Karl K. Turekian

Distribution of the Elements in Some Major Units of the Earth's Crust

This paper presents a table of abundances of the elements in the various major units of the Earths lithic crust with a documentation of the sources and a discussion of the choice of units and data.

The fate of metals in the oceans

The role of particules in cleaning the aqueous system of metals from land to sea can be demonstrated using natural 210Pb.

Transport and residence times of tropospheric aerosols inferred from a global three‐dimensional simulation of 210Pb

A global three-dimensional model is used to investigate the transport and tropospheric residence time of 210Pb, an aerosol tracer produced in the atmosphere by radioactive decay of 222Rn emitted from soils. The model uses meteorological input with 4°×5° horizontal resolution and 4-hour temporal resolution from the Goddard Institute for Space Studies general circulation model (GCM). It computes aerosol scavenging by convective precipitation as part of the wet convective mass transport operator in order to capture the coupling between vertical transport and rainout. Scavenging in convective precipitation accounts for 74% of the global 210Pb sink in the model; scavenging in large-scale precipitation accounts for 12%, and scavenging in dry deposition accounts for 14%. The model captures 63% of the variance of yearly mean 210Pb concentrations measured at 85 sites around the world with negligible mean bias, lending support to the computation of aerosol scavenging. There are, however, a number of regional and seasonal discrepancies that reflect in part anomalies in GCM precipitation. Computed residence times with respect to deposition for 210Pb aerosol in the tropospheric column are about 5 days at southern midlatitudes and 10–15 days in the tropics; values at northern midlatitudes vary from about 5 days in winter to 10 days in summer. The residence time of 210Pb produced in the lowest 0.5 km of atmosphere is on average four times shorter than that of 210Pb produced in the upper atmosphere. Both model and observations indicate a weaker decrease of 210Pb concentrations between the continental mixed layer and the free troposphere than is observed for total aerosol concentrations; an explanation is that 222Rn is transported to high altitudes in wet convective updrafts, while aerosols and soluble precursors of aerosols are scavenged by precipitation in the updrafts. Thus 210Pb is not simply a tracer of aerosols produced in the continental boundary layer, but also of aerosols derived from insoluble precursors emitted from the surface of continents. One may draw an analogy between 210Pb and nitrate, whose precursor NOx is sparingly soluble, and explain in this manner the strong correlation observed between nitrate and 210Pb concentrations over the oceans.

The osmium isotopic composition of the continental crust

Osmium isotopic compositions and concentrations have been determined for deltaic and continental shelf sediments from three major rivers (the Amazon, Changjiang, and Mississippi) and for two loesses from the upper Mississippi River Valley. The loesses have heavy mineral and clay compositions characteristic of Mississippi river and delta sediment. Concentrations range from 15 to 120 pg Os/g, and 187Os/186Os ratios range from 8.2 to 10.9. Sediments which are polluted (surface sediment from the Changjiang Estuary) or which have scavenged metals from seawater (Gulf of Mexico slope surface sediment and distal Amazon Shelf sediment) have high osmium concentrations and less radiogenic isotopic compositions. Purely terrigenous sediments have 15–90 pg Os/g, and 187Os/186Os= 10.0–10.9. North American loess and Mississippi Delta sediment osmium isotopic compositions are indistinguishable. We believe that 187Os/186Os= 10–11 is a good average for currently eroding upper continental crust—these sediments have neodymium isotopic compositions close to the global mean for river sediment. Terrigenous 187Os/186Os is significantly more radiogenic than seawater 187Os/186Os(8.6), requiring a nonriverine source of osmium to seawater. Terrigenous osmium is also significantly more radiogenic than bulk pelagic clay osmium. We use the improved estimate of terrigenous 187Os/186Os to recalculate the flux of extraterrestrial osmium to Pacific marine sediment. If the residence age of osmium in an upper crustal region is similar to the neodymium model age of a sediment derived from that region, a model-dependent upper crustal 187Re/186Os ratio can be calculated. For the model ages of the sediments analyzed in this study (Tdm = 1.5–1.7 Ga), model 187Re/186Os ratios are 340–360 and can be used to correct for bias resulting from the preferential erosion of young crust. The 187Os/186Os ratio of 2.2 Ga crust with 187Re/186Os= 400 is 14 ± 2. We have also used the data from this study to estimate upper crustal osmium and rhenium concentrations of 50 pg/g and 390 pg/g, respectively.

