S. Krishnaswami

The global-average production rate of10Be

Precipitation collected in continuously open containers for about a year at seven sites around the United States was analyzed for10Be,90Sr,210Pb and238U. Based on these data and long-term precipitation,90Sr and210Pb delivery patterns, the stratospheric, tropospheric and recycled10Be components in the collections were estimated and the global10Be production rate was assessed. Single station production rate estimates range from 0.52 × 106 atoms cm−2 yr−1 to 2.64 × 106 atoms cm−2 yr−1. The mean value is 1.21 × 106 atoms cm−2 yr−1 with a standard error of 0.26 × 106 atoms cm−2 yr−1.

The record of sea water 187Os/186Os variation through the Cenozoic

The variation of sea water 187Os/186Os during the Cenozoic has been determined by analyzing the hydrogenous Os released by acidic hydrogen peroxide leaching of layers in an abyssal North Pacific pelagic clay core. Overall, sea water 187Os/186Os has increased from 3.2 at 58 Ma to the present-day value of ∼ 8.6. This pattern is similar to the well-known 87Sr/86Sr evolution of sea water. The increase in the proportion of continental Os (187Os/186Os⩾ 10) could be due to enhanced weathering rates of black shales associated with the uplift of the Himalayas. Other elements associated with black shale weathering, such as U and P, should show effects in the sedimentary record.

Atmospherically-derived radionuclides as tracers of sediment mixing and accumulation in near-shore marine and lake sediments: Evidence from7Be,210Pb, and239,240Pu

Abstract Cosmogenic 7 Be(t 1/2 = 53.3days) has been used to estimate particle-mixing rates in the upper layers of lacustrine and near-shore marine sediments. Excess 210 Pb and/or 239,240 Pu have provided limits on rates of sediment accumulation in these environments and indices of the efficiency of the sediments as collectors of reactive nuclides over longer time scale. In sediment cores from Long Island Sound (marine) and Lake Whitney (fresh-water) 7 Be was measurable in the top 2–3 cm. Diffusion-analog particle-mixing coefficients calculated from these data are in the range of 10 −7 cm 2 /s. For Long Island Sound the coefficients are lower by factors of 3–6 than those estimated from the depth distributions of excess 234 Th at the same stations [14]. For Lake Whitney the calculated mixing coefficient is an upper limit because of the possibility of a sampling artifact. Measurements of total (wet + dry) atmospheric deposition of 7 Be in New Haven give an average flux of 0.07 dpm/cm 2 day during March-November, 1977; this is equivalent to a steady-state inventory of 5.4 dpm/cm 2 in a perfect collector. Sediment cores from Long Island Sound contain about half this 7 Be inventory, consistent with either a mean residence time for 7 Be in the water column of about one half-life or with post-depositional loss of 7 Be from Long Island Sound sediments. The Lake Whitney cores contain about 5 dpm/cm 2 , much nearer the atmospheric delivery. A higher inventory of 7 Be in fresh-water, as compared to marine, sediments could be due either to a shorter mean residence time for 7 Be in fresh water or to lateral transport processes in the lake or its catchment. High inventories of excess 210 Pb and 239,240 Pu in Lake Whitney sediments demonstrate the importance of lateral transport on longer time scales at least.

