Description: Highlights • First systematic dissolved neodymium isotope distributions in Angola and Cape Basins. • Deep water neodymium isotopes dominated by conservative mixing in study area. • Neodymium isotopes trace surface waters from Mozambique Channel in Angola Basin. • Cape Basin bottom water neodymium isotopes are unaffected by neodymium from sediments. • Glacial Cape Basin bottom waters may show effects of sedimentary neodymium inputs. Abstract In contrast to the vigorous deep ocean circulation system of the north- and southwestern Atlantic Ocean, no systematically sampled datasets of dissolved radiogenic neodymium (Nd) isotope signatures exist to trace water mass mixing and provenance for the more restricted and less well ventilated Angola Basin and the Cape Basin in the southeastern Atlantic Ocean, where important parts of the return flow of the Atlantic Meridional Overturning Circulation are generated. Here, to improve our understanding of water mass mixing and provenance, we present the first full water column Nd isotope (expressed as εNd values) and concentration data for a section across the western Angola Basin from 3° to 30° S along the Zero Meridian and along an E-W section across the northern Cape Basin at 30° S sampled during GEOTRACES cruise GA08. Compared with the southwestern Atlantic basin we find overall less radiogenic εNd signatures reaching −17.6 in the uppermost 200 m of the Angola and Cape basins. In the western Angola Basin these signatures are the consequence of the admixture of a coastal plume originating near 13° S and carrying an unradiogenic Nd signal that likely resulted from the dissolution of Fe-Mn coatings of particles formed in river estuaries or near the West African coast. The highly unradiogenic Nd isotope signatures in the upper water column of the northern Cape Basin, in contrast, originate from old Archean terrains of southern Africa and are introduced into the Mozambique Channel via rivers like the Limpopo and Zambezi. These signatures allow tracing the advection of shallow waters via the Agulhas and Benguela currents into the southeastern Atlantic Ocean. The Nd isotope compositions of the deep water masses in both basins primarily reflect conservative water mass mixing with the only exception being the central Angola Basin, where the signatures are significantly overprinted by terrestrial inputs. Bottom waters of the Cape Basin show excess Nd concentrations of up to 6 pmol/kg (20%), originating from resuspended bottom sediments and/or dissolution of dust, but without significantly changing the isotopic composition of the waters due to similar εNd values of particles and bottom waters ranging between −9.6 and −10.5. Given that bottom waters within the Cape Basin today are enriched in Nd, non-conservative Nd isotopic effects may have been resolvable under past glacial boundary conditions when bottom waters were more radiogenic.

Description: Key Points Calibration of XRF core scanning data highlights the need for careful examination of sediment properties such as porosity/water Grain size and water content in the sediment trigger systematic artifacts in the signal intensity of light elements (e.g. Si and Al) Known terrigenous flux proxies (e.g Ti/Ca, Fe/Ca) are influenced by sea level variations X‐ray fluorescence (XRF) core scanning of marine and lake sediments has been extensively used to study changes in past environmental and climatic processes over a range of timescales. The interpretation of XRF‐derived element ratios in paleoclimatic and paleoceanographic studies primarily considers differences in the relative abundances of particular elements. Here we present new XRF core scanning data from two long sediment cores in the Andaman Sea in the northern Indian Ocean and show that sea level related processes influence terrigenous inputs based proxies such as Ti/Ca, Fe/Ca, and elemental concentrations of the transition metals (e.g. Mn). Zr/Rb ratios are mainly a function of changes in median grain size of lithogenic particles and often covary with changes in Ca concentrations that reflect changes in biogenic calcium carbonate production. This suggests that a common process (i.e. sea level) influences both records. The interpretation of lighter element data (e.g. Si and Al) based on low XRF counts is complicated as variations in mean grain size and water content result in systematic artifacts and signal intensities not related to the Al or Si content of the sediments. This highlights the need for calibration of XRF core scanning data based on discrete sample analyses and careful examination of sediment properties such as porosity/water content for reliably disentangling environmental signals from other physical properties. In the case of the Andaman Sea, reliable extraction of a monsoon signal will require accounting for the sea level influence on the XRF data.

