David J. Hirschberg

Export flux of carbon at the equator during the EqPac time-series cruises estimated from 234Th measurements

Abstract Distributions of 234 Th were determined in three particle-size classes ( > 53, 1–53 and 0.7–1.0 μm) and in filtered seawater during each of the two time-series cruises of the U.S. JGOFS Process Study in the equatorial Pacific. Four vertical profiles were measured on the equator at 140°W from the sea surface to 400 m depth between 24 March and 9 April 1992 (Time-series I) and again between 3 and 18 October 1992 (Time-series II). In addition, both organic and inorganic carbon were measured in each of the particle fractions. The results were used with a one-dimensional model, which includes the equatorial upwelling, to estimate the flux of particulate carbon sinking out of the surface layer. The flux of particulate organic carbon (POC) at the base of the euphotic zone (0.1 % light level, 120 m depth) was estimated to average 1.9 mmol m −2 day −1 during El Nino (Time-series I) and 2.4 mmol m −2 day −1 during the cold period that followed (Time-series II). These values amount to only ∼ 2% of the primary production measured during each of the same periods and are insufficient to balance the new production, estimated previously to be ∼ 17% of primary production. These results are consistent with the hypothesis that the major part of the new production is removed from the region by advection in the form of dissolved organic matter. The POC flux profile indicates a net remineralization below the 1 % light level (80 m depth) such that the flux reaching 200 m depth has been reduced by ∼ 55%, giving a remineralization length scale of ∼ 155 m. For particulate inorganic (carbonate) carbon the flux at 200 m averaged 0.54 mmol m −2 day −1 during Time-series I and 0.71 mmol m −2 day −1 during Time-series II, very similar to the fluxes reported in deep sediment traps deployed at the same time. Estimates of the average large-particle sinking velocity give values −1 in the upper part of the euphotic zone, show a sharp increase near the base of the euphotic zone and level off to values of 30–60 m day −1 at 200 m depth.

Accretion rates and sediment accumulation in Rhode Island salt marshes

In order to test the assumption that accretion rates of intertidal salt marshes are approximately equal to rates of sea-level rise along the Rhode Island coast,210Pb analyses were carried out and accretion rates calculated using constant flux and constant activity models applied to sediment cores collected from lowSpartina alterniflora marshes at four sites from the head to the mouth of Narragansett Bay. A core was also collected from a highSpartina patens marsh at one site. Additional low marsh cores from a tidal river entering the bay and a coastal lagoon on Block Island Sound were also analyzed. Accretion rates for all cores were also calculated from copper concentration data assuming that anthropogenic copper increases began at all sites between 1865 and 1885. Bulk density and weight-loss-on-ignition of the sediments were measured in order to assess the relative importance of inorganic and organic accumulation. During the past 60 yr, accretion rates at the eight low marsh sites averaged 0.43±0.13 cm yr−1 (0.25 to 0.60 cm yr−1) based on the constant flux model, 0.40±0.15 cm yr−1 (0.15 to 0.58 cm yr−1) based on the constant activity model, and 0.44±0.11 cm yr−1 (0.30 to 0.59 cm yr−1) based on copper concentration data, with no apparent trend down-bay. High marsh rates were 0.24±0.02 (constant flux), 0.25±0.01 (constant activity), and 0.47±0.04 (copper concentration data). The cores showing closest agreement between the three methods are those for which the excess210Pb inventories are consistent with atmospheric inputs. These rates compare to a tide gauge record from the mouth of the bay that shows an average sea-level rise of 0.26±0.02 cm yr−1 from 1931 to 1986. Low marshes in this area appear to accrete at rates 1.5–1.7 times greater than local relative sea-level rise, while the high marsh accretion rate is equal to the rise in sea level. The variability among the low marsh sites suggests that marshes may not be poised at mean water level to within better than ±several cm on time scales of decades. Inorganic and organic dry solids each contributed about 9% by volume to low marsh accretion, while organic dry solids contributed 11% and inorganic 4% to high marsh accretion. Water/pore space accounted for the majority of accretion in both low and high marshes. If water associated with the organic component is considered, organic matter accounts for an average of 91% of low marsh and 96% of high marsh accretion. A dramatic increase in the organic content at a depth of 60 to 90 cm in the cores from Narragansett Bay appears to mark the start of marsh development on prograding sand flats.