210Po and210Pb distributions in the central and eastern Indian Ocean

Abstract Disequilibrium between 210 Po and 210 Pb and between 210 Pb and 226 Ra has been mapped in the eastern and central Indian Ocean based on stations from Legs 3 and 4 of the GEOSECS Indian Ocean expedition. 210 Po/ 210 Pb activity ratios are less than 1.0 in the surface mixed layer and indicate a residence time for Po of 0.6 years. 210 Po and 210 Pb are generally in radioactive equilibrium elsewhere in the water column except at depths of 100–500 m, where Po may be returned to solution after removal from the surface water, and in samples taken near the bottom at a few stations. 210 Pb excesses relative to 226 Ra are observed in the surface water but these excesses are not as pronounced as in the North Pacific and North Atlantic. The difference is attributable to a lower flux of 210 Pb from the atmosphere to the Indian Ocean. Below the main thermocline, 210 Pb activities increase with depth to a broad maximum before decreasing to lower values near the bottom. Departures from this pattern are especially evident at stations taken in the Bay of Bengal (where 210 Pb/ 226 Ra activity ratios as low as 0.16 are observed) and near the Mid-Indian Ridge. The data suggest that removal of 210 Pb at oceanic boundaries, coupled with eddy diffusion along isopycnals, can explain gradients in 210 Pb near the boundary. Application of a simple model including isopycnal diffusion, chemical removal, production and radioactive decay produces fits the observed 210 Pb/ 226 Ra gradients for eddy diffusion coeffients of ∼ 10 7 cm 2 /s. High productivity in surface waters of the Bay of Bengal makes this region a sink for reactive nuclides in the northern Indian Ocean.

Uranium and thorium series nuclides in oriented ferromanganese nodules: growth rates, turnover times and nuclide behavior

Three ferromanganese nodules handpicked from the tops of 2500 cm2 area box cores taken from the north equatorial Pacific have been analysed for their U-Th series nuclides.230Thexc concentrations in the surface 1–2 mm of the top side of the nodules indicate growth rates of 1.8–4.6 mm/106 yr. In two of the nodules a significant discontinuity in the230Thexc depth profile has been observed at ∼0.3 m.y. ago, suggesting that the nodule growth has been episodic. The concentration profiles of231Paexc (measured via227Th) yield growth rates similar to the230Thexc data. The bottom sides of the nodules display exponential decrease of230Thexc/232Th activity ratio with depth, yielding growth rates of 1.5–3.3 mm/106 yr. The230Thexc and231Paexc concentrations in the outermost layer of the bottom face are significantly lower than in the outermost layer of the top face. Comparison of the extrapolated230Thexc/232Th and230Thexc/231Paexc activity ratios for the top and bottom surfaces yields an “age” of (5−15) × 104 yr for the bottom relative to the top. This “age” most probably represents the time elapsed since the nodules have attained the present orientation. The210Pb concentration in the surface ∼0.1 mm of the top side is in large excess over its parent226Ra. Elsewhere in the nodule, up to ∼1 mm depth in both top and bottom sides,210Pb is deficient relative to226Ra, probably due to222Rn loss. The absence of210Pbexc below the outermost layer of the top face rules out the possibility of a sampling artifact as the cause of the observed exponentially decreasing230Thexc and231Paexc concentration profiles. The flux of210Pbexc to the nodules ranges between 0.31 and 0.58 dpm/cm2 yr. The exhalation rate of222Rn, estimated from the226Ra-210Pb disequilibrium is ∼570 dpm/cm2 yr from the top side and >2000 dpm/cm2 yr from the bottom side. 226Ra is deficient in the top side relative to230Th up to ∼0.5–1 mm and is in large excess throughout the bottom. The data indicate a net gain of226Ra into the nodule, corresponding to a flux of (24−46) × 10−3 dpm/cm2 yr. On a total area basis the gain of226Ra into the nodules is <20% of the226Ra escaping from the sediments. A similar gain of228Ra into the bottom side of the nodules is reflected by the high228Th/232Th activity ratios observed in the outermost layer in contact with sediments.

10Be concentrations and the long-term fate of particle-reactive nuclides in five soil profiles from California

