Alkenone δ2H values – a viable seawater isotope proxy? New core-top δ2HC37:3 and δ2HC37:2 data suggest inter-alkenone and alkenone-water hydrogen isotope fractionation are independent of temperature and salinity

Abstract

A growing body of work has demonstrated that the δ2H values of alkenones reflect the δ2H values (δ2HH2O) and / or salinity of the fluid in which they are produced. If so, δ2Halkenone values would act as a surface seawater isotope / salinity proxy, similar to foraminiferal δ18O values, but advantaged in locations with poor carbonate preservation and / or high organic content. Nevertheless, laboratory culture, sediment trap, and water column studies have failed to consistently characterize the effects of temperature, alkenone-producing species, and salinity itself on the δ2Halkenone-salinity and -seawater isotope relationships, and a robust sedimentary alkenone-based calibration remains elusive.Most δ2Halkenone datasets report δ2HC37, i.e., combined δ2HC37:3 and δ2HC37:2 values, and differ in how they address inter-alkenone fractionation (i.e., αC37:3-C37:2). To constrain controls on alkenone hydrogen isotope systematics in the natural environment, we measured δ2H values of C37 and C38 alkenones from 20 open ocean core tops by gas chromatography-stable isotope ratio mass spectrometry after separation of di- and tri-unsaturated forms. Core-top δ2Halkenone data points are currently concentrated in extreme-salinity regions; in combination with our new values from a more moderate range of open ocean δ2HH2O / salinity values, for sedimentary alkenones, we show that 1) mean inter-alkenone hydrogen isotope fractionation is negligible (αC37:3-C37:2 = 1.002 ± 0.006), and therefore that δ2HC37:3 and δ2HC37:2 values can be measured in bulk; 2) temperature and salinity have little impact on alkenone-water fractionation (i.e., αC37-H2O) (mean 0.803 ± 0.010) relative to their expected variability in the ocean; and 3) δ2HC37 and δ2HH2O values are correlated such that statistically identical δ2HC37-, δ2HC37:3-, and δ2HC37:2-δ2HH2O regressions yield a core-top-based calibration of δ2HC37 = 1.44 (± 0.13) * δ2HH2O – 191.62 (± 1.13) ‰. This is indistinguishable from water column calibrations, suggesting a consistent response of environmental δ2HC37 values to changes in δ2HH2O values.This calibration still contains a high amount of scatter (∼ 7 ‰), perhaps attributable to irradiance, growth rate, intra- or interspecies variability, or other factors difficult to constrain in sedimentary material. Nevertheless, when applied to the well-constrained Last Glacial Maximum-to-present mean ocean δ2HH2O change of ∼ 8.8 ‰, it (1.44 ‰ δ2HC37 per 1 ‰ δ2HH2O change) reproduces the mean δ2HC37 Modern – δ2HC37 LGM shift observed from the handful of extant down-core records, legitimizing the observed lack of temperature or salinity effects on αC37-H2O. This suggests that combined δ2HC37:3 + δ2HC37:2 values are a valid proxy for δ2HH2O values in open ocean settings where E. huxleyi and G. oceanica dominate, although additional efforts will be required to refine the core-top calibration for universal use.