DISTRIBUTION OF HEAVY METAL AND PCB CONTAMINANTS IN THE SEDIMENTS OF AN URBAN ESTUARY : THE HUDSON RIVER

Abstract Surficial sediments obtained from sediment cores were collected over 100km along the axis of the lower Hudson River in June 1994, November 1994, May 1995 and April 1996 and showed the presence of anthropogenic Ag, Cd, Cu, Pb, Zn, PCBs in all samples. Contaminant distributions in the Hudson River estuary show two types of trends: Ag, Cu and Pb show an increasing trend down-estuary with maximum values in New York Harbor sediments; in contrast, Cd, Zn and total PCBs display a decreasing trend toward New York Harbor where urban sources are also apparent. Silver is a useful tracer of urban sources of contaminants in the Hudson River estuary and polychlorinated biphenyls (PCBs) are useful source indicators of upriver sources. Correlations of Cu and Pb with Ag suggest that Ag, Cu and Pb are dominated by down-estuary sources such as wastewater effluent. The history of their inputs suggests that they have been progressively transported downstream. Correlations of Cd and Zn with total PCBs indicate that these contaminants are dominated by upriver sources, where they are removed and diluted downstream along with the sediment transport.

Determination of thorium isotopes in seawater by nondestructive and radiochemical procedures

Abstract Procedures have been developed for the analyses of dissolved and particulate 234Th, 228Th, 230Th and 232Th in seawater. Large volume samples (>1000 1) are collected using in situ pumps. Seawater is pumped sequentially through a filter cartridge and two MnO2 adsorbers for the collection of particulate and dissolved Th, respectively. Both filters adsorbers are analysed for 234Th using a simple gamma counting technique. This newly developed 234Th procedure can be conducted at sea, and thus provides an easy and efficient method for 234Th analyses on large volume samples. Subsequent radiochemical purification procedures and low-level alpha counting techniques are used in the laboratory for the analyses of 228Th, 230Th and 232Th on these same samples.

210Pb scavenging in the North Atlantic and North Pacific Oceans

The radionuclide210Pb shows significant geographic variations in the extent of its removal from the open ocean water column. This “texture of scavenging” is defined by mapping: (1) the integrated deficiency of210Pb in the water column, relative to its supply from the atmosphere and from in situ decay of dissolved226Ra, and (2) inventories of excess210Pb in deep-sea sediments. The ratio of210Pb deficiency to its supply, termed the scavenging effectiveness, is ∼ 20% in the North Equatorial Pacific and ∼ 50% in the North Atlantic. This variation is related to the combined effects of uptake of210Pb onto sinking particles and lateral transport of210Pb to areas of more intense removal. Sediment inventories of excess210Pb, normalized to the210Pb deficiency in the overlying water column, permit evaluation of the relative importance of these effects. In the North Equatorial Pacific virtually all of the210Pb removed from the water column is present in the underlying sediments but in the mid-latitude North Atlantic, the sediments comprise only about 50% of the210Pb removed. The deficiencies of210Pb in the mid-latitude North Atlantic sediments south of 50°N are qualitatively offset by surpluses in high-latitude sediments north of 50°N. Higher primary productivity and new production in the surface waters of the high-latitude North Atlantic and North Equatorial Pacific, relative to the oligotrophic central North Atlantic, may account for the greater fluxes of210Pb to bottom sediments in those areas.

Natural and anthropogenic radionuclide distributions in the northwest Atlantic Ocean

