Steven Emerson

The geochemistry of redox sensitive trace metals in sediments

Abstract We analyzed the redox sensitive elements V, Mo, U, Re and Cd in surface sediments from the Northwest African margin, the U.S. Northwest margin and the Arabian Sea to determine their response under a range of redox conditions. Where oxygen penetrates 1 cm or less into the sediments, Mo and V diffuse to the overlying water as Mn is reduced and remobilized. Authigenic enrichments of U, Re and Cd are evident under these redox conditions. With the onset of sulfate reduction, all of the metals accumulate authigenically with Re being by far the most enriched. General trends in authigenic metal accumulation are described by calculating authigenic fluxes for the 3 main redox regimes: oxic, reducing where oxygen penetrates ≤1 cm, and anoxic conditions. Using a simple diagenesis model and global estimates of organic carbon rain rate and bottom water oxygen concentrations, we calculate the area of sediments below 1000 m water depth in which oxygen penetration is ≤1 cm to be 4% of the ocean floor. We conclude that sediments where oxygen penetrates ≤1 cm release Mn, V and Mo to seawater at rates of 140%–260%, 60%–150% and 5%–10% of their respective riverine fluxes, using the authigenic metal concentrations and accumulation rates from this work and other literature. These sediments are sinks for Re, Cd and U, with burial fluxes of 70%–140%, 30%–80% and 20%–40%, respectively, of their dissolved riverine inputs. We modeled the sensitivity of the response of seawater Re, Cd and V concentrations to changes in the area of reducing sediments where oxygen penetrates ≤1 cm. Our analysis suggests a negligible change in seawater Re concentration, whereas seawater concentrations of Cd and V could have decreased and increased, respectively, by 5%–10% over 20 kyr if the area of reducing sediments increased by a factor of 2 and by 10%–20% if the area increased by a factor of 3. The concentration variations for a factor of 2 increase in the area of reducing sediments are at about the level of uncertainty of Cd/Ca and V/Ca ratios observed in foraminifera shells over the last 40 kyr. This implies that the area of reducing sediments in the ocean deeper than 1000 m (4%) has not been greater than twice the present value in the recent past.

Ocean anoxia and the concentrations of molybdenum and vanadium in seawater

Abstract Molybdenum and vanadium are relatively unreactive in seawater and are preferentially concentrated in sediments overlain by anoxic or near-anoxic waters. We investigate the geochemistry of these metals by interpreting measurements from anoxic basins and sediment porewaters. The concentrations of Mo and V in the water column of anoxic basins are usually lower than in oxic seawater because of uptake into highly anoxic sediments. The mechanism of removal from solution appears to be diffusion across the sediment-water interface and immobilization within the sediments. Measurements from open ocean porewaters imply fluxes of vanadium from sediments that are greater than 10 times the dissolved and particulate river inflow, which is inconsistent with the distribution of vanadium in deep waters of the ocean and previous geochemical studies; this inconsistency shows that early diagenesis of this element is poorly understood. We estimate that the area of ocean sediment at present overlain by anoxic and near-anoxic water is accumulating 25 ± 15% and 8 ± 5% of the river influx of Mo and V, respectively. This fraction is 17 ± 10% for uranium, which has been shown to share some of the redox behavior of Mo and V. A case is made that changes in the concentrations of these elements in the ocean should be sensitive indicators of the portion of the ocean bottom overlain by anoxic seawater through geologic time.

Distribution of rose bengal stained deep-sea benthic foraminifera from the Nova Scotian continental margin and Gulf of Maine

Abstract Analysis of Rose Bengal stained benthic foraminifera in six boxcores taken from the Nova Scotian margin and Gulf of Maine reveals that foraminifera are vertically stratified within surficial sediments raised from 200 to 3000 m water depth. The consistent presence of infaunal taxa within the sediments demonstrates that these tolerant of a range of low-oxygen conditions as determined by pore-water manganese profiles. The habitat depth of foraminiferal populations, defined as the depth within which 95% of the fauna is found in a subcore, varies between the stations. A shallow habitat depth of 3 cm exists in shallow water (a 202 m core) in the Gulf of Maine. The habitat depth gradually increases on the continental slope with increasing water depth, reaching 11–13 cm in cores at 2225 and 3000 m, and then shoals to 4 cm in a core taken from 4800 m water depth. We suggest that the habitat pattern is related to the flux of organic carbon to the seafloor. At shallow depths, relatively high organic carbon flux results in a shallow oxic layer. The inferred oxic layer gradually increases on the continental slope and rise with increasing water depth, due to decreased organic carbon flux to the sediment-water interface. Carbon fluxes in the deep ocean are so low that pore waters are oxic and the organic carbon content is low, creating a food-limiting environment best suited to epifaunal taxa and reflected in a shallow habitat depth.

