Search

We succeeded in determining the Ce isotopic composition (138Ce/142Ce) in seawater with an error of 2σm = 0.3–0.7 of e unit. In this study, 1000–3000 L of seawater samples were passed through MnO2 fibers to concentrate Ce and Nd for precise measurement of their isotope ratios. Four surface seawater samples of the northwestern Pacific and a coastal sample in Tokyo Bay were analyzed. Most Ce isotope ratios in the surface water showed positive eCe values (+ 0.8 to + 1.4) in the northwestern Pacific Ocean. These values indicate that Ce in the surface water originates from the continental crust preferentially over mantle-derived materials. We examined binary mixing model between the continental crust and mid-ocean ridge basalt. However the model could not explain both isotopic compositions and concentrations, which implies that the atmospheric input was a possible pathway for Ce into the ocean. A negative eCe value was observed in Tokyo Bay, suggesting mantle-derived sources.

cating that two or more factors contribute to the REE distribution. We examined four factors making the regional variation of surface REE concentrations mainly; a) particle scavenging, b) atmospheric dust input, c) vertical mixing and d) lateral transport from the coastal region. Flux calculations for Nd showed that the influence of atmospheric dust was less significant than the vertical input even in the western upwelling zone. Moreover, the longitudinal and latitudinal transitions of surface REE seem to reflect the lateral supply from the coastal areas. We constructed the diagram of surface Er/Lu and Er/Yb molar ratios in order to assess the origin and the input processes of the surface REE. Both molar ratios showed increasing trend toward PEW (Er/Lu (>10.5) and Er/Yb (>1.4)) from PSUW (Er/Lu (>7) and Er/Yb (>1.2)). The high Er/Lu and Er/Yb ratios in PEW indicate that the lateral supply of terrestrial materials from the coastal area is possibly the important factor in PEW, because only weathering and dissolution of rocks can explain such high Er/Lu and Er/Yb ratios to our knowledge.

Concentrations of Re and Os, and the isotopic composition of Os have been measured in the Japan Sea sediments to assess the response of the Japan Sea to glacial–interglacial climate change and associated weathering fluxes. The osmium concentrations in the sediment samples analyzed vary from 59 to 371 pg/g, and 187Os/188Os from 0.935 to 1.042. Only 187Os/188Os of sediment samples from dark laminations deposited under suboxic to anoxic conditions and having elevated concentrations of Re and Os, and with ≥ 80% hydrogenous Os are explained in terms of seawater composition. Lower 187Os/188Os were observed for sediments deposited during the last glacial maximum (LGM) when planktonic foraminifera from the Japan Sea recorded lighter oxygen isotopic composition. Decrease in dissolved Os fluxes from continents and/or change in the composition of the dissolved load to the Japan Sea are suggested as the driving mechanisms for the observed lower LGM 187Os/188Os. The results of this study, coupled with lower 187Os/188Os during the last glacial observed at other sites from ocean basins with different lithology and contrasting sediment accumulation rates, suggest that this trend is characteristic of the global oceans. Data from this study show that the Japan Sea recorded higher 187Os/188Os during the current interglacial coinciding with excursions of oxygen isotopic compositions of planktonic foraminifera to heavier values. This is explained in terms of preferential release of 187Os during deglacial weathering and/or higher continental Os flux driven by warm and wet climate. This study demonstrates that Os isotopic composition of reducing margin sediments has immense potential to track variations in the seawater composition. In addition, 187Os/188Os of reducing sediments may be used to draw inferences about local paleoceanographic processes in semi-enclosed basins such as the Japan Sea.

The concentration of dissolved molybdenum (Mo) was measured in the Chao Phraya River estuary, Thailand to examine the sources and geochemical behavior of Mo. Water samples in the salinity range of 0.02–24.5 collected during periods of contrasting river discharge have Mo concentrations of 3.2–117 nM. In the estuary, Mo and salinity show strong linear correlations suggestive of conservative mixing. However, at salinity>5, Mo concentrations are much higher than that can be achieved from mixing of river water and average seawater. Desorption from river-borne and estuarine sedimentary particles contribute Mo that is significantly less than the observed “excess” Mo in the Chao Phraya estuary. Groundwater discharge is unlikely to be an important source of Mo in the estuary. Reductive mobilization of Mo in the sediments of the Gulf of Thailand, its transport to the overlying water and finally to the estuary is likely to be the dominant mechanism that is responsible for the observed “excess” Mo. This inference is drawn from higher salinity and “excess” Mo in deepwater of the Gulf of Thailand, strong correlation between dissolved Mo and 228 Ra, and the observation of suboxic diagenesis in the Gulf sediments. Contribution of Mo to the estuary via porewater transport is likely to be enhanced by higher water temperature, tidal currents, sediment re-working and bottom water upwelling which are all documented in the northern Gulf of Thailand. Porewater flux of Mo, estimated using 228 Ra flux, is much higher than the average riverine Mo flux. Temporal changes in redox condition and circulation pattern in the Gulf seem to be responsible for the observed seasonal variation of Mo abundances in the estuary.

New data on vertical profiles of dissolved rare earth elements (REEs) in the southeastern Indian Ocean are presented. The compositions of dissolved REEs are uniquely characterized by strong depletion in the light REEs; the extent of depletion decreasing systematically from La to Sm, as compared to those in the western Indian Ocean and the South Pacific. Dissolved Gd is also anomalously depleted in many of the water masses occupying this region. The LREE depleted patterns with a Gd deficit appear to be derived by input from volcanic sources around the Indonesian Archipelago, although the exact location and mechanism are not yet well understood. Nevertheless, the unique REE patterns can be used to trace the pathways of the water masses in the eastern Indian Ocean. The salinity minimum centered at ∼1100 m at 40°S in the Antarctic Circumpolar region, southwest of Australia, can be considered as a boundary between the upper and the deeper layers, since the REE patterns around that depth are clearly different from those of the Antarctic Intermediate Water (AAIW) at 30°S on the Perth Basin and can be produced largely by mixing of the two waters. The unique REE patterns of AAIW with no significant Gd deficit between 497 and 2474 m in the Perth Basin indicate that the waters must be derived from somewhere else by lateral transport. The upper portion of the intermediate waters may be derived from a near-surface region in the Subantarctic zone, including the southernmost station in this study (40°S), and the deeper part may be contributed from the South Pacific reaching the area from south of Tasmania. The northward flow of the intermediate waters can be traced by the characteristic REE patterns at least to 10°S in the southeastern Indian Ocean, but does not reach the Bay of Bengal. The surface waters in the North Australia Basin have REE patterns similar to those of the surface waters in the western equatorial Pacific, suggesting a strong influence of Pacific waters to the region north of Australia. The light REE and Gd-deficient patterns are also found in the deep and bottom waters of the southeastern basins as well as in the surface waters from subtropical to temperate offshore regions, and they are presumably due to the influence of the Indonesian Archipelago. The unique pattern is clearly traced to the station at 40°S, 110°E, in the South Australia Basin, suggesting a southward flow of deep and bottom waters along the eastern edge of the Indian Ocean.