We have used in-situ pumps which filter large volumes of sea water through a 1 μm cartridge prefilter and two MnO2-coated cartridges to obtain information on dissolved and particulate radionuclide distributions in the oceans. Two sites in the northwest Atlantic show subsurface maxima of the fallout radionuclides137Cs,239,240Pu and241Am. Although the processes of scavenging onto sinking particles and release at depth may contribute to the tracer distributions, comparison of predicted and measured water column inventories suggests that at least 35–50% of the Pu and241Am are supplied to the deep water by advection. The depth distributions of the naturally occurring radionuclides232Th,228Th and230Th reflect their sources to the oceans.232Th shows high dissolved concentrations in surface waters, presumably as a result of atmospheric or riverine supply. Activities of232Th decrease with depth to values ⩽ 0.01 dpm/1000 l.228Th shows high activities in near surface and near bottom water, due to the distribution of its parent,228Ra. Dissolved230Th, produced throughout the water column from234U decay, increases with depth to ∼ 3000 m. Values in the deep water (> 3000 m) are nearly constant (∼ 0.6–0.7 dpm/1000 l), and the distribution of this tracer (and perhaps other long-lived particle-reactive tracers as well) may be affected by the advection inferred from Pu and241Am data. The ratio of particulate to dissolved activity for both230Th and228Th is ∼ 0.15–0.20. This similarity precludes the calculation of sorption rate constants using a simple model of reversible sorption equilibrium. Moreover, in mid-depths228Th tends to have a higher particulate/dissolved ratio than230Th, suggesting uptake and release of230Th and228Th by different processes. This could occur if228Th, produced in surface water, were incorporated into biogenic particles formed there and released as those particles dissolved or decomposed during sinking.230Th, produced throughout the water column, may more closely approach a sorption equilibrium at all depths.230Th,241Am and239,240Pu are partitioned onto particles in the sequence Th > Am > Pu with ∼ 15% of the230Th on particles compared with ∼ 7% for Am and ∼ 1% for Pu. Distribution coefficients (Kd) are 1.3–1.6 × 107 for Th, 5–6 × 106 for Am and 7–10 × 105 for Pu. The lower reactivity for Pu is consistent with analyses of Pu oxidation states which show ∼ 85% oxidized (V + VI) Pu. However, theKd value for Pu may be an upper limit because Pu, like228Th, may be incorporated into particles in surface waters and released at depth only by destruction of the carrier phase.

Natural and anthropogenic radionuclide distributions in the Nansen Basin, Artic Ocean: Scavenging rates and circulation timescales

Abstract Determination of the naturally occurring radionuclides 232 Th, 230 Th, 228 Th and 210 Pb, and the anthropogenic radionuclides 241 Am, 239,240 Pu, 134 Cs and 137 Cs in water samples collected across the Nansen Basin from the Barents Sea slope to the Gakkel Ridge provides tracers with which to characterize both scavenging rates and circulation timescales in this portion of the Arctic Ocean. Large volume water samples (∼ 15001) were filtered in situ to separate particulate (> 0.5 μm) and dissolved Th isotopes and 241 Am. Thorium-230 displays increases in both particulate and dissolved activities with depth, with dissolved 230 Th greater and particulate 230 Th lower in the deep central Nansen Basin than at the Barents Sea slope. Dissolved 228 Th activities also are greater relative to 228 Ra, in the central basin. Residence times for Th relative to removal from solution onto particles are ∼1 year in surface water, ∼10 years in deep water adjacent to the Barents Sea slope, and ∼20 years in the Eurasian Basin Deep Water. Lead-210 in the central basin deep water also has a residence time of ∼20 years with respect to its removal from the water column. This texture of scavenging is reflected in distributions of the particle-reactive anthropogenic radionuclide 241 Am, which shows higher activities relative to Pu in the central Nansen Basin than at the Barents Sea slope. Distributions Of 137 Cs show more rapid mixing at the basin margins (Barents Sea slope in the south, Gakkel Ridge in the north) than in the basin interior. Cesium-137 is mixed throughout the water column adjacent to the Barents Sea slope and is present in low but detectable activities in the Eurasian Basin Deep Water in the central basin. At the time of sampling (1987) the surface water at all stations had been labeled with 134 Cs released in the 1986 accident at the Chernobyl nuclear power station. In the ∼1 year since the introduction of Chernobyl 134 Cs to the Nansen Basin, it had been mixed to depths of ∼800 m at the Barents Sea Slope and to ∼300 m in the central basin. “PreChernobyl” inventories of 137 Cs (as well as 239,240 Pu) are 10 times those expected from global atmospheric fallout from nuclear weapons testing and are derived principally from releases from the Sellafield, U.K., nuclear fuel reprocessing facility on the Irish Sea. Based on the sources Of 137 Cs to the Nansen Basin, mixing time scales are 9–18 years for the upper water column (to 1500 m) and ∼40 years for the deep water. These mixing time scales, combined with more rapid scavenging at the basin margin relative to the central basin, produce residence times of particle-reactive radionuclides in the Nansen Basin comparable to other open ocean areas (e.g. north-west Atlantic) despite the presence of permanent ice cover and long periods of low-light levels that limit productivity in the Arctic.