Description: Increased carbon sequestration in the ocean subsurface is commonly assumed to have been one of the main causes responsible for lower glacial atmospheric CO2 concentrations. Remineralized carbon must have been stored away from the atmosphere for thousands of years, yet the water mass structure accommodating such increased carbon storage continues to be debated. Here, we present new sediment-derived bottom-water neodymium isotope records that allow fingerprinting of water masses and provide a more complete picture of the Atlantic Meridional Overturning Circulation geometry during the Last Glacial Maximum. These results suggest that the vertical and meridional structure of the Atlantic water mass distribution only experienced minor changes since the last ice age. In particular, we find no compelling evidence supporting glacial southern-sourced water substantially expanding to shallower depths and farther into the Northern Hemisphere than today, which had been previously inferred from stable carbon isotope (δ13C) reconstructions. We argue that depleted δ13C values observed in the deep Northwest Atlantic do not necessarily indicate the presence of southern-sourced water. Instead, these values may represent a northern-sourced water mass with lower than modern preformed δ13C values that were further modified downstream by increased sequestration of remineralized carbon, facilitated by a more sluggish glacial deep circulation, corroborating previous evidence.

Description: The Asian summer monsoon affects the lives of billions of people. With the aim of identifying geochemical tracers for the monsoon-related freshwater input from the major rivers draining into the Bay of Bengal (BoB) and the Andaman Sea (AnS), we have analyzed the yttrium and rare earth element (YREE) concentration of surface seawater samples from various locations spanning the Andaman Islands in 2011 to 2013. In some locations, samples have been taken in March, July, and November 2011, thus spanning the seasonal cycle and including different monsoon phases. Generally, the YREE patterns are similar to those reported for offshore samples from the BoB and AnS in January 1997, with seawater-normalized patterns of most samples characterized by middle REE enrichments. An enhancement of these middle REE bulges accompanies large increases in dissolved REE concentrations from streams and sediment-rich areas such as mangrove environments. Conversely, some samples, in particular those taken 1–2 days after heavy rainfall in March 2011, show pronounced REE scavenging accompanied by the preferential removal of dissolved light REEs (LREEs) and by higher Y/Ho ratios. The Nd isotope signature of the remaining dissolved REE phase of these low YREE samples is more radiogenic than local rocks and sediments. The time series at a location away from local input sources show remarkably similar REE patterns and concentrations in March and July. Then in October–November, following the peak in monsoon-induced river discharge, the dissolved REE concentrations increase by almost a factor of two, whereas Nd isotopes become less radiogenic by 1.5 εNd units. These unradiogenic values are found at the same site in the winter dry season of the following year, demonstrating the decoupling of sea surface salinity (SSS) and Nd. The large sub-annual variability of YREE concentrations and Nd isotopes encountered was likely caused by the conversion of YREE from the dissolved (probably colloidal) pool to the labile particulate fraction. The comparison of unfiltered and filtered sample concentrations reveals the existence of a large labile particulate pool in the BoB and AnS that most likely originates from the massive river sediment fluxes and is instrumental in the seasonal changes observed.

