OS23E-1288
Dust Deposition and the Burial of Organic Carbon
Understanding the behavior of the carbon cycle necessitates analyzing the differing processes involved in
the production, transport, and eventual preservation of organic carbon in marine sediments. To quantify the
potential role of dust in these processes, we propose a connection between dust fluxes to the ocean and the
physical protection of organic matter in ocean sediments. Previous work suggests a linear relationship
between the preservation of organic matter and mineral surface area [4]. Assuming that dust contributes to
this mineral fraction, we calculate the organic carbon that would be associated with its surface area. We find
that the organic matter associated with dust would be about 10% of the total organic carbon buried in ocean
sediments. This finding is consistent both with global sedimentation rates [3] and with recent work based on
marine particle flux data [1,2]. Our work suggests a potential connection between the physical protection of
organic matter and the associations found between the lithogenic and organic carbon fluxes in the ocean
[1]. It also strengthens the idea that large perturbations to dust fluxes could directly impact carbon dioxide
levels in the atmosphere.
[1] R.A. Armstrong, C. Lee, J.I. Hedges, S. Honjo, and S.G. Wakeham. A new, mechanistic model for
organic carbon fluxes in the ocean based on the quantitative association of POC with ballast minerals.
Deep-Sea Research Part II, 49(1-3):219--236, 2001.
[2] J.P. Dunne, J.L. Sarmiento, and A. Gnanadesikan. A synthesis of global particle export from the surface
ocean and cycling through the ocean interior and on the seafloor. Global Biogeochem.
Cycles, 21, 2007.
[3] R.A. Jahnke. The global ocean flux of particulate
organic carbon: Areal distribution and magnitude. Global Biogeochem. Cycles, 10:71--88,
1996.
[4] L.M. Mayer. Surface area control of organic carbon accumulation
in continental shelf sediments. Geochim. Cosmochim. Acta, 58(4):1271--1284, 1994.
OS23E-1289
High Precision and High Sensitivity Measurements of Osmium Isotopes in Natural Waters
Direct measurements of Os in water are critical in understanding the geochemical cycle of Os in the environment. However, measurements of Os isotopes in natural waters are challenging due to a) low concentrations (~10 fg/g or less; 1 fg/g = 10-15 g/g) and b) the differences in oxidation states of naturally occurring and tracer Os that prevent accurate determination of Os concentration by isotope dilution [Sharma et al., GCA 61:5411, 1997]. It has been recognized for more than a decade that the best way to chemically separate and purify Os and at the same time achieve tracer-sample equilibration is to oxidize Os in sample-tracer mixture to OsO4. Three techniques have been developed: 1) heating of sample-tracer mixture with Br2 and Cr6+ in Teflon bombs at 90°C and solvent-extraction of OsO4 with Br2 [Levasseur et al., Science 282:272, 1998]; 2) heating to 180°C with Cr6+ in sealed glass (carius) tubes and its extraction by distillation [Sharma et al., GCA 63:4005, 1999]; 3) heating of water in the presence of H2O2 and H2SO4 and distillation of OsO4 [Woodhouse et al., EPSL 173:223, 1999]. The blanks for these techniques are [Os] = 22 fg 187Os/188Os = 0.47, [Os] = 19 fg 187Os/188Os = 0.27, and [Os] = 120 fg 187Os/188Os = 0.31, respectively. We have modified the carius tube technique by using a High Pressure Asher at 300°C and a confining pressure of 100 bars. This method is an improvement over previous techniques because the time required to achieve complete oxidation is much shorter due to the increased temperature of reaction and the blanks are significantly lower ([Os] = 2.2 fg, 187Os/188Os = 0.18) due to smaller amounts of reagents used. Additionally, we have modified the mass spectrometry associated with measuring low level Os samples. Typically, Os is measured on a single Pt filament as OsO3-, but we have modified the technique to include a double filament geometry. We use Ta for the ionization filament and Pt for the evaporation filament. The double filament geometry allows us to run samples at a lower temperature, thus reducing the blank contribution from the Pt filament. Not only do we find total Re counts to be reduced, but organic interferences on mass 233 (185Re.16O4-) are no longer present, allowing for more accurate Re correction. Repeated measurements of a 200 fg standard gives 2σ uncertainty of 1.38% for 187Os/188Os and 0.32% for 190Os/188Os. The improvements in chemistry and mass spectrometry allow the analysis of low level water samples such as snow from Antarctica and to understand the transport of Os from atmosphere to the oceans. We have applied this technique to surface and deep seawater samples collected from the Atlantic Ocean through GEOTRACES (Summer '08) and find that the surface seawater has been impacted by inputs of anthropogenic Os.
