PP21C-1340:
Early Diagenetic Imprint on Temperature Proxies in Holocene Corals: A Case Study from French Polynesia

Tuesday, 16 December 2014
Rashid Juma Rashid1, Anton Eisenhauer1, Volker Liebetrau1, Jan Fietzke1, Florian Boehm1, Marlene Wall1, Stefan Krause1, Andres Rüggeberg2, Wolf-Christian Dullo1 and Elias Samankassou3, (1)GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany, (2)University of Fribourg, Earth Sciences, Fribourg, Switzerland, (3)University of Geneva, Geneva, Switzerland
Abstract:
Coral based reconstructions of sea surface temperatures (SST) using Sr/Ca, U/Ca and δ18O ratios are an important tool for quantitative reconstructions of past climate. However, post-depositional alteration of coral aragonite due to early diagenesis is restricting the accuracy of calibrated proxies even on relatively young corals. Here we present Mid to Late Holocene SST reconstructions using well dated (U/Th: ~70yr to 5.4ka) fossil Porites collected from the Society Islands, French Polynesia. Microscopic observations and microprobe mapping show the presence of aragonite needles inside the coral pores. They are characterized by high Sr/Ca ratios of up to 14.7±0.3mmol/mol, shifting the corresponding SST to ~6.2±1.1°C. Inhomogeneous distribution of Sr/Ca reflects partial dissolution, pore fluid percolation and re-precipitation of Sr within the coral. The resulting effect of early diagenesis on the SST records is reflected by significant temperature offsets up to ±6°C. A relative SST reconstruction is still feasible by normalizing the records to their individual mean value. Latter approach shows that all three records are in phase with simultaneous SST variability of up to ±2.4°C with respect to the Holocene mean value. Low SSTs are observed between ~4.4 to ~4.9ka BP and ~2.8 to ~3.4ka BP. Relatively high temperatures are observed at ~5ka BP, between ~3.7 and 3.9ka BP as well as at ~1.8ka BP to ~2.6ka BP. This indicates SST oscillations with the periodicity of 1500yrs in the Society island record which is in general accord with the solar activity reconstructed from 10Be and 14C production (Vonmoos et al., 2006, JGR 111, A10105). In particular, the relatively high SST reconcile empirically determined sea level variations (Rashid et al., 2014, G3, 15) indicating a higher sea level than predicted from numerical modelling taking only “Glacial Isostatic Adjustment” into account as the primary controlling factor for the Late Holocene sea level variation.