Glycerol dialkyl glycerol tetraethers and TEX86 index in sinking particles in the western North Pacific

► Sinking flux of GDGTs measured in a sediment trap experiment. ► Flux was enhanced with phytoplankton bloom. ► TEX86 did not change in response to SST change. ► Sinking particle/surface sediment TEX86 values corresponded to mean annual SST. ► Only small portion of GDGTs preserved at sediment surface, but change in TEX86 small. Seasonal and depth variation in the flux of glycerol dialkyl glycerol tetraethers (GDGTs) and TEX86 (TEX86H and TEX86L) values in sinking particles was examined by conducting a 21month time-series sediment trap experiment at a mooring station (WCT-2, 39°N, 147°E) in the mid-latitude NW Pacific. The aim was to understand the sinking process of GDGTs in the water column and the preservation of the TEX86 signal in the water column and sediment surface. In the shallow trap, the sinking flux of GDGTs showed maxima from May 1998 to February 1999. The maximal peaks in sinking flux corresponded to peaks in the sinking flux of organic carbon, opal and lithogenic material. GDGT concentration in the total fine fraction and the caldarchaeol/crenarchaeol ratio at three depths (ca. 1300–4800m) varied synchronously, implying rapid vertical transport of GDGTs to deeper water with a sinking velocity >260mday−1 below ca. 1300m. The changes in TEX86-based temperature were different from those in contemporary sea surface temperature (SST). The former was lower than the SST from May to December and corresponded to the temperature at the thermocline, whereas it was higher than the SST from December to May. The annual average sinking flux of the GDGTs decreased with depth. The GDGT half-depth, the depth range over which half of the GDGT is lost, was calculated to be 3108–3349m, implying that GDGTs were well preserved during sinking. The flux-weighted average TEX86-based temperature was constant with depth and corresponded roughly to mean annual SST. The findings support a previous hypothesis that the GDGTs produced in surface water are preferentially delivered to the deeper water column via grazing and repackaging in larger particles. The constant TEX86 at different depths indicates that it was not affected by degradation in the water column. The preservation efficiency of GDGTs was 1.0–1.3% at the water–sediment interface. Despite significant GDGT degradation, there was a small difference in TEX86 values between sinking particles and surface sediment.