OS23E-1290
Transition metal stable isotope data for North Atlantic aerosols: first data and implications for atmospheric inputs to the ocean
The biogeochemical cycling of trace metals in the oceans has been an important pursuit for some time and, with the GEOTRACES programme, is set to become a major research endeavour in the next few years. The isotope geochemistry of the trace metals is a newer field, but is likely to become important in the future. For example, isotopic fractionations of the transition metals can yield quantitative information on the processes controlling the distribution of these metals in the ocean as a result of biological and other processes1,2. Any study of the cycling of these trace metals in the oceans, however, must be preceded by a quantitative understanding of the inputs, particularly to the surface ocean. Atmospheric deposition is one of the most important of these inputs for many metals. Here we report isotopic data for a range of transition metals in aerosols. The samples were collected in June-August 2003 as part of the CLIVAR/CO2 Repeat Hydrography program and derive from the central Atlantic between 63°N and 3°S. Total aerosol trace metal contents, as well as amounts of soluble trace metals, have been reported previously3.Here we concentrate on the isotopic composition of transition metals instantly soluble in pH 5.6 de-ionized water. There are concerns about isotopic fractionation during the leaching process but this fraction may also best reproduce that released from aerosols during wet deposition from the atmosphere. The Zn isotopic dataset is the largest we currently have available, and the δ66Zn = 0.4±0.3 (2 s.d.) for the entire sample set down the section. Within this, however, there is a small but significant amount of isotopic variability. For example, samples dominated by Saharan dust at 13-26°N have an isotopic composition very close to the small dataset for rocks from the continental crust at around 0.3 per mil. On the other hand, samples collected at 0-7°N, and thought to be influenced by biomass burning, form a very tight cluster at δ66Zn = 0.50±0.06. Cu concentrations in the soluble fraction are low and so the dataset is smaller but these latter samples appear to be significantly lighter than rocks while Saharan dust is slightly isotopically heavy. Preliminary Fe isotopic data suggest a positive correlation between δ56Fe and δ66Zn but with a much larger range in isotopic composition, from -1 to +2. Overall, these initial data suggest that the average input from aerosols to the surface ocean, if the DI- soluble fraction can be taken as representative, is not significantly different from that expected from what we know about the continental crust. 1Vance et al. (2008) EPSL, in press. 2Andersen et al. (2008), abstract of paper presented at Fall AGU 2008, San Francisco. 3Buck, C.S. et al. (2008) Mar. Chem, in press.
OS23E-1291
Linking Atmospheric Deposition and Ocean Flux of Fe and Al in the Sargasso Sea
A comparison of atmospheric Fe and Al fluxes to their oceanic fluxes at Bermuda and nearby Sargasso Sea from June 1999-2000 shows a seasonal decoupling on monthly timescales, yet good agreement (within ±20-30 percent) on an annual timescale. This suggests that other than atmospheric input, the intrinsic physical and biogeochemical processes within the water column could differentially affect the deeper sinking fluxes of Fe and Al. Such processing evidently causes the temporal decoupling between atmospheric inputs and oceanic export on sub-annual time scales. The annual balance between the atmospheric and oceanic fluxes of Fe and Al support atmospheric dust deposition to be the most important lithogenic source of Fe and Al to the deeper Sargasso Sea. However at depth (3200 m), both Fe and Al suggest minor additional inputs (20-30 percent), apparently from lateral advection associated with benthic circulation. Thus the eventual oceanic removal of Fe and Al to the benthos appears linked to the sequencing of atmospheric dust deposition, surface production, and mid-water processing. As such, mesopelagic waters appear to serve as an important reservoir of atmospherically derived Fe during the high dust season. However, Al appears to be efficiently removed from the surface layer to deeper depths by mid-water organisms, even during periods of low pelagic production. Such sub-annual redistribution of Fe and Al is presumed to reflect basic biogeochemical differentiation during the oceanic processing of these two diverse lithogenic elements.