Concentration-depth profiles of cosmic-ray-produced 10 Be ( t 1/2 = 1.5m.y.) have been measured by accelerator-mass spectrometry in five soil profiles. These measurements were made in an effort (1) to understand the retentivity of soil surfaces for particle-reactive tracers depositing from the atmosphere on time scales of 10 4 –10 6 years, and (2) to explore the application of 10 Be as a chronometer of geomorphic surface age. The profiles sampled are from two wave-cut terraces located near Mendocino, California, a table mountain top and an alluvial fan, both located near Friant, California. The ages of the Mendocino terraces are inferred to be (1–5) × 10 5 years based on amino-stratigraphic correlations and models of terrace evolution; those of the table mountain top and alluvial fan are 9.5 × 10 6 years and 6.0 × 10 5 years, respectively, based on K-Ar analyses. All the surfaces sampled are nearly flat and exhibit few erosional features. In addition to 10 Be we measured 210 Pb, 239,240 Pu and 7 Be to ascertain the retentivity of the soils for particle-reactive nuclides and to assess the present-day delivery rate of nuclides from the atmosphere. The 7 Be inventory is 4.0 dpm/cm 2 similar to those observed at nearby locations. The inventories of 210 Pb and Pu isotopes conform to those predicted from model calculations and suggest that the soil surfaces sampled retain the entire burden of particle-reactive nuclides delivered to them over short time scales, ∼ 100 years. The 10 Be concentrations in the sample range between (0.2 and 7) × 10 8 atoms/g soil and show strong correlations with leachable Fe and/or Al. The inventory of 10 Be in the soil domain sampled is 1–2 orders of magnitude lower than that expected from the geological age of the surface and an average delivery rate of 10 Be from the atmosphere, 5.2 × 10 5 atoms/cm 2 yr. The low inventory of 10 Be is attributed to its loss from the soil domain sampled by solution transport. Based on a simple ☐-model type calculation with a first-order removal process for 10 Be, the residence time of 10 Be in the soil domains sampled is determined to be of the order of 10 4 years. The low residence time of 10 Be in the soil domains sampled requires that it be found either deeper in the regolith or in ground waters. In either case, the application of 10 Be as a chronometer of geomorphic surface age is severely constrained. However, the study of 10 Be in soils provides the only entry into the long-term (10 4 –10 6 years) behavior of particle-reactive nuclides in soils and, hence, could be important for understanding the behavior of analogous nuclides introduced into soils by natural and anthropogenic processes.

10Be and Th isotopes in manganese nodules and adjacent sediments: Nodule growth histories and nuclide behavior

Abstract The concentration profiles of 9 Be, 10 Be, 230 Th, 232 Th, 231 Pa (via 227 Th) and 238 U have been measured in three manganese nodules, one each from the North Pacific (A47-16(4)), the South Pacific (TF-5) and the Indian Ocean (R/V Vitiaz). In addition the 10 Be concentration in deep water from the GEOSECS reoccupation station 500 of the North Pacific, and in ☐ cores raised from the manganese nodule field in the North Pacific have been measured. The 10 Be concentration in nodule and seawater samples was measured by the accelerator masss spectrometric technique employing the Yale Tandem Van de Graaff accelerator. The concentrations of 10 Be, 230 Th exc and 231 Pa exc and ratios of 10 Be/ 9 Be and 230 Th exc / 232 Th all decrease with depth in the nodules. This decrease, interpreted in terms of nodule growth, yields “average” growth rates of a few millimeters per million years for the nodules. The growth rates of the nodules exhibit temporal variations, both on short time (∼ 50,000 years) and long time (several million years) scales. Of the three nodules studied, only in TF-5 is the short-term average growth rate based on 230 Th exc in the top 0–0.5 mm the same as the long-term average rate based on Be isotope data for the 0.5–17 mm interval. For the other two nodules, the recent average growth rates based on 230 Th exc data differ significantly from the long-term average growth rates based on Be isotopes. In A47-16(4) the 10 Be based rate is less than the 230 Th exc rate and in R/V Vitiaz the 10 Be based rate is greater than the 230 Th exc rate. This observation, coupled with measurable changes in growth rates even during the past few hundred thousand years, suggests, but does not prove, that the discordant growth rates deduced from 230 Th exc and 10 Be profiles document changes in nodule growth rate with time rather than mixing effects on 230 Th exc profiles. The 10 Be concentration in the GEOSECS North Pacific deep water is 6100±1200 atoms/g. This value coupled with the average surface 10 Be/ 9 Be ratio of North Pacific nodules predicts a 9 Be concentration within the limits of measured values. The inventory of 10 Be and 230 Th exc in the nodules is only ∼ 10% of the total, the remaining being in sediments. The 10 Be concentrations in the upper portions of two adjacent cores studied are nearly the same, but the deposition fluxes of both 10 Be and 230 Th based on 230 Th dating vary by a factor of two. This difference is attributable to local redistribution of sediment at the time of deposition prior to accumulation.