Effects of biological sediment mixing on the210Pb chronology and trace metal distribution in a Long Island Sound sediment core

An experiment was designed to assess the relative importance of sediment accumulation and bioturbation in determining the vertical distribution of nuclides in estuarine sediments. A diver-collected core, 120 cm long, was raised from central Long Island Sound and analyzed down its length for:210Pb and226Ra;239, 240Pu; and Mn, Zn, Cu, and Pb. Sampling for chemical analysis was guided by X-radiography of the core. Excess210Pb (relative to226Ra) is roughly homogeneous in the top 2–4 cm of the core, then decreases quasi-exponentially to zero at (or above) 15 cm.239, 240Pu and excess Zn, Cu, and Pb, relative to background values at greater depths in the core, are distributed like excess210Pb in the top 10–15 cm. The absence of Mn enrichment at the top of the core, in contrast to other cores raised from this station, suggests that 1–3 cm of sediment was lost by erosion at the site of this core sometime prior to sampling. Below 15 cm excess210Pb and excess Zn, Cu, and Pb are found only in the bulk sample from 25 to 30 cm and in clearly identifiable burrow fillings dissected from 70 cm and 115 cm depth. Infilling of large burrows, excavated and then abandoned by crustaceans, is therefore a mechanism for transfer of surficial material to depth in these sediments. The bioturbation rate in the top several centimeters at this station has been determined previously using234Th (24-day half-life). The distribution of239, 240Pu can be used to estimate a bioturbation rate for the underlying layer (to ∼10 cm depth); this rate is found to be 1–3% of the maximum mixing rate for the top 2–3 cm. Using these two mixing rates in a composite-layer, mixing + sedimentation model, the distribution of excess210Pb in the top 15 cm was used to constrain the sediment accumulation rate, ω. While the apparent rate of sediment accumulation (assuming no mixing below 2–4 cm) is 0.11 cm/yr, the model requires ω < 0.05 cm/yr. Thus in an area of slow sediment accumulation, a low rate of bioturbation below the surficial zone of rapid mixing causes an increase of at least a factor of two in apparent accumulation rate.

The distribution of 210Pb and 210Po in the surface waters of the Pacific Ocean

Abstract By modelling the observed distribution of 210 Pb and 210 Po in surface waters of the Pacific, residence times relative to particulate removal are determined. For the center of the North Pacific gyre these are τ Po = 0.6years andτ Pb = 1.7years . The surface ocean τ Pb is determined by particulate transport rather than plankton settling. The fact that it is about two orders of magnitude smaller than τ Pb for the deep ocean implies a sharp change in the adsorptive quality of particles during descent through the water column.

The investigation of the geographical and vertical distribution of several trace elements in sea water using neutron activation analysis

Abstract Methods for the determination of 18 trace elements in sea water by neutron activation analysis have been developed. Of these elements sufficient analyses have been completed for gold, selenium, antimony, silver, cobalt and nickel to permit a discussion of their distributions in the world ocean. The distributions of gold, selenium and antimony are more uniform than those of silver, cobalt and nickel, so little can be said regarding the processes important to their supply to and removal from the world ocean. The variations observed for silver, cobalt and nickel, however, make possible the evaluation of several factors which influence their concentrations in the ocean: The Atlantic Ocean north of 10°S receives 60 per cent of the total dissolved material supplied to the ocean by streams, but is lower in silver, cobalt and nickel than is the remainder of the world ocean indicating that continental run-off is not important in the deep-sea economy of these elements. Relatively high concentrations of cobalt and nickel are found in water of the central Pacific Ocean in areas where the cobalt/manganese and nickel/manganese ratios in manganese nodules are high. The evidence that the nodule material in these areas is of volcanic origin may also imply a volcanic origin for the cobalt and nickel in the overlying water. No such source is indicated for silver. Unweathered rock material supplied directly to the ocean by Antarctic glaciers seems to be sufficient to affect the silica economy of the ocean and may in part affect the economy of other trace elements such as cobalt and nickel. The data are not sufficient, however, to permit an unambiguous interpretation. Cobalt and nickel are relatively low in Long Island Sound compared to the North Atlantic indicating near-shore removal of these elements from sea water. In the deep sea much of the cobalt and nickel may be removed by co-precipitation with manganese oxide, but no mechanism seems singularly important in the removal of silver. Increase in concentration of silver, cobalt and nickel with depth in areas of high organic productivity indicates alteration of trace element concentration by organic reactions. In areas where upwelling currents oppose the downward movement of organic material a relatively high steady state concentration may be produced.