Experimental determination of the organic carbon flux from open-ocean surface waters

The flux of biologically produced organic carbon from the euphotic zone of the ocean to the deep waters below—the ‘biological organic carbon pump’—is one of the main controls on the carbon dioxide partial pressure in the atmosphere. Accurate determination of this flux is therefore critically important for understanding the global carbon cycle and its response to climate change. Our goal is to assess how accurately the biological organic carbon pump can be determined at a single location and to constrain estimates of the global value. As there are no standards against which such environmental fluxes can be measured, we assess accuracy by comparing results from three independent experimental approaches for measuring the net annual export of organic carbon from the euphotic zone in the subtropical North Pacific Ocean near Hawaii. Mass balances of dissolved oxygen, inorganic carbon and organic carbon yield estimates of the organic carbon export flux of 2.7 ± 1.7, 1.6 ± 0.9 and 2.0 ± 0.9 mol C m−2 yr−1, respectively. These three estimates are not significantly different, and establish the present analytically attainable accuracy at this location to be about ±50%. If 2.0 mol C m−2 yr−1 is typical of the organic carbon export flux in the subtropical ocean, then this vast region, often considered to be a biological desert, may be responsible for up to half of the global-ocean biological organic carbon pump.

Partitioning and transport of metals across the O2H2S interface in a permanently anoxic basin: Framvaren Fjord, Norway☆

Abstract The geochemical processes operating on metals in anoxic marine waters influence metal mobility and mode of transport to the sediments in a manner different from that observed in oxic regimes. In order to better understand these processes, dissolved and particulate Mn, Fe, Co, Ni, Cu, Zn, and Cd concentrations were determined in the water column of a permanently anoxic basin, Framvaren Fjord, Norway. Class specific behavior determines the degree to which these metals are involved in the processes of redox cycling at the O 2 H 2 S interface and metal sulfide precipitation in the sulfidic water. Metal sulfide precipitation influences the magnitude of metal enrichment in the sediments. The transition metals, Mn, Fe, and Co, show active involvement in redox cycling, characterized by dissolved maxima just below the O 2 H 2 S interface. Nickel concentrations appear unaffected by processes influencing the profiles of the other metals. The metals, Cu, Zn, and Cd, display a dramatic solubility decrease across the interface, are not involved in redox cycling, and are enriched in the sediments relative to a lithogenic component by factors of 11, 105, and 420, respectively. Ion activity products of the metals and sulfide provide evidence that chemical equilibria with a pure metal sulfide solid phase is not the dominant process controlling dissolved metal concentrations in the sulfide containing waters.

Organic carbon dynamics and preservation in deep-sea sediments

Abstract Three methods of estimating the particulate organic carbon fluxes to the sediment-water interface of the deep Pacific Ocean agree to within the error of the measurements at MANOP sites M, H, and C. Sediment trap experiments, pore water results, and surface sediment organic carbon data suggest that a major fraction of the particulate organic carbon raining to abyssal depths at these locations is degraded within the surface sediments rather than at the sediment-water interface or in the nephloid layer. Organic carbon rain rates at the three sites are similar—within a factor of two; however, the preservation rate of organic carbon and the chemistry of sediment pore waters are very different. A model developed to describe the pore water oxygen and sedimentary carbon distributions indicates model developed to describe the pore water oxygen and sedimentary carbon distributions indicates that changes in the rate constant for organic matter degradation and the bioturbation rate may contribute significantly to the observed differences in character of both pore water and sediment chemistry at these locations. The implication with respect to interpreting the sedimentary record is that cycles of organic carbon and redox sensitive metals (i.e., manganese) are not simply related to particulate organic carbon flux or surface water primary productivity. The residence time of organic carbon with respect to degradation in the surface sediments is on the order of 15 to 150 y.