The isotopic composition of leachable osmium from river sediments

The 187Os186Os ratios of acidic hydrogen peroxide leaches of river sediments are inferred to be representative of the isotopic composition of Os delivered to sea water from the continents. The 187Os186Os values for river sediment leaches from the eastern and central U.S., the Ganges River, and the Rio Maipo range from 10.1 to 21.5. These results suggest that the 187Os186Os of the soluble river load delivered from the major continental areas to the oceans is, on average, more radiogenic than average currently eroding continental crust (10.5). Sediment leaches from three rivers from the western U.S. that drain ophiolitic peridotites have distinctly lower 187Os186Os values, 1.4–7.1; such rivers may be significant contributors to sea water of the relatively unradiogenic osmium required to yield the contemporary oceanic 187Os186Os of 8.6.

Sedimentary processes in the inner New York Bight: Evidence from excess210Pb and239,240Pu

Abstract Particle-reactive radionuclides were determined in sediments from the inner New York Bight to trace transport and storage of fine-grained sediments and associated reactive materials. Seven sediment ☐ cores 20–50 cm in length were analyzed for water content, loss on ignition (LOI) and excess 210 Pb; three of these were also analyzed for 239,240 Pu. Excepting some depth horizons in a core from a dredge-spoil dumpsite, every sample analyzed contained excess 210 Pb. Variations in the concentration of excess 210 Pb with depth in the sediment at all stations correlated strongly with LOI, which apparently traces that fraction of the sediment which is active in removing reactive elements from the water column. In the cores analyzed for 239,240 Pu, every sample contained finite Pu, and Pu concentrations correlated strongly with excess 210 Pb. The radionuclide distributions may be simply viewed as products of steady-state sediment accumulation or of mixing. Geochemically reasonable accumulation rates are very high (0.5–2.6 g/cm 2 y) and could probably only be sustained by offshore transport of dumped materials. At the other extreme the relationships between excess 210 Pb and LOI are compatible with rapid mixing of a 210 Pb carrier phase (traced by LOI) into the pre-existing substrate with little or no actual accumulation. Other non-steady-state processes, such as sediment gravity flow, could also explain the observed distributions. Measured sediment inventories (dpm/cm 2 ) of excess 210 Pb and Pu at these stations are greatly in excess of those supportable by direct atmospheric deposition: lateral supply is required. Incorporation of sedimentary fines into the sand substrate could make the inner New York Bight an important repository of reactive materials.

Retardation of 238U and 232Th decay chain radionuclides in Long Island and Connecticut aquifers

Abstract Knowledge of the ability of an aquifer to retard the groundwater transport of toxic or radioactive ions can be inferred from the analysis of groundwater for the radionuclides of the 238U- and 232Th-decay chains. Groundwaters of varying chemical composition were analyzed from wells in Long Island, New York, and Connecticut. Aquifer lithologies are arkose, quartz sandstone, granite, and glacial drift. Adsorption (k1) and desorption (k2) coefficients for Ra and Pb were calculated using 222Rn activity as a measure of the supply of other α-recoil nuclides. Laboratory tests of the validity of this assumption were made by measuring the flux of 224Ra and 222Rn from aquifer solids. The ratio k 1 k 2 is the distribution coefficient, K, which is effectively equal to R ƒ , the retardation factor. The average value of K for Ra is 6 × 102 in Long Island aquifers and 5 × 104 in Connecticut. The distribution coefficient for Pb is 104 in Long Island and 105 in Connecticut. Results from this and other studies reveal a strong dependence of retardation on pH, ƒ O 2 , and ionic strength that tends to overwhelm any dependence of retardation on lithology.