To carry out comparative geochemical investigation of refractory and reactive metals in different oceanic settings covering different θ-S characteristics, productivity, dissolved oxygen profiles, water and sediment discharge, etc., we have determined the vertical profiles of dissolved (Ce>In, depending upon particle reactivity. Although dissolved Al decreases along the water trajectory by particle scavenging, variations of dissolved In and Ce are relatively small which may be due to less scavenging for both elements. Compared with significantly high (>4 pM) dissolved Ce throughout the water column in the Bay of Bengal, dissolved Al concentration remains low, suggesting that it has higher affinity to particles and hence is scavenged by sinking particulate matter. This is consistent with the observation that the dissolved Al in the Antarctic Intermediate Water (AAIW) decreases from 4 to 6 nM in the 30°S Perth Basin to

Four vertical profiles of the concentration and isotopic composition of Nd in seawater were obtained in the western North Pacific. Two profiles from the Kuroshio Current regime showed congruently that although the Nd concentration increases gradually with depth, its isotopic composition varies significantly with depth depending upon the water mass occupying the water column. The high-salinity Kuroshio waters originating from the North Pacific Tropical Water (NPTW) carry the least radiogenic Nd (ϵ Nd = −7.4 to −8.7) to this region at ∼250 m from the western margin continental shelves, most likely from the East China Sea. The Nd isotopic compositions in the North Pacific Intermediate Water (NPIW) that occurs at 600 to 1000 m in the subtropical region are fairly uniform at ϵ Nd = −3.7. The profile data from the ∼38° to 40°N Kuroshio/Oyashio mixed water region off Sanriku of Honshu, Japan, also suggest that the newest NPIW with ϵ Nd = −3.2 is formed there by the mixing of various source waters, and the radiogenic component of Nd is derived mainly from the Oyashio waters. In the Pacific Deep Water (PDW) below ∼1000 m, the Nd isotopic composition is neither vertically nor horizontally homogeneous, suggesting that it serves as a useful tracer for sluggish deep water circulation as well. Two profiles from the Izu-Ogasawara Trench showed a minimum ϵ Nd value at ∼2000 m, suggesting that there exists a horizontal advective flow in the vicinity of Honshu, Japan. There is some evidence from other chemical properties to support this observation. The waters below 4000 m including those within the trench in the subtropical region have ϵ Nd values of around −5, suggesting that the deep waters are fed from the south along the western boundary, ultimately from the Antarctic Bottom Water (AABW) in the South Pacific. This extends up to ∼40°N along the Japanese Islands. In the subarctic region (>∼42°N), the waters have more radiogenic Nd with ϵ Nd > −4.0 throughout the water column, presumably due to the supply of Nd by weathering in such igneous provinces as the Kuril-Kamchatska-Aleutian Island chain. The lateral inhomogeneity of the Nd isotopic composition in PDW suggests that there may be different circulation and mixing regimes in the North Pacific Basin.

Nd isotopic composition and concentration distributions were determined in surface waters of the Northwest Pacific Ocean and its adjacent seas, i.e., the Okhotsk and Japan Seas. The Okhotsk and Japan Seas samples showed higher Nd concentrations (19∼31 pmol/kg) than the Northwest Pacific Ocean samples, suggesting a large amount of Nd input from the land areas surrounding the Okhotsk and Japan Seas. Although the Nd isotopic composition data were limited for those regions, each oceanic area showed a distinctive value. The lone sample from the Okhotsk Sea had radiogenic Nd (eNd = −3.6) that was comparable to samples from the Pacific Ocean, indicating a large amount of radiogenic Nd supply from the Kuril Islands. On the other hand, the Japan Sea samples had the most highly unradiogenic Nd (eNd = −8.9∼−7.2). The southernmost sample had the most negative value, due to the influence of the Tsushima Warm Current (TWC) having a low eNd value. The latitudinal distributions of Nd isotopic composition and concentration in the surface waters of the Northwest Pacific Ocean showed marked variations. Nd concentration was low at low latitudes (3∼8 pmol/kg) and was increased with increasing latitude (16∼20 pmol/kg). On the other hand, Nd isotopic composition showed a mid-latitude minimum (eNd = ∼−6) and high eNd values at low and high latitudes (eNd = −2∼−1). The distribution of the eNd values seemed to be controlled by three currents: Kuroshio Extension, Oyashio Current and North Equatorial Current (NEC). The Kuroshio Extension, which showed a low eNd value, seemed to play an important role in transporting continentally derived Nd to the central Pacific. The Oyashio Current and NEC are supplied with large amounts of radiogenic Nd from the Kuril and Aleutian Islands, and the Hawaiian Islands, respectively. Based on the Nd isotopic composition and concentration distributions, we calculated the radiogenic Nd flux to the Pacific Ocean surface waters and estimated the Nd residence time. Our calculation suggested that the radiogenic Nd flux required to account for the Nd isotopic composition is higher than 70% of the unradiogenic Nd flux including Nd remobilization from coastal and shelf sediments and atmospheric input. It was also revealed that a large Nd remobilization flux is necessary to attain a global Nd residence time of ∼400 y as estimated in previous studies.

Detailed vertical profiles of dissolved Ag in the Bering Sea, the central North and South Pacific and the Southern Ocean have been present here. These profiles are comparable to our previous report in the western North Pacific, the Okhotsk Sea and the Japan Sea (Zhang et al., 2001). There is a systematic enrichment of Ag concentrations in the deep waters (North Atlantic < South Atlantic < South Pacific < North Pacific), which is parallel with the distributions of dissolved Si and consistent with the route of the global ocean circulation. In the surface waters, the latitudinal distributions of both Ag and Si showed higher concentrations at higher latitude, suggesting that Ag in the oceanic surface waters are supplied from the deep water by upwelling and vertical mixing similar to that of Si. However, the Ag/Si ratios in surface waters are significantly higher than those in deep waters. Neither anthropogenic nor natural input from the atmosphere can play an important role for the elevated Ag/Si ratios in the surface waters. It is assumed that the elevated Ag/Si ratios in the surface waters are due to the difference in the biological uptake between the two elements.

Dissolved yttrium and rare earth elements (YREEs) have been measured in the Antarctic Circumpolar Current Region, southwest of Australia. Their vertical features are fairly smooth, irrespective of the different water masses in the water column, suggesting that the biogeochemical processes including reversible scavenging reactions and sinking particulate matter are important control on their distribution. Comparison to the previous data at nearly the same latitude of ~40 °S in the Southern Ocean of the Atlantic sector (German et al., 1995) indicates that our concentrations are significantly depleted in the lighter REEs. The difference in their mean concentrations systematically decreases with increasing atomic number, i.e., 67% for La, 40% for Nd, 15% for Sm and less than 5% for the REEs heavier than Tb and Y. Gadolinium is an exception to this trend with a deficit that is anomalously high at ~36% compared to those of its neighbors, Eu and Tb. Negative Gd-anomalies exceeding 30% are recorded for the subsurface waters, which contrast with positive Gd anomalies of up to ~20% in the North Pacific Deep Water (Alibo and Nozaki, 1999). The relative deficit of 62% for Ce(IV), perhaps fortuitously, fits well between those of La and Nd, despite its vertical profile being different from those of strictly tri-valent YREE(III)s. These observations cannot be explained simply by the REE fractionation during scavenging of particulate matter and/or regional variation of scavenging intensity alone, and suggest that there must be a REE compositional difference in the external sources, e.g., from shallow water sediments and lithogenic materials sinking through the water column. The unique REE(III) patterns of the Circumpolar Deep Water (CDW) and the Antarctic Bottom Water (AABW) with a LREE depletion in the southeastern Indian Ocean appear to be originated from igneous rocks around the Indonesian Archipelago. This is in contrast to the REE(III) patterns of a LREE enrichment with a maximum at Nd-Sm, reflecting continental sedimentary rocks, in the CDW and AABW in the South Atlantic and the western Indian Ocean. The markedly different REE patterns in the eastern and western components of CDW and AABW indicate that they can serve as novel tracers in elucidating the deep water circulation of the Indian Ocean.