Inhomogeneous accumulation of the earth from the primitive solar nebula.

Model for accumulation of earth and planets from primitive solar nebula, implying inhomogeneous chemical composition of bodies in solar system

Radiocarbon and 210Pb distribution in submersible-taken deep-sea cores from Project FAMOUS

During the FAMOUS survey of the Mid-Atlantic Ridge in August and September, 1974 by the research submersible “Alvin” two cores were taken for radiochemical analysis. One core (527-3) was 24 cm long and the other (530-4) was 17 cm long. Both were from water depths of about 2500 m. Slices of the cores were analyzed for radiocarbon and 210 Pb. In the top 8 cm layer of 527-3 dates are constant with depth at about 2400 yr B.P. Below 8 cm radiocarbon dates increase linearly yielding an accumulation rate of 2.9 cm/10 3 yr. The constant age from the surface to a depth of 8 cm can be attributed to biogenic mixing to that depth with no significant mixing below 8 cm. The excess 210 Pb pattern yields a mixing coefficient of 0.6 × 10 −8 cm 2 /sec. The top 2 cm of core 530-4 has a 14 C date of 13,000 yr B.P. Below 4 cm dates increase from 16,400 to 18,000 yr B.P., but this increase probably is not statistically significant. The data indicate physical disruption of the section. The date of this disruption is not defined by the data but the restriction of excess 210 Pb to the top centimeter of the core implies either that sediment accumulation at this site has only recently resumed or that both the rate of accumulation and rate and depth of bioturbation have been very small since the disrupting event.

The geochemical cycle of rhenium: a reconnaissance

Rhenium (Re) is one of a suite of elements (including uranium and molybdenum) that display conservative behavior in seawater and are enriched in anoxic sediments. The decay of187Re to187Os provides a geochronometer in ancient sedimentary rocks and gives rise to Os-isotopic variations in nature. In order to better characterize its sources to seawater, Re was measured in three major rivers (Amazon, Orinoco, Ganges-Brahmaputra) and some of their tributaries. Re concentrations span four orders of magnitude (from < 0.02 to 400 pmol/kg), with the highest concentrations found in rivers draining black shales in the Venezuelan Andes. Mainstream Re levels in the three rivers are between 1 and 10 pmol/kg, with a flux weighted average of 2.3 pmol/kg. The residence time for Re in the oceans is estimated to be 750,000 yr with respect to river inputs. Re profiles from the Atlantic and Pacific Oceans confirm that Re behaves conservatively in seawater, with no significant uptake onto particles and/or recycling within the water column. This is also true in the anoxic water column of the Black Sea. Re removal into anoxic sediments occurs at or below the sediment water interface, as demonstrated in sediment pore waters from Chesapeake Bay. In oxic sediments, Re is not cycled with manganese oxides, and it is not enriched in very slowly accumulating pelagic sediments with a large hydrogenous iron and manganese component, or in manganese nodules. Burial of Re in anoxic sediments, which accumulate on 0.3% of the ocean floor, removes approximately 50% of the riverine Re flux to the oceans. Hence, oceanic Re concentrations may be very sensitive to changes in the area of anoxic sedimentation.