GDGT and alkenone flux in the northern Gulf of Mexico: Implications for the TEX 86 and U K' 37 paleothermometers

Abstract

The TEX86 and U-37(K') molecular biomarker proxies have been broadly applied in downcore marine sediments to reconstruct past sea surface temperature (SST). Although both TEX86 and U-37(K') have been interpreted as proxies for mean annual SST throughout the global ocean, regional studies of glycerol dibiphytanyl glycerol tetraethers (GDGTs) and alkenones in sinking particles are required to understand the influence of seasonality, depth distribution, and diagenesis on downcore variability. We measure GDGT and alkenone flux, as well as the TEX86 and U-37(K') indices in a 4 year sediment trap time series (2010-2014) in the northern Gulf of Mexico (nGoM), and compare these data with core-top sediments at the same location. GDGT and alkenone fluxes do not show a consistent seasonal cycle; however, the largest flux peaks for both occurs in winter. U-37(K') covaries with SST over the 4 year sampling interval, but the U-37(K') SST relationship in this data set implies a smaller slope or nonlinearity at high temperatures when compared with existing calibrations. Furthermore, the flux-weighted U-37(K') value from sinking particles is significantly lower than that of underlying core-top sediments, suggesting preferential diagenetic loss of the tri-unsaturated alkenone in sediments. TEX86 does not covary with SST, suggesting production in the subsurface upper water column. The flux-weighted mean TEX86 matches that of core-top sediments, confirming that TEX86 in the nGoM reflects local planktonic production rather than allochthonous or in situ sedimentary production. We explore potential sources of uncertainty in both proxies in the nGoM but demonstrate that they show nearly identical trends in twentieth century SST, despite these factors.