Impact of sedimentary degradation and deep water column production on GDGT abundance and distribution in surface sediments in the Arabian Sea: Implications for the TEX86 paleothermometer

The TEX86 is a widely used paleotemperature proxy based on isoprenoid glycerol dibiphytanyl glycerol tetraethers (GDGTs) produced by Thaumarchaeota. Archaeal membranes are composed of GDGTs with polar head groups (IPL-GDGTs), most of which are expected to be degraded completely or transformed into more recalcitrant core lipid (CL)-GDGTs upon cell lysis. Here, we examined the differences in concentration and distribution of core lipid (CL)-and intact polar lipid (IPL)-GDGTs in surface sediments at different deposition depths, and different oxygen bottom water concentrations (<3-83 mu mol L-1). Surface sediments were sampled from 900 to 3000 m depth on a seamount (Murray Ridge), whose summit protrudes into the oxygen minimum zone of the Arabian Sea. Concentrations of organic carbon, IPL-and CL-GDGTs decreased linearly with increasing maximum residence time in the oxic zone of the sediment (t(OZ)), suggesting increasing sedimentary degradation of organic matter and GDGTs. IPL-GDGT-0 was the only exception and increased with t(OZ), indicating that this GDGT was probably produced in situ in the surface sediment. Concentrations of crenarchaeol with glycosidic headgroups decreased with increasing t(OZ), while crenarchaeol with a hexose, phosphohexose head (HPH) group, in contrast, showed an increase with increasing t(OZ), indicating that the concentration of HPH crenarchaeol was primarily determined by in situ production in surficial sediments. TEX86 values of both IPL-derived GDGTs and CL-GDGTs decreased by similar to 0.08 units with increasing water depth, in spite of the sea surface temperatures being identical for the restricted area studied. In situ production in sediments could be excluded as the main cause, due to the slow production rates of GDGTs in sediments, and previous observations of the same trends in TEX86 in sediment trap material. Instead, the incorporation of GDGTs produced in the oxygen minimum zone (with high TEX86 values) and their preferential degradation during the sinking through the water column, or differential degradation of IPL-GDGTs per head group could be the causes for the observed change in TEX86 values. The effect of differential degradation might cause differences between oxic and anoxically deposited sediments, and, together with a potential deep water contribution on TEX86 values, could translate into changes in reconstructed temperature of <3 degrees C, which might have to be accounted for in TEX86 calibration and paleotemperature studies of deep water sedimentary records. (C) 2014 Elsevier Ltd. All rights reserved.