Vertical profiles of dissolved rare earth elements (REEs) were obtained in the Bay of Bengal and the Andaman Sea. The REE concentrations at various depths in the Bay of Bengal are the highest in the Indian Ocean. This is attributable ultimately to the large outflow of the Ganges–Brahmaputra and Irrawaddy rivers, but the dissolved REE flux to surface waters alone cannot explain the large and near-constant REE enrichment throughout the entire water column. The underlying fan sediments serve as not a source but a sink for dissolved REE(III)s. Absence of excess 228Ra in the deep waters suggests that lateral input of dissolved REEs from slope sediments is also small in these regions. Partial (

This paper presents the first in situ flow-through chemical analyzer using a chemiluminescence (CL) method in the deep sea to a depth of 5200 m. The analyzer, called GAMOS (Geochemical Anomalies MOnitoring System), determines the concentration of dissolved manganese continuously using a H 2 O 2 –luminol CL method. A detection limit of 0.23 nM was obtained. The GAMOS was used successfully for hydrothermal plume observation.

The activities of 228Ra and 226Ra in the seawaters of the eastern Indian Ocean and its adjacent seas were measured during the 1996/1997 “Piscis Austrinus” Expedition using R/V Hakuho-Maru. The surface Ra data are in good agreement with those in literature and indicate that the continental shelves of the southeast Asian Seas are strong sources for 228Ra and, to much lesser extent, for 226Ra which are largely transported by the surface currents into the Indian Ocean. The vertical profiles of 228Ra obtained at six stations from the different oceanic regimes along the cruise track generally showed an almost exponential decrease from the surface to the middepths of 1200- 1500 m. Coupling the vertical profiles of 228Ra and nitrate and using the vertical one-dimensional model developed by Ku et al., [1995], we estimated the nitrate fluxes into the euphotic zone ranging from 1.5 mmol N m−2 d−1 (or 0.55 mol m−2 yr−1) in Andaman Sea to 4.0 mmol N m−2 d−1 (1.5 mol m−2 yr−1) in the Southern Ocean. In steady state, these nitrate fluxes correspond to the new production of 11–26 mmol C m−2 d−1 (or 3.9–9.6 mol C m−2 yr−1). In the sub-Antarctic region between 30° and 40°S, north of the polar front, dissolved Si is depleted relative to nitrate in the upper water column as compared to their ∼1:1 molar ratio demanded by diatoms. In such high-nitrate but low-silicate regions, growth of diatoms is likely to be limited by available silicate. It also favors calcareous coccoliths to grow by utilization of nitrate, as deduced from a decrease in alkalinity in the surface waters. During the last glacial period, supply of atmospheric dust containing silicon and iron to the surface ocean was much more intense than that during the interglacial period, and hence, diatom production might have been enhanced in the Si-depleted regions. Diatoms would utilize nitrate readily and inhibit growth of coccoliths, resulting in increases in alkalinity and pH in the surface water. This species change of phytoplankton may account for the lowered atmospheric CO2 level during the glacial period without causing a large-scale depletion of dissolved oxygen in the deep sea. The records of opal and carbonate contents in the Antarctic sediment cores studied by Charles et al. [1991] seem to be consistent with this silicon- induced alkalinity pump hypothesis.

In this study, a control algorithm of an autonomous vehicle is proposed on the basis of risk level to simulate control motion of a real driver. The normal traffic situation can be expressed by risk level. The risk level is affected by several risk elements: roadside edges, curves, the other vehicles, obstacles, and so on. Each risk element is represented by an exponential function. The risk elements make risk potential field on the road. It is assumed that the desirable course to follow is determined as the point of minimum risk potential in the cross section of the road. Tree prediction models are examined to predict the future position of vehicle. The change of preview time is considered on the curved road. A lateral and longitudinal control algorithm with the prediction model proposed in this study shows similar control motion to that of a real driver.

New data on dissolved Ba, 226 Ra and 228 Ra were obtained in the Chao Phraya River estuarine mixing zone in 1996. Although desorption of Ba from river-suspended sediments gives rise to elevated dissolved Ba concentrations in the low salinity ( S S >10) region have been observed depending upon the sampling periods. Under a dinofragellates ( Noctiluca ) red tide condition in January, significant biological removal of dissolved Ba took place in the mid-salinity ( S= 10 – 16) region, whereas under normal condition in July and November, the dissolved Ba followed a conservative linear mixing trend. 226 Ra is also supplied to the coastal waters by desorption from river-transported sediments in a delayed fashion compared to Ba, at least, in the latter seasons, whereas 228 Ra is predominantly supplied by diffusion from underlying sediments to the water largely in the higher salinity ( S >15) region. These geochemical observations in July and November fit well to the previous studies done for other estuarine environments. Total riverine fluxes to the Gulf of Thailand were estimated to be 6.2×10 6 mol Ba/yr and 1.5×10 13 dpm 226 Ra/yr, about half of which is contributed by the Chao Phraya River. Although there is a sign indicating that ground water input may be contributing to the fluxes of 226 Ra and Ba, we were unable to quantify them due to lack of information on the exchange rate of the coastal waters with open seawater.

We present here the detailed vertical profiles of dissolved ( Our dissolved Ag concentrations (4.2–8.0 pmol/kg) in the surface waters are significantly higher than those reported previously for other locations, and it appears that the variability of Ag in the surface waters is greater than an order of magnitude. Although many factors are involved in controlling the Ag concentration in the surface waters, removal by biological uptake appears to be particularly important. Although there may be some indication of anthropogenic sources in the surface waters, it seems to be local, and the large-scale contamination of Ag in the ocean is not seen in our data.

Although the rare earth element (REE) geochemistry of hydrothermal systems have intensively been investigated at mid-oceanic ridges, studies are relatively few at the converging plate zones like the Okinawa Trough. The “chimney” deposits obtained by using submersible “SHINKAI 2000” from the Iheya Ridge site, Okinawa Trough are largely composed of massive carbonates and silicates with minor amounts of sulfides, and appear black and white depending upon their composition. The concentrations of REEs are 2–3 orders of magnitude higher in the sulfide-rich black portion than in the white portion. Both deposits show a pronounced positive Eu anomaly and the light REE enrichment which are characteristic of the high-temperature fluids venting from sediment-free oceanic ridges. Whilst the Okinawa Trough REE data are unique in the hydrothermal vent systems in having a flat chondrite-normalized pattern for the REEs heavier than Tb. The Eu-anomaly is greater in the white deposit than in the black deposit, suggesting that Eu2+ is more mobile (resistant to scavenging by sulfides) than stricktly trivalent REEs in the hydrothermal solution and may be incorporated into the former by substitution of Sr2+ in the calcite lattice. The Calyptogena shell shows the REE pattern typical for seawater without any positive Eu anomaly, suggesting that the influence of hydrothermal solution is negligible. Although the light REE concentrations are comparable, the heavy REEs are significantly higher in the shell than in the white deposit. This suggests that the concentrations of the heavy REEs in the carbonate-forming hydrothermal solution may be lower than those of ambient seawater. Thus, the hydrothermal system in the Okinawa Trough may not be a source but a sink for dissolved REEs in seawater.

