Organic biomarker sea surface temperature proxies in Mediterranean Sea sediment traps

Abstract

Paleoclimate proxies allow researchers to assess past climate and how it changed over time. Two sea surface temperature (SST) proxies, TEX86 and Uk37, are based on biomarkers called alkenones and glycerol dialkyl glycerol tetraethers (GDGTs), respectively. Specifically, Uk37 is the degree of unsaturation of long-chain alkenones, and TEX86 is based on the relative number of ring moieties in GDGTs. In the Mediterranean Sea, TEX86 tends to overestimate SST by 2-6 ⁰C on average, while Uk37 underestimates SST by 2-4 ⁰C when using calibrations based on globally distributed core tops. Examining these biomarkers in sinking particles collected by sediment traps can provide insight into why these offsets occur. Here, we study the potential impact of seasonality and possible biases in export on the TEX86 and Uk37 by analyzing alkenones and GDGTs in biweekly sediment trap material collected at 510 m, 1745 m, and 2920 m depth in the Urania Basin in the Eastern Mediterranean Sea from August 1999 until July 2000 and from August 2005 until July 2006. The results show a general agreement of the individual sediment traps using either paleotemperature proxy. Reconstructed temperatures of TEX86 show little variation on an annual scale. A large influence of Deep Dwelling Thaumarchaeota is found to be a possible cause for a lack of seasonality and an annual overestimation of SST. The annual flux of GDGTs does show seasonality and peaks from late winter through early spring causing annual temperature predictions by TEX86 to be weighted toward this time of the year. The long-chain alkenone flux peaks around the same time as the GDGT flux. The Uk37 temperature prediction shows seasonality and correlates relatively well with SST during autumn and winter. During spring and summer, the UK37 signal better represents sub-mixed layer temperatures at 40 m depth. This is likely caused by the increasingly stratified water column in the Mediterranean at this time of the year forcing Haptophyte algae to migrate to deeper waters. Even though export biases are found for each proxy concerning the timing and depth causing offsets in the temperature predictions, both paleotemperature proxies are useful tools to reconstruct past environments. However, they are to be used with caution and awareness of the uncertainties that come with using them.