Foraminiferal magnesium in Globeriginoides sacculifer as a paleotemperature proxy

Foraminiferal magnesium shows increasing promise as a paleothermometer, but the accuracy and precision are limited by biases introduced by partial dissolution, salinity variations, Mg-rich gametogenic calcite, and contaminant phases. We improved cleaning methods and reduced errors introduced by partial dissolution by sampling from well-preserved cores in the equatorial Atlantic and the Caribbean Sea with different dissolution histories. All cores reveal a synchronous 25% increase in Mg/Ca from the stage 2/3 boundary to the Holocene core top, indicating that dissolution is not a controlling factor. Modern temperatures estimated from core top Mg/Ca are 24.5°–25.0°C, equal to mean annual water temperatures at 50–100 m. We estimate that sea surface temperature increased by 2.6°C (±1.3) from the last glacial maximum to the Holocene. Holocene values were comparable to those during isotope stage 5e. Our data indicate that biases from contaminant phases and partial dissolution can be reduced. This paleothermometer holds promise if uncertainties introduced by salinity variations and gametogenic calcite can be constrained.

Dissolution of calcite in deep-sea sediments: pH and O2 microelectrode results

Abstract We present microelectrode profiles of oxygen, pH, and porosity measured at in situ depth across the sediment-water interface from stations above and below the calcite saturation horizon in the equatorial Atlantic. Resistivity electrode data indicate that the diffusion coefficient for solutes is attenuated by nearly a factor of two within the top centimeter of the sediment lattice. Oxygen and porosity data are used to estimate that the organic carbon respiration rate of the sediment is 10–15 μmol cm −2 yr −1 in these locations. Comparison of the pH data with numerical model results reveals that sedimentary calcite dissolves in response to respiration in the top few centimeters of sediment even above the saturation horizon. We calculate that the effective in situ dissolution rate constant for calcite required to explain these data is the neighborhood of 10–100% d −1 . Although more data are needed to confirm this result, it appears to be at least an order of magnitude less than the rate observed in laboratory stirred-reactor experiments.

Environmental oxidation rate of manganese(II): bacterial catalysis

Abstract A simple mass balance for dissolved manganese(II) in waters containing low levels of oxygen in Saanich Inlet indicates that the residence time for Mn(II) removal to the solid phase is on the order of a few days. The average oxidation state of Mn in particulate material sampled from the region of Mn removal was 2.3–2.7, and electron micrographs revealed structures characteristic of bacterially formed Mn precipitates. Radiotracer experiments utilizing 54 Mn(II) indicated that removal of Mn from solution in the region of active uptake was substantially blocked by a poison mixture, demonstrating that Mn(II) binding to particulates is catalyzed by bacteria in this environment.

Early diagenesis in sediments from the eastern equatorial Pacific, I. Pore water nutrient and carbonate results

Interstitial waters were extracted from cores at three locations in the eastern equatorial Pacific and analyzed for nutrients, dissolved carbonate species, Mn and Fe. From the depth variation in pore water chemistry, we infer that organic matter oxidation reactions occur with depth in the following sequence: O2 reduction, NO3− and MnO2 reduction, and then ferric iron reduction. From NO3− results we infer that O2 is largely or totally consumed within the top few centimeters of sediment. NO3− is completely reduced at a sediment depth of 20 cm at a site near the crest of the East Pacific Rise, but is preserved at levels of 20–30 μmol/kg at 40 cm depth at a Guatemala Basin site. We have calculated the alkalinity for pore water samples assuming ions diffuse according to relative ionic diffusion coefficients, that the stoichiometry of organic matter oxidation reactions is that of “Redfield” organic matter, and that the pore waters are saturated throughout with respect to CaCO3. The measured alkalinity increase is only about half of the predicted value. The difference is probably a result of either enhanced mixing of the pore water in the top few centimeters of sediments by biological or physical processes, or the occurrence of an inorganic reaction which consumes alkalinity. At depths of oxygen and nitrate reduction in the sediments, the ion concentration product of CaCO3 is the same, within the analytical error, as the solubility product of Ingle et al. [34] at 1 atm and 4°C. This result indicates CaCO3 resaturation on pressure change during coring. Where pore water Mn concentrations become measurable, the ion concentration product increases, indicating either supersaturation with respect to calcite or that another phase is controlling the carbonate solubility.