We have measured the vertical profiles of dissolved rare earth elements (REEs) and yttrium in the South China Sea together with conductivity-temperature-depth and hydrographic measurements to compare with those in the western North Pacific and the SuIu Sea. Although the South China Sea is rapidly flushed by the Pacific through the Luzon Strait with a sill depth of ∼2500 m [Broecker et al., 1986], a unique REE pattern is developed within the sea. The most striking difference exists in the dissolved Ce profiles. Dissolved Ce generally decreases from high values (6–9 pmol/kg) at the surface to a minimum of ∼3 pmol/kg at around 300–500 m where the North Pacific Intermediate Water penetrates. In deepwaters of the North Pacific and the Sulu Sea it remains at a relatively low and nearly constant concentration level of ∼5 pmol/kg throughout the water column, whereas in the South China Sea, it gradually increases with depth to a maximum of 12.9 pmol/kg at ∼2500 m, resembling the “nutrient-like” profiles of other strictly trivalent REEs, and then sharply drops to a constant value of ∼6 pmol/kg in the bottom water below 2900 m. Some lighter REEs like Pr, Nd, and Gd, though to a much lesser extent, also show similar concentration breaks at the sill depth, but the other hydrographic properties like dissolved oxygen, nutrients, pH, and alkalinity do not. Therefore dissolved REEs may best be utilized to characterize the water masses. Two major sources for dissolved REEs in the South China Sea are fluvial and coastal input to the surface ocean and a bottom release into the deep water during the passage over the Luzon Strait. Redox chemistry including reduction of Ce(IV) to Ce(III) in the pore water of hemipelagic sediments and subsequent release of dissolved Ce(III) to the overlying deep water may be involved in the latter. The middle REE-enriched patterns with a significant Gd depression relative to that of the North Pacific Deep Water are characteristic of the South China Sea and prevail throughout the water column by physical circulation. Since the bottom water of the basin is also fed by the North Pacific, the dissolved Ce(III) must be oxidized through bacterial mediation and removed from the bottom water presumably by scavenging near/at the sediment-water interface along the slopes of the basin.

New data on the dissolved ( (Nozaki et al., in press) . Like Gd, the high dissolved In in the study area can also be ascribed to recent use of In-containing organic compound, In(DTPA)2− in medical diagnosis. Thus, in the highly populated and industrialized area, dissolved heavy metal concentrations in rivers and estuaries may be significantly perturbed by human activities and the fate of those anthropogenic soluble substances in the marine environment needs to be investigated further.

The Nd isotopic composition and dissolved rare earth elements (REEs) have been measured in the surface waters along the 1996/97 R.V. Hakuho-Maru Expedition route from Tokyo to the Southern Ocean, southwest of Australia, through the Philippine and Indonesian Archipelago, the eastern Indian Ocean, the Bay of Bengal and the South China Sea. The radiogenic eNd values of −1.3 and −1.4 were found in the Sulu Sea and near the Lombok Strait, indicating the strong influence of surrounding volcanic islands, whereas non-radiogenic eNd values of less than −10 were found in the Southern Ocean and the Bay of Bengal suggesting Nd of continental origin. The dissolved Nd concentrations also showed a wide range of variation from 2.8 to 19.6 pmol/kg and the trivalent REE patterns exhibited characteristic features that can be grouped into each different oceanic province. The geographical distribution of dissolved Nd is different from that of atmospherically derived 210Pb, but generally resembles that of coastally derived 228Ra. This strongly suggests that fluvial and coastal input predominates over eolian input for dissolved Nd in the surface ocean. However, the riverine dissolved Nd flux appears to be relatively minor, and remobilization of Nd from coastal and shelf sediments may play an important role in the total Nd input to the ocean. By modeling the distributions of the isotopic composition and concentration of Nd together with the activity ratio of 228Ra/226Ra in the southeastern Indian Ocean, we estimate a mean residence time of Nd in the surface mixed layer to be 1.5–2.6 years. The short mean residence time is comparable with, or slightly longer than that of 210Pb suggesting similar chemical reactivity.

A flow-through analytical method has been developed for the determination of trace amounts of aluminum in seawater with the use of Lumogallion fluorometric detection. Because iron was selectively removed by passing the sample through an 8-quinolinol immobilized chelating resin column (MAF-8HQ), this method could be applied to samples collected at the hydrothermal oceanic regime. Several known organic complexing agents were studied as a model group to examine the effect of naturally occurring organic ligands on the detection of aluminum in seawater. The detection limit was 0.17 nM for 10 ml of sample. The standard deviation was 2.7% at 2 nM of aluminum. This method permitted the sensitive, precise, and rapid determination of subnanomolar Al in seawater without contamination onboard ship. © 2000 John Wiley & Sons, Inc. Field Analyt Chem Technol 4: 274–282, 2000

Experiments have been carried out to clarify the effect of alteration on Re–Os system, near infrared (NIR)–infrared (IR) spectroscopic characteristics and polytype of a natural molybdenite mineral (MoS2). The molybdenite sample was placed in H2O and various media of 0.1 mol/L NaCl, NaHCO3, CaCl2, and AlCl3 solutions, and heated in a sealed quartz tube at a temperature of 180°C for 20 d. The unaltered and altered samples were subsequently used for analysis of Re and Os, NIR microscopic observation, and NIR–IR spectroscopy, and microfocus X-ray diffraction (XRD). Molybdenites subjected to NaCl and NaHCO3 solutions give younger Re–Os ages than that of the original unaltered molybdenite. No significant changes in d spacing and width of micro-XRD patterns can be found in these altered molybdenite, indicating the possibility of Re–Os fractionation without significant structural conversion of molybdenite mineral. These results strongly suggest that the Re–Os system in molybdenite would be frequently disturbed if it has experienced alteration, because alteration by the low salinity (