Description: The sedimentary concentration and stable isotope composition of molybdenum (Mo) is widely used as a proxy for paleo redox conditions in the marine environment. However, the behavior of Mo during early diagenesis is still not fully understood, which complicates the application of the Mo proxy in ancient continental margin environments. Here, we present Mo concentrations and isotope compositions of sediment and pore water samples from the Guaymas Basin in the Gulf of California. Our sample set covers a broad range of depositional environments, including sediments from within the eastern equatorial Pacific oxygen minimum zone (OMZ), from a semi-restricted oxic graben, and from near a hydrothermal vent-field. By investigating Mo cycling in these different settings, we provide new insights into different modes of Mo fixation and the associated isotope fractionation. Sediments from the OMZ have authigenic Mo concentrations (Moauth) between 3.3 and 17.2 µg/g and δ98Mo between +1.64 and +2.13 ‰. A linear decrease in pore water Mo concentrations to the depth were hydrogen sulfide accumulates along with sedimentary authigenic δ98Mo values (δ98Moauth) close to seawater indicate diffusion of Mo from the bottom water into the sediment with little isotope fractionation during quantitative Mo removal. Sediments from the site with oxic bottom water within the basin reveal Moauth concentrations ranging from 1.2 to 14.7 µg/g and δ98Moauth signatures between –1.39 to +2.07 ‰. Pore water Mo concentrations are generally higher than ambient bottom water concentrations and the light δ98Moauth signatures of the pore waters between +0.50 and +0.80 ‰ and of the sediments indicate continuous Mo exchange between the pore water Mo pool and Mn and Fe oxides during early diagenesis. Sediment samples from the vent field mainly consist of black smoker debris and are characterized by Moauth concentrations ranging from 8.6 to 33.2 µg/g and δ98Moauth values as high as +2.20 ‰. The relatively high Mo concentrations and seawater-like δ98Mo can be explained by near-quantitative Mo scavenging from hydrothermal solutions with little isotope fractionation at high temperatures. Comparison of our new data for the OMZ sediments in the Gulf of California with previously published data for sediments from the Peruvian OMZ highlights that Mo isotope compositions in this kind of setting strongly depend on how Mo is delivered to the sediment. If Mo delivery into the sediment contributes to Mo accumulation in the solid phase, as is the case in the Guaymas Basin, sedimentary Moauth concentrations are relatively low but the isotope values are close to the δ98Mo signal of seawater. If Mo is exclusively delivered by particles, like on the Peruvian margin, much higher sedimentary Moauth concentrations can be attained. In the latter case, Moauth isotope values will be lighter because the sediments preserve the isotopic offset that was generated during adsorption or uptake of Mo by particles. Our findings de-emphasize the role of dissolved Mo speciation in pore waters but highlight the importance of the mode of Mo delivery for the Mo concentration and isotope composition preserved in the paleo-record.

Description: Highlights • Exchange between pore and bottom water with sediment affects authigenic REE in the Labrador Sea (LS). • REE-based correction allows reconstruction of past bottom water ϵNd signatures. • Evidence for lack of southern sourced water in the LS through the last 35 ka. • Glacial LS bottom waters were less radiogenic than today with ϵNd = −16 ± 1. • Similarity to Nordic Seas ϵNd record suggests uninterrupted supply of overflow waters. Abstract Deep waters of the Labrador Sea (LS) are important contributors to the Atlantic Meridional Overturning Circulation, but their water mass structure has been highly variable and sensitive to climatic changes on different time scales. The LS is also an area of intense exchange of rare earth elements (REE) between seawater and the underlying sediments, which complicates the reconstruction of past deep water provenance based on radiogenic neodymium (Nd) isotopes. Most notably, Northwest Atlantic Bottom Water exchanges Nd with Archaean age Laurentian detritus, resulting in a significant shift to less radiogenic Nd isotope signatures before it enters the North Atlantic to form the deep part of North Atlantic Deep Water. Here we show that the authigenic fractions of LS core top sediments carry Nd isotope signatures intermediate between bottom water and detritus and thus reflect pore waters that incorporate a mixture of both signatures. We furthermore find that detrital imprints on pore waters led to shifts of REE patterns in the authigenic fraction towards detrital signatures in the past during times of enhanced supply of glacially eroded material from Hudson Bay to the LS, as recorded by radiogenic lead isotopes. This allows an estimation of the intensity of past benthic REE exchange inside the LS. We exploit variations in the mid REE enrichment in the authigenic phase to propose a correction to one LS Nd isotope record for detrital imprints originating from pore water exchange. The corrected ϵNd signatures are argued to more accurately reflect those of past bottom waters. This correction results in past LS bottom water signatures of −16 ± 1 during MIS 2 and 3, considerably less radiogenic than today. This implies that no southern sourced waters advanced into the LS during the last 35 ka and instead supports continuous bottom water sourcing from the Nordic Seas. It thus seems likely that LS bottom waters supplied unradiogenic Nd to abyssal Glacial North Atlantic Bottom Water in the Northwest Atlantic, as was previously hypothesized.

