Seasonal variations in the long chain diols fluxes in the East Sea: Implication for the paleotemperature proxy

Abstract

Long chain diols (LCDs) were reported for the first time in the Black Sea sediments (de Leeuw et al., 1981). Generally, C28-C32 diols which have alcohol groups at C1 and a mid-chain position (e.g. carbon number 13, 14, 15) are dominantly observed in the various marine environments. Rampen et al. (2012) showed that the fractional abundances of C30 1,15-diols were positively correlated with sea surface temperatures (SSTs) in the globally distributed sea surface sediments. In contrast, the fractional abundances of C28 1,13-diols and C30 1,13-diols revealed the negative correlations with SSTs. Based on this observation, the long chain diol index (LDI) was proposed as a paleotemperature proxy. However, there applicability of this proxy has not been tested in the East Sea yet. Hence, we deployed a sediment trap in the Ease Sea (37.33？N, 131.45？E, 2300 m water depth) for the time period of March 2011-Feburary 2012 and analyzed alkenones and LCDs. The fluxes of alkenones and LCDs showed the seasonal variations with the higher fluxes of alkenones at the period of spring (April) and early summer (June) while the fluxes of LCDs were higher at the period of summer (July) to early autumn (September). The estimated SSTs based on the UK'37 and the LDI also showed the seasonal variations between 6.8 °C and 22.7 °C for the UK'37 and between 2.4 °C and 24.7 °C for the LDI with a time lag of satellite-derived SSTs. The flux-weighted SST based on the UK'37 was 11.4 °C while that of the LDI was 17.3 °C. Accordingly, the UK'37-based annual mean SST was 4.9°C lower than the in-situ annual mean SST (16.3 °C), while the LDI-based SST was 1°C lower. In summary, our results indicate that the LDI-based SSTs might be warm-biased than the UK'37-based ones in the East Sea.