Because most Southeast Asian basins are flushed rapidly by waters from the western Pacific, the effects of respiration and silica dissolution within those basins are hardly discernible based on general hydrographic and geochemical observations (Broecker et al., 1987). However, a different situation is expected for the Sulu Sea because its deep water is isolated from other deep sea basins by shallow sills of ,400 m depth. Hence, any heterogeneity of elemental distributions within the basin must be ascribed to vertical physical and biogeochemical processes. We have obtained detailed vertical profiles of dissolved rare earth elements (REEs) together with conductivity-temperature-depth (CTD) and hydrographic measurements in the Sulu Sea during the 1996 -97 Hakuho-Maru cruise. Vertical profiles clearly indicate that the REE(III)s are enriched in the deep water due to regeneration and are involved in vertical biogeochemical cycling. The light REE(III)s (La, Pr, and Nd) showed a subsurface maximum at 500 m and a minimum;1500 m in depth. None of the hydrographic properties or nutrients shows such features; therefore, the REE(III)s and their elemental ratios may be useful as tracers of water masses. The relationship of REE(III)s versus dissolved Si suggests that the REE(III)s are regenerated in a delayed fashion relative to Si, and it is likely that their dissolution largely occurs at the bottom interface. The mean residence time of the deep water is estimated to be 300 6 150 yr. Consequently, the calculated benthic flux of REE(III)s becomes roughly comparable with the river and atmospheric fluxes. Because of oxidation to Ce(IV), the Ce profile shows a decrease from ;6 pmol/kg at the surface to a constant value at 3.9 pmol/kg below 1500 m, unlike the other REEs. The REEs in the Sulu Sea are enriched in the middle REEs and Ce relative to those of the North Pacific Deep Water and must derive from local sources around the Sulu Sea. Copyright © 1999 Elsevier Science Ltd

Vertical profiles of dissolved indium and yttrium were determined in the eastern North Atlantic and the Mediterranean Sea to compare with those of the North Pacific reported earlier. The Y concentrations in the surface waters are 120 pmol/kg in the North Atlantic and 205 pmol/kg in the Mediterranean Sea, which are significantly higher than 80 pmol/kg in the North Pacific. The difference may be attributable to the different strength of input of Y to the oceans from fluvial and aeolian sources. In contrast, the deep water concentration of Y increases in the order of North Atlantic North > North Pacific exhibits an inverse trend of Y and nutrients. Indium is highly particle-reactive (47% association in the Mediterranean Sea), and must have a short mean oceanic residence time. However, the featureless dissolved In profile in the Mediterranean Sea is clearly different from the profiles of dissolved Al, showing increase with depth Hydes et al 1988 , Measures and Edmond 1988 , suggesting that significant fractionation of the two elements is taking place in the ocean. The interoceanic variations of dissolved In and Al may be ascribed to the different intensities of external input of which aeolian has been considered to be major rather than fluvial. However, the difference of In and Al concentrations in the deep waters of the above oceanic basins are significantly greater than those of other refractory elements, such as Ce, Ti, Hf, and Zr, whose major sources to the ocean are also considered to be aeolian. Furthermore, the In/Al ratios in seawater are about two orders of magnitude greater than the average crustal ratio. Thus, some additional sources, though not yet certain, may be required to explain the high concentrations of In in the Atlantic and the Mediterranean deep waters.

Dissolved ( Shale-normalized patterns of dissolved REEs were examined in detail, based on three datasets of composite shales available in the literature. Distinctively positive La and only slightly positive Gd anomalies were identified together with well-documented negative Ce-anomaly as common features of seawater. These anomalies systematically change with depth. Rapid changes occur in the upper several hundred meters suggesting that their distributions are largely governed by ocean circulation and biogeochemical cycling.

Two methods have been developed to measure indium (In) in natural waters by flow injection inductively coupled plasma mass spectrometry (ICPMS). One is the isotope dilution technique using an113In enriched spike and the other utilizes natural yttrium present in the sample as an internal standard. In the former, optimization of the113In spike to minimize error is often difficult for samples in which In concentrations are variable, whereas in the latter method, a separate determination of Y in the sample is necessary and hence more sample is required. Using about 1 liter of a water sample, 200 fold preconcentration of In was performed by solvent extraction and back extraction technique and then introduced into the ICPMS to measure the113In/115In or115In/89Y ratios. The detection limits were 0.01–0.02 pmol kg−1 for both methods.

A flow through analytical method has been developed for the determination of trace amounts of manganese in sea water using luminol–hydrogen peroxide chemiluminescence (CL) detection. The method was very simple and almost all the metal ions did not interfere with the CL detection except for iron species. The iron species were thoroughly removed by passing the sample solutions through a 8-quinolinol immobilized chelating resin column. The lower and upper limits of detection were 0.029 nM and 4 μM, respectively. The relative standard deviation was 0.58% at 10 nM of Mn(II). The method permitted the selective and sensitive determination of sub-nM levels of Mn in sea water sample with sufficient precision.

210 Po, 210 Pb, 228 Ra, and 226 Ra were measured in surface sea waters along the 1989–1990 global traverse of the oceans using the new R.V. Hakuho-Maru. Where the traverse intersects other expedition routes, the data are generally confirmatory. In the high-productivity regimes like the Red Sea, and the Arabian Sea 210 Po is removed from the mixed layer at much faster rates than 210 Pb. This fractionation occurs during scavenging presumably because 210 Po is strongly sorbed by organic particles, whereas 210 Pb is more likely associated with inorganic detritus. The 210 Po/ 210 Pb activity ratios leaving the mixed layer by particulate transport can be estimated from the steady state balance of 210 Pb and 210 Po in the surface waters for different oceanic regions, and are compared with those in the literature obtained by sediment-trap experiments. Although this comparison appears to merge, there exist some inconsistencies, which may be attributable to one of the following possibilities: (1) the model-derived atmospheric 210 Pb flux is overestimated for the North Pacific and the North Atlantic, or (2) the sediment-trap data do not represent the real 210 Po/ 210 Pb ratio in the vertical particulate flux due to some experimental artifacts, such as incomplete trapping and size fractionation. Further careful studies of sediment trapping including seasonal variation are needed to resolve this issue. Our Ra data confirmed that strong sources for 228 Ra exist in the Bay of Bengal and the Southeast Asian continental shelf zone, whereas the Mediterranean and the Red Sea, though they are surrounded by land, are not effective sources of 228 Ra in the surface water.

Settling particles were collected with time-series sediment traps at four depths (1174 m, 3680 m, 5687 m, and 8688 m) moored at the deepest site (water depth 9200 m) of the Japan Trench for the period from March 5, 1991 to March 2, 1992. The composition of rare earth elements (REEs) in the sinking particles selected for four seasons were measured to investigate the REE fractionation during scavenging and settling transport through the water column. The REE-particle associations were also investigated by sequential digestion with three chemical treatments (acetic acid, mild HCl/HNO3, and HF/HNO3/HClO4). The measured vertical particulate flux increased with depth suggesting that there is a significant addition of particles, at least, to the deeper traps from horizontal sources. No systematic variation was observed in the REE concentration in the particles with respect to depth and season. The shale-normalized REE data show either flat or slightly heavy REE-enriched patterns with a middle REE enrichment in which the maximum is usually centred at Gd, but sometimes at Eu and a significantly positive Ce anomaly. The chemical digestion experiments indicated characteristic patterns for shale-normalized REEs. The acetic acid leach showed a marked middle REE-enriched pattern with a maximum at Eu and distinctively negative Ce anomaly. The mild HCl/HNO3 leach had a middle REE-enriched pattern with a maximum at Gd but roughly no Ce anomaly. The residue showed a rather flat REE pattern with a marked positive Ce and often middle REE enrichment with a maximum at Gd. These results are compared with those reported previously for sinking and suspended particles in other parts of the oceans. It is suggested that REEs adsorbed on the suspended particles are altered into some refractory phases prior to or during transport by settling. The possibility exists that authigenic barite formed in the microenvironment around aggregates may effectively coprecipitate the REEs attracted to the particle surfaces.

