Earth Observatory, Columbia University, 61 Route 9W, Palisades, NY 10964, (2)University of Victoria, School of Earth and Ocean Sciences, Victoria, BC V8P 5C2, Canada

Understanding sea level changes during the Last Interglacial (LIG; ~129-116 ka) is important for accurately predicting the future of sea level rise due to ongoing climate change, as past sea level histories are used to calibrate climate-ice sheet models. Coral reefs play a pivotal role in providing reliable estimates for the LIG sea-level, since they are often well preserved in the fossil record and have the potential for obtaining precise ages using U-Th dating. Nevertheless, many fossil corals show evidence of post-depositional alteration of their primary geochemistry and their utility is limited by their preservation. Here we present a new series of high-precision U-Th ages of in-situ corals of LIG origin from Crooked Island, Long Cay, Long Island, and Eleuthera of the Bahamas archipelago. To identify unaltered corals, we apply rigorous threshold criteria that screen samples by their appearance, mineralogy, [U], and amount of detrital component. Nearly all analyzed samples show a narrow δ234Uinitial range between 143.8-151.3 ‰. Reliable ages range from ~126-118 ka; typical age uncertainties are ~1 ka.

We translate these local sea level observations into LIG global mean sea level by correcting our data for Earth deformation processes, i.e., glacial isostatic adjustment (GIA) and longterm subsidence, that occurred since the corals’ deposition. For our GIA correction, we use sea level predicted from a set of ice histories and Earth viscosity structures. We infer, via a Bayesian inversion, the most likely range by identifying those parameters that produce sea level most consistent with LIG data from across the Bahamian archipelago including our new chronology. In turn, the U-Th ages together with the estimated paleo-water depth correspondent to each reef, provide important constraints on GIA model parameters. This approach allows to revise estimates of LIG global mean sea level and to assess the amount of ice melt during this past pronounced warming.