Distribution of Crenarchaeota tetraether membrane lipids in surface sediments from the Red Sea

The Red Sea represents an extreme marine environment, with high salinity, high temperature and low level of nutrients, complicating the application of standard geochemical palaeotemperature proxies. In order to investigate the applicability of the TEX86 (TetraEther indeX of GDGTs with 86 carbons) proxy for sea surface temperature (SST) in the Red Sea, the distribution of glycerol dialkyl glycerol tetraether membrane lipids (GDGTs) in sediments from the Red Sea and the Gulf of Aden were examined. Against expectations, TEX86 values for the Red Sea do not show a simple linear relationship with SST and deviate from the global core top calibration. In the northern Red Sea, at temperatures between 25 and 28 degrees C, the values increase linearly with SST, whereas in the southern Red Sea, at temperatures above 28 degrees C, TEX86 decreases with increasing temperature. Factors like seasonality and depth of production, salinity and nutrient availability, as well as diagenetic overprint or influence of allochtonous terrestrial lipids, cannot explain this pattern. However. the observed TEX86 relationship with SST could be explained by the presence of a hypothetical endemic Crenarchaeota population in the Red Sea with a specific TEX86 vs. SST relationship. In the Southern Red Sea, a two-component mixing model implies an exponential decrease in the endemic population towards the Gulf of Aden. Thus, the application of the TEX86 as a palaeotemperature proxy in the Red Sea is likely only possible for the northern Red Sea area with the specific SST vs. TEX86 relationship determined in this study and potentially for the whole Red Sea basin during glacials, when water exchange with the Indian Ocean was more restricted than today and the endemic archaeal lipid distribution was not affected by transport from the Indian Ocean. Our results suggest that distinct populations of Crenarchaeota in extreme environments such as evaporitic basins may have different membrane composition, necessitating application of another calibration than the global core top calibration for TEX86 palaeothermometry. (C) 2009 Elsevier Ltd. All rights reserved.