Presented here is the first detailed geochemical data set on the U/Th series Th, Pa, Ac, and Pb isotopes and artificial fallout radionuclides (Sr-90, Cs-137, and Pu isotopes), and some trace elements (V, Zn, Cd, Cu, Mn, and Ni) in two water columns of the Japan and Bonin trenches down to the bottom depths of 7585 m and 9750 m, respectively. Hydrographic properties such as temperature, salinity, dissolved oxygen, and nutrient content within the trench valley remain constant at the same levels as those in the bottom water of the Northwest Pacific basin (typically similar to 6000 m in depth). The radionuclide activities and most trace metal concentrations are also not very different from those in the overlying water at depths of around 5000-6000 m. This means that any chemical alteration which sea water undergoes during its residence within the trench was not obviously detected by the techniques used here. The suggestion follows that the trench water is rather freely communicating by isopycnal mixing with the bottom water overlying the Northwest Pacific abyssal plain. The trench waters contain high Pu-239,Pu-240 activities throughout, indicating that Pu is actively regenerating from rapidly sinking, large particles at the bottom interface, probably due to a change in the oxidation state. On the other hand, the vertical profiles of Pb-210 and Pa-231 show lower activities within the trench than those in the overlying deep waters, suggesting that the effect of boundary and bottom scavenging is significant in controlling their oceanic distributions. However, none of the trace metals studied here obviously follows the behaviour of the above nuclides. The Th-228 data show scattering within the Bonin Trench that is largely ascribable to analytical errors. If, however we accept that the scatter of Th-228 data is real and the variation is caused solely by decay of its parent Ra-228, We,,, set an upper limit of similar to 5 years for the renewal time of the trench water., La première série détaillée de données géochimiques obtenues dans deux colonnes d'eau des fosses du Japon et des îles Bonin, respectivement jusqu'à 7585 m et 9750 m, est présentée ici. Elle porte sur la série uranium/thorium, sur les isotopes de thorium, protactinium, actinium et plomb et sur les retombées des nucléides radioactifs artificiels (strontium-90, césium-137 et plutonium) ainsi que sur quelques éléments à l'état de traces (vanadium, zinc, cadmium, cuivre, manganèse et nickel). Les caractéristiques hydrologiques telles que température, salinité, oxygène dissous et teneurs en nutriments prennent les mêmes valeurs aux mêmes profondeurs, dans les fosses comme dans l'eau de fond du bassin nord-occidental du Pacifique (typiquement, vers 6000 m). De même, les valeurs de l'activité des nucléides radioactifs et les concentrations de la plupart des traces métalliques ne sont pas très différentes dans la fosse et dans l'eau sus-jacente, vers 5000–6000 m. Si, au cours de son transit dans la fosse, l'eau a subi une modification chimique, celle-ci n'a pas été détectée par les techniques utilisées. Cela suggère que l'eau de la fosse communique assez librement par mélange isopycne avec l'eau de fond qui couvre la plaine abyssale dans le nord-ouest du Pacifique. Partout dans les eaux de la fosse l'activité du 239, 240Pu est élevée, ce qui indique que le plutonium est activement régénéré à partir des grosses particules qui sédimentent rapidement sur le fond, probablement par un changement de leur état d'oxydation. De plus les profils verticaux de 210Pb et 231Pa indiquent des activités plus faibles dans la fosse que dans les eaux profondes sus-jacentes, suggérant que l'effet de barrière et le balayage sur le fond sont significatifs dans la répartition de ces éléments dans l'océan. Aucun des éléments-traces étudiés ici ne se comporte comme les nucléides radioactifs. Dans la fosse des îles Bonin, les données de 228Th montrent une dispersion qui est due essentiellement aux erreurs analytiques. Cependant, si la dispersion est réelle et résulte uniquement de la désintégration du 228Ra, on peut fixer une limite supérieure d'environ cinq ans pour le temps de renouvellement de l'eau dans la fosse.

Using a flow injection ICP-MS method, yttrium and all the lanthanides in seawater were determined in the coastal/offshore mixing regime near Japan. Marked enrichments in the trivalent rare earth elements (REEs) in the surface waters are ascribed to the influence of coastal waters from Tokyo Bay in March and perhaps somewhere else in September and October to this region. At less than half of the stations where the detailed vertical profiles were obtained throughout the water column, we found marked excess REE(III) concentrations near the bottom. These bottom concentration anomalies are probably caused by resuspension of underlying sediments. Excluding those surface and bottom anomalous concentrations, the distribution of REE(III)s in the midwater column are very similar to those of the open ocean and are highly correlated with each other (R2 > 0.95). This suggests that the boundary effects on the chemical fractionation of REE(III)s are small. We find no effect of the boundary scavenging known for 210Pb and 231Pa on the REE distribution, whereas there may be preferential release of light and middle REE(III)s over heavy REEs from slope sediments to the seawater much like 227Ac. From the shale-normalized REE patterns, we obtained up to 90% negative Ce and 10% positive Gd anomalies in the water column.

I have currently worked on the chemistry and geochemistry of the ocean with special focus on seawater for which the Prize of the Oceanographic Society of Japan was awarded in 1998. Coincidentally, during the last few decades, my work is closely followed by the world-wide development of new analytical and sampling techniques that enabled us to establish the oceanic distributions of almost all the elements existing in seawater. It seems that a long desire of world chemical oceanographers is now going to be achieved. Nevertheless, kinetics that govern the present-day elemental composition and distribution are not well understood as yet, because fluxes to and out of the ocean are poorly quantified for most elements. Exceptions are natural radionuclides that have served as useful tracers to determine how the elements behave in the ocean (Nozaki, 1991). By coupling knowledge from the natural radionuclides and those of elements with similar chemical reactivities, for example, 227Ac versus lanthanide, the dynamic aspects of geochemical cycling in the ocean will be better elucidated. This is an aspect described in this article.

New measurements of Y and Ho in seawater, rivers and rain are presented. Based on the data and a two-box model calculation, we suggest that fractionation between Y and Ho takes place during their removal by particulate matter from the surface ocean to the deep sea. The fractionation factor, K D is calculated to be 0.50, implying that Ho is scavenged two times faster than Y. This presumably occurs due to differences between Y and Ho complexation behavior with respect to seawater inorganic ligands (mainly carbonate ions) and soft organic ligands (though unspecified) of the surface of particulate matter. Fractionation of Y and Ho during weathering and fluvial transport to the ocean appears to have minor influence on the relative abundance of Y and Ho in seawater. We also estimated the mean oceanic residence time to be 5100 years for Y and 2700 years for Ho. Y is less effectively scavenged from seawater than any of the trivalent rare earth elements and theY/Ho ratio in seawater is higher than those in rain, rivers and estuarine waters.