Transport and sources of metal contaminants over the course of tidal cycle in the turbidity maximum zone of the Hudson River estuary

Transport and source of metal contaminants (Ag, Cd, Cu, Pb and Zn) in the turbidity maximum zone of the Hudson River estuary were studied over the course of a tidal cycle in November 1994 and August 1995. This study showed that the metal/Fe ratios in suspended particles varied more widely than those in the local sediments, implying that sources of metal contaminants to the water column are not only from local sediment resuspension but also from lateral advection. Although the metal/Fe ratios oscillate with time at the anchor stations, a general increasing trend with salinity was found over the course of a tidal cycle, suggesting that the lower estuary could be a source of metal contaminants to the upper estuary regions. These results support that sediment resuspension and lateral advection are important factors responsible for transporting the lower estuary contaminants up river and redistributing metal contaminants in the estuary.

234Th and 7Be as tracers for transport and sources of particle-associated contaminants in the Hudson River Estuary.

The natural radionuclides 234Th and 7Be have known sources in estuaries and strongly and rapidly associate with particles. They are thus good tracers of the transport and sources of particles in the suspended sediment reservoir, and, by implication, of particle-associated contaminants such as heavy metals. In the Hudson River estuary, 234Th is produced from decay of dissolved 238U, which varies linearly with salinity, while 7Be is added directly to the estuary from the atmosphere. The 234Th/7Be activity ratio eliminates variations in radionuclide specific activities that result from changes in sediment grain size or composition. Sampling in the Hudson under conditions of low and high river flow shows that 234Th/7Be in suspended sediment increases from low salinity to higher salinity waters at all sampling times. The Th/Be activity ratio may thus be used as a tracer of where particles are labeled with these radionuclides in the estuary and of the relative importance of resuspension and particle transport through the estuary. The distribution of heavy metals in the suspended sediment reservoir can be compared with that of 234Th/7Be by normalizing the metals to Fe. Iron and Al are well correlated in suspended sediments of the Hudson, indicating that either element can be used to normalize grain size effects on specific concentrations of metals. Ratios of Ag and Pb to Fe in suspended particles generally increase down-estuary at all sampling times, whereas Cu, Zn and Cd-to-Fe ratios show more scatter and less consistency in trends. Both source(s) and chemical behavior affect the metal/Fe ratios. The strongest positive correlation with 234Th/7Be is seen for Ag/Fe, reflecting the dominance of lower estuary sources of this trace metal to the Hudson. Relationships between Pb/Fe, Cu/Fe and 234Th/7Be are less strong, reflecting multiple sources of these elements. No correlation is seen for Cd/Fe with 234Th/7Be, indicating either a source up-river or release of Cd as its Kd decreases with increasing salinity.

Small-scale spatial variations of natural radionuclide and trace metal distributions in sediments from the Hudson river estuary

Multiple sediment cores were collected in June 1994 in the turbidity maximum zone of the Hudson River estuary off Manhattan, New York. Results from X-radiography of the sediments and measurements of natural radionuclides (234Th,7Be, and210Pb) and trace metals (Ag, Cd, Cu, Pb, and Zn) show significant spatial variability of sediment composition and structure and patchy distributions of radionuclides activities and trace metal concentrations in this small area (0.6 km × 0.5 km). Radionuclide and trace metal analyses confirm prior work (Olsen et al. 1978; Olsen et al. 1981; Hirschberg et al. 1996) that show the western margin area of the river acts as a repository of these chemical constituents at least for the short-term period (0.5–1 yr), and the mid-channel area is not a depositional area for sediments and associated chemical constituents.7Be profiles reveal short-term sediment deposition rates ranging from 6 cm yr−1 to 26 cm yr−1 in the western margin area. Significant spatial variations in excess234Th and7Be inventories (up to a factor of 10 and 5 for234Th and7Be, respectively) are found in the western margin depositional area, although the inventories are balanced, on average, with in situ production in water column and atmospheric supply. The spatial variation of surficial excess210Pb and trace metal concentrations in depositional areas of the western margin are ≤10% for Ag, Cu, Pb, and Zn and 29% for Cd. However, the variations in the transition zone range from 28% to 93%. This variability is likely related to variations in tidal current velocity, bottom shear stress, and river channel morphology.