Description: Highlights • Improved understanding of the behaviour of instrumental mass fractionation (IMF). • The effect of matrix elements on IMF is largely associated with plasma conditions that can be quantified with the NAI. • Matrix effects can be systematically and significantly attenuated by tuning of instrumental operating parameters. • A matrix tolerance plasma state is defined for stable barium isotope analysis. • The suggested analytical protocol is expected to be applicable to other stable isotope measurements with MC-ICP-MS. Abstract Stable barium isotope measurements with multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) remain an analytical challenge and can be considerably affected by the presence of matrix elements, even when applying double spiking. Therefore significant efforts were invested in previous studies to develop efficient barium purification methods. However, due to the high variability in matrix/barium ratios for diverse sample matrices, potential matrix effects can still not be excluded. While a lot of effort has been invested into improving the chemical separation protocols, the impact of plasma conditions on the accuracy and precision of stable isotope measurements has rarely been considered. Here we present a systematic investigation of the relationship between plasma conditions, instrumental mass fractionation (IMF) and impurity (i.e. matrix) concentrations. The Normalised Ar Index (NAI) and Matrix-Ar Index (MA) are used to quantify MC-ICP-MS plasma conditions and plasma mass loading, respectively. Our results show that the effect of matrix elements on IMF is largely linked to plasma conditions (i.e. NAI) and behaves as a linear function of mass loading (i.e. MA). Accordingly, the matrix effects can be significantly attenuated by increasing the NAI thereby minimising the risk of plasma “over-loading”. The improved understanding of the behaviour of the matrix-induced IMF allows us to define a matrix tolerance plasma state for barium isotope analysis. The accuracy of this recommended method is further assessed by analyses of two well-studied reference materials, the GEOTRACES seawater reference sample SAFe D2 and the carbonate reference material JCp-1. We expect that the analytical protocol described in this study is applicable not only to barium isotope analysis, but also to a wide range of other stable isotope measurements with MC-ICP-MS.

Description: The marine chromium (Cr) cycle is still insufficiently understood, in particular the mechanisms modulating the spatial distribution of dissolved stable Cr isotopes in seawater. Redox transformations between its main oxidation states, Cr(VI) and Cr(III), have been held accountable for the observed tight inverse logarithmic relationship between the dissolved Cr concentration [Cr] and its isotopic composition (δ53Cr), whereby isotopically light Cr(III) is removed in surface waters and oxygen minimum zones (OMZs), and subsequently released to deeper waters from remineralized particles or sediments. Seawater [Cr] and δ53Cr were investigated in a series of depth profiles across the Peruvian margin OMZ, covering a wide spectrum of dissolved oxygen concentrations ranging from 2 to 242 µmol/kg. We found [Cr] ranging from 1.5 to 5.5 nmol/kg, associated with δ53Cr variations between +1.59 and +0.72 ‰, but no systematic relationship to dissolved oxygen concentrations. However, distinctly different seawater profiles were observed above the suboxic/anoxic shelf compared to those located further offshore, with substantial Cr removal restricted to suboxic or anoxic environments on the shelf. This suggests that suboxic conditions ([O2] 〈 5 µmol/kg) alone are not sufficient to account for substantial Cr removal. Given that environmental conditions under which Cr can be reduced remain restricted spatially, the role of this sink in the marine Cr cycle may therefore be small. Additionally, some observations corroborate the assumption that Cr reduction is not necessarily accompanied by immediate adsorption of the formed Cr(III) onto particles, leading to its removal from the dissolved phase. Instead, partial removal of Cr(III) via particles, leaving a residual dissolved Cr(III) pool, may be more widespread in suboxic waters.