New data are presented for210Po and210Pb concentrations in seawater of the equatorial Pacific and the Bering Sea. Their implications are discussed in terms of productivity and boundary scavenging. The210Pb concentrations in the equatorial waters are high at the surface (0.10.15 d.p.m. 1−1, approximately twice the226Ra activity) with significant210Po deficits in the upper 1000 m water column ranging from 2.6 to 4.3 d.p.m. cm−2. Clearly210Po is preferentially removed relative to210Pb from the surface waters. Along the Tokyo-San Diego-Panama transect in the midNorth Pacific, the removal rate of210Po is correlated with the chlorophyll-a content in the surface waters, suggesting that scavenging of210Po to particles is virtually controlled by phytoplankton in the open ocean. In the deep water,210 Pb is always deficient relative to226Ra. Box-model210Pb residence times, based on mean 210Pb/226Ra activity ratios, are approximately 90 years at the open-ocean equatorial sites. These are significantly shorter than those of the North Pacific gyre (greater than 200 years). The short residence times can be ascribed to the intensified scavenging of210Pb because of the high particle flux regime of the equatorial zone. The deep-sea210Pb residence time becomes significantly shorter as the western topographic boundary is approached, and is only 8 years in the Bismarck Sea. This tendency cannot be ascribed to the difference in the particle flux in the equatorial zone. It seems that210Pb is actively taken up from the deep water by sediments at or near the bottom interface presumably in association with manganese redox cycling. The Bering Sea data support this mechanism. Compilation of deep-sea210Pb residence times available in the literature and the new data presented here suggests that the lateral transport of210Pb via isopycnal mixing followed by scavenging at the sediment-water interface is a major control on the210Pb-226Ra disequilibrium in the deep sea. The surface ocean productivity may be of secondary importance in this context.

Using inductively coupled plasma mass spectrometry (ICP-MS), vertical profiles of yttrium and all the rare earth elements (REEs) in seawater were obtained at three locations in the western South Pacific (the East Caroline, Coral Sea, and South Fiji Basins). Based on the data, we have analyzed the inter-REE(III) relationships and found that REE(III)s heavier than Dy and particularly close neighbors show strong coherencies in their oceanic behavior. Considering that the heavy REE(III) s are less particle-reactive than the light and middle REE (III) s and that they have very tight correlations with regressions passing almost through the origin, the neighboring element ratios of the heavy REE(III)s behave virtually conservatively and are suitable as tracers of water masses. The Dy-Ho-Er system is practically good, because its dynamic range (signal-to-noise ratio) is large in the ocean. The highest Ho/Dy ratios (∼0.31) are found in intermediate and deep waters throughout the western South Pacific which overlie the Antarctic Bottom Water with a low Ho/Dy ratio (∼0.27). These high Ho/Dy waters are probably formed in the Polar and Subantarctic Frontal Zone and advect northward. REE(III) data provide better insights into the deep water recirculation in the South Pacific than those discussed based upon the regular oceanographic properties alone.

New and precise data on the 138 Ce 142 Ce ratio in marine manganese nodules obtained from the Atlantic and Pacific Oceans fall within a narrow range of −0.6 to + 1.1 in terms of ϵCe (one tenth of per mil difference from the standard). The nodule Ce isotopic data are used as a surrogate of surrounding seawater, and the ϵCe, values ranging from −0.6 to −0.2 for the central Pacific are clearly more negative than those (+0.2 to + 1.1) for the Atlantic. By coupling the ϵCe with ϵNd for 143Nd/144Nd, it is inferred that the relative contribution of the continental-derived to mantle-derived sources is smaller in the Pacific than in the Atlantic. If we assume that the oceanic 138 Ce 142 Ce and 143 Nd 144 Nd ratios are determined by mixing of two end-members, i.e. the continental crust and MORB (mid-ocean ridge basalt), the oceanic variation of ϵCe is greater than that of ϵNd relative to their difference in the two end-members. This strongly suggests that Ce and Nd are fractionated within the ocean because of the difference in their mean residence times (90–165 yr for Ce and 1000–1500 yr for Nd) and the 138 Ce 142 Ce and 143 Nd 144 Nd ratios modified by scavenging and oceanic mixing are recorded in the manganese nodules.

The concentration of indium in the seawaters of the western North Pacific was measured by using isotope dilution inductively coupled plasma mass spectrometry (ID-ICPMS). The data suggest a surface maximum and the profiles show a possible decrease with depth. The measured concentrations of indium range from 0.047 to 0.155 pmol/kg and are almost three orders of magnitudes lower than those reported for the North Pacific by Chow and Snyder [1969]. We believe that the earlier data reflect experimental artifacts. Our data are also lower by about one order of magnitude than those in the North Atlantic reported by Matthews and Riley [1970]. The possibility exists that their data also suffered from contamination problems. It is also likely that the concentration of indium in the Atlantic is indeed significantly higher than those in the Pacific due to the difference in the atmospheric input if indium behaves like A1. Redetermination of the Atlantic samples is essential to resolve the two possibilities.

Yttrium has long been recognized as an ekalanthanide, because of its chemical contiguity relative to the chemistries of rare earth elements which are, in recent years, intensively utilized for elucidation of metal scavenging processes in the ocean. Here, we present the first detailed depth-profile of Y(III) in seawater together with the lanthanides in the North Pacific Ocean. The concentrations of Y(III) range 70–290 pmol/kg and show the “nutrient-like” profile best resembling that of Ho(III) amongst the other rare earth elements. The results agree well with an expectation based on the similarity in their ionic radii and hence stability constants of complexation with carbonate ions. Yet the Ho(III)/Y(III;) ratios in seawater systematically increase with depth, suggesting that Y and Ho are fractionated during scavenging by natural marine particulates. This is likely to result from the different complexation behavior in that Y(III) is more weakly complexed than Ho(III) with soft organic ligands on the surface of particulate matter during scavenging in the surface water but, once released into seawater in the deep sea, Y(III) is complexed with carbonate ions equally or stronger than Ho. The pattern of deep water enrichment in the lanthanide series appears to be consistent with the recent observation of partitioning between suspended particles and seawater. Our precise measurements also indicated that Pr and Tb best resemble Nd and Dy, respectively in their oceanic behavior, whereas Ho and Tm are intermediate between their neighboring rare earth elements.

Two sediment cores (pilot gravity and piston) were obtained from the bottom of the Izu-Ogasawara Trench at 9750 m and analyzed for various elements and radioisotopes. The results showed a history of complex and frequent turbidite deposition: In the gravity core, eight layers rich in manganese were observed, of which five are enriched in Cu and Co as well. The other three are also enriched in Mo but no other heavy metals, suggesting the presence of at least two mechanisms of formation. Trapping of iron manganese micronodules can account for the enrichment of Mn, Cu and Co. The other three layers rich in Mn and Mo appear to be formed by a post-depositional diagenetic process of Mn mobilization and redeposition in the sediment column. A strong correlation between 226 Ra and Cu in the gravity core suggests that the 226 Ra was also carried into the bottom of the trench in turbidites in association with Mn micronodules. Little excess of 210 Pb over 226 Ra was found at the top but the excess was significant at mid-depths from 30 to 70 cm, indicating that those sediments were deposited within the last 200 y. In the piston core there is a sharp discontinuity of chemical and radiochemical composition around a depth of 250 cm. Below that depth the sediments appear to be dominated by materials derived from terrestrial sources, as compared with those in the upper layer which are of contemporary marine origin. 226 Ra is deficient relative to 230 Th throughout the sediment column down to ∼ 6 m. This finding is consistent with the finding that the average rate of sediment accumulation is 1–2 orders of magnitude faster than that in the western North Pacific abyssal plain, suggesting the convergence of materials into the bottom of the trench.

The features of 228 Ra and 226 Ra distribution covering the salinity range from less than 0·1% to 35% in the Ota River estuary and extending to the Seto Inland Sea have been presented. Using this data and a mass balance of radium in this region, the source of radium from the river-estuarine sediment is shown to be more important than that from the sediments themselves of the Seto Inland Sea. The flux of Ra from the sediments is 0·84 dpm 228 Ra cm -2 year -1 and 0·08 dpm 226 Ra cm -2 year -1 about 17 times higher than those from river for 228 Ra and 8 times for 226 Ra. By applying the One-Dimensional Advection-Diffusion model and assuming a steady state of radium in the Seto Inland Sea, the appropriate value of the horizontal eddy diffusion coefficient is in the range 1·1 × 10 6 - 1·4 × 106 cm 2 s -1 .

An idea has been proposed to recover and dispose of the fossil-fuel CO2 generated by stationary power plants into the deep sea, thereby reducing the CO2 emission to the atmosphere. The assessment of its environmental effects and risks is essential for developing new technology for the recovery/disposal system, which has recently received growing attention in Japan.

228Ra, 226Ra, 210Pb, and 210Po were measured in the surface waters of the East China and Yellow seas. Using mass balance equations for the Ra isotopes, we estimated the total flux of diffusion from sediments and desorption from suspended particles to be 0.1 dpm 226Ra cm−2 a−1 and 1 dpm 228Ra cm−2 a−1, and residence times to be 2–3 years for the waters on the East China Sea Shelf and 5–6 years for Yellow Sea waters. Box-model calculations yielded generally congruent scavenging residence times for 210Pb and 210Po in the waters of ~2 months on the shelf and ~7 months in the Kuroshio Current. These suggest that reactive heavy metals and pollutants discharged through rivers from the continent to the East Asian continental shelf are largely deposited on the bottom sediments prior to transport to the pelagic ocean by lateral mixing.

{sup 227}Ac with a half life of 21.8 years has a potential utility as a tracer of deep water circulation and mixing studies on time scales less than 100 years. Here the authors present the first measurement of {sup 227}Ac profile in the pore water of Northwest Pacific deep-sea sediment and in the {approximately}10,000 m long water column of Izu-Ogasawara Trench. The results clearly show that {sup 227}Ac is supplied from the sediment to the overlying water through migration in the pore water. The model calculation indicates that the molecular diffusion alone through sediment porewater can support only a half of the standing crop of excess {sup 227}Ac in the water column and the enhanced supply of {sup 227}Ac by particle mixing is necessary to account for the remainder. Thus, bioturbation in the deep sea plays an important role in controlling the flux of some short-lived radionuclides such as {sup 227}Ac and {sup 228}Ra across the sediment-water interface.

The isotopic composition of Ce in seawater has been measured for the first time, together with that of Nd. In spite of rather great uncertainties of measurement, we can say that the extent of 138Ce/142Ce variation appears considerably large, and that the ratio can be used as a geochemical indicator characteristic of water mass. According to our data, 138Ce/142Ce ratio appears to be related with Ce content or Ce anomaly, reflecting sample locality. 143Nd/144Nd ratio also decreases with Nd content, but their relationship is less localized than that observed for 138Ce/142Ce ratio.

New data on 228Ra and 226Ra concentrations in the northern North Pacific are presented. Due to upwelling in the northern latitudes, 226Ra concentration is twice as much as that of the mid Pacific. The trend in the variation of surface 226Ra and dissolved SiO2 concentrations appears to be the same in the Pacific and the Atlantic oceans. A map of 228Ra/226Ra activity ratio in the surface water of the Pacific Ocean was constructed based on our data and those in the literature. The high 228Ra/226Ra activity ratio follows the prevailing surface currents such as the Kuroshio and the California Currents with low values at the center of the subarctic and the mid North Pacific gyres. The map is useful to identify the fluvial, coastal and shelf influences in the surface open ocean.

[1] Dissolved aluminum, indium, and cerium were measured in seawater from the Sea of Japan and the Sea of Okhotsk. Surface maxima of Al in the eastern Japan Basin indicate a signature of atmospheric mineral dust deposition resembling that of In. Data from all stations in this study show that Ce was markedly high in well-oxygenated bottom waters. The Ce:Al ratios of settling particles and surface sediments implicate the benthic flux of Ce as its cause. The In:Al ratios found in this study are not as high as those of the South China Sea and Sulu Sea, suggesting the minor influence of atmospheric deposition on the In:Al ratios in seawater. Fractionation between In and Al during scavenging processes, in addition to input from the bottom sediment, might contribute to the In enrichment of seawater. Apparent differences between In and Ce distributions in the marginal seas show that the benthic supply mechanism of In is not merely a simple redox-related process.

[1] 230Th activities were determined in seawater collected from the Bay of Bengal and the Andaman Sea. The Andaman Sea is a semi-closed basin that is isolated from the Bay of Bengal by the Andaman-Nicobar Ridge. The noticeable trend in the vertical distribution of 230Th activity in the Andaman Sea is its uniformity below 1250 m depth. We applied the scavenging-mixing model to this vertical profile in the Andaman Sea, and estimated the upper limit of renewal time to be 6 yr. We compared 210Pb/226Ra ratios among deep waters of various oceanic basins to constrain the application of 230Th as a tracer of deep-water renewal. The 210Pb/226Ra ratios were almost identical for the deep-water of the Andaman Sea and the inflow water from the Bay of Bengal, suggesting that the deep water in the Andaman Sea is replaced rapidly without bottom scavenging effects on the 210Pb/226Ra ratio.

International audience; Depth profiles of cerium and neodymium isotopic ratios and REE abundances are reported for three water columns (DE-4, MF52-59, and MF62-67) of the North Pacific Ocean. REEs in seawater of the DE-4 samples are collected by coprecipitation with Fe(OH)3, while others are by in situ direct adsorption onto MnO2-impregnated fiber. Abundance ratios of (HREE)/(LREE) are lower for the MF samples than for the DE-4 samples, indicating lower recoveries of HREEs in the MnO2-fiber method than in the Fe(OH)3 coprecipitation method. In the epsilon(Nd) depth profile for the DE-4 site, the surface water shows less radiogenic epsilon(Nd) of -8, suggesting large contribution of aeolian dust input. The deeper waters have epsilon(Nd) = -5 similar to other Pacific deep waters. This profile proves that bottom water keeps epsilon(Nd) value around -5 at the North Pacific Ocean even if surface waters show extremely low epsilon(Nd) values due to local source-inputs. Epsilon(Ce) values for seawater samples studied range from -1.21 to +1.82, showing chiefly positive values. The positive epsilon(Ce) values for most of seawaters contrast with the negative epsilon(Ce) values for most of Pacific ferromanganese nodules. The positive epsilon(Ce) and negative epsilon(Nd) values for most of the Pacific seawaters can be interpreted by the mixing model of continental and mantle-like oceanic/hydrothermal sources. Only surface waters have MORB-like negative epsilon(Ce) values comparable to nodules, perhaps suggesting preferential photo-reduction of MORB-derived water-borne Ce to continental-derived airborne Ce. This photo-reduction, together with a large aeolian-dust input, is suggested to be responsible for surface maxima of Ce anomaly observed for the DE-4 surface samples.