OS23E-1292
Concentration, Complexation and Chemical Speciation of Zinc and Cadmium in the Western North Pacific Ocean : Exploring Sources and Transport of Trace Metals and Complexing Ligands.
We determined Zn and Cd total dissolved (0.45 ”m-filtered) concentrations, organic complexation and chemical speciation in surface water samples collected along the transect of the 2002 IOC Baseline Contaminant Survey expedition in the Western North Pacific and in vertical profile water samples at nine stations. The goals of this work were (1) to compare and contrast various trace metal sources, including both natural and anthropogenic atmospheric deposition, upwelling, marginal seas and others; (2) to study the organic ligand sources, generally thought to be phytoplankton; and (3) to investigate metal and ligand transport mechanisms, residence times and eventual upwelling in the Eastern North Pacific. Total dissolved (TD) Zn and Cd values were obtained using a combination of differential pulse stripping anodic voltammetry (DPASV), preconcentration with 8-HQ or APDC/DDC and quantification at ICPMS or AA. Organic complexation and chemical speciation of Zn and Cd were determined simultaneously using DPASV at a thin-mercury-film, glassy-carbon-disk-electrode. Surface transect TDZn and TDCd concentrations were low in the Subtropical Gyre (STG), in contrast with high values in the Western Subarctic Gyre (WSG). Zn and Cd were organically complexed in most surface samples: at least one ligand class was detected for Zn and Cd, whose conditional stability constants (log K') averaged 10.2 and 10.5, respectively. These ligands were found in excess of the total dissolved metal throughout the region of study except in the WSG for Cd. Vertical distributions of TDZn and TDCd exhibited nutrient-type profiles for all the STG stations. While constant Zn/Si and Cd/P values were observed throughout the water column in the WSG, some deviations were observed within the STG. In addition, the mode and intermediate water masses of the STG displayed very high concentrations of a Zn-complexing ligand (log K' 10.0) in excess of TDZn. As these water masses moved eastward, we observed that the ligand concentrations decreased. In contrast to the STG, the upper 1000m of the WSG showed elevated concentrations of both metals. Despite elevated surface (0-200m) Zn concentrations (~2nM), a Zn-complexing ligand (log K' 9.8) was found in excess of TDZn; below the photic layer, even higher TDZn concentrations might have saturated the ligand. A ligand for Cd was present in lower-than-TDCd concentrations in the same surface waters; below them, organic complexation of Cd was observed rarely in both STG and WSG regions. By studying the geographic distribution of the total dissolved metals and ligands, along with other dissolved and particulate tracers, possible sources and transport mechanisms can be contrasted and evaluated. Furthermore, the influence of these sources and transport mechanisms on the distribution of Zn and Cd chemical species and, ultimately, the bioavailability of these micronutrient metals can be studied.
OS23E-1293
Influence of Smoke From Biomass Burning on the Dissolution of Iron From Australian Aeolian Dusts
Primary production by phytoplankton throughout much of the remote ocean, especially the Southern Ocean, is limited by the availability of iron. Aeolian dust is a major source of iron to remote waters. Surface water concentrations of iron are controlled by processes that affect the solubility of iron from aeolian dust. Australia is a major supplier of aeolian dust to iron-limited waters of the Southern Ocean. Australia is also subject to large biomass burning bushfires. The smoke from bush fires contains many complex ligand-like organic substances that could affect iron solubility. In a simple proof of concept laboratory experiment we report results from the dissolution of iron and aluminium from smoke-altered aeolian dust in simulated acidic cloud waters. We conclude that biomass burning smoke has no observable affect on the solubility of iron or aluminium. We also note that the ratio of Fe:Al during dissolution of both "smoked" and "unsmoked" dust changes with time and may take months to reach equilibrium and that therefore caution must be used with the common practice of using Al concentration as a proxy for iron.
OS23E-1294
Iron Speciation Influences Solubility in Dust Sources of HNLC Waters
A primary source of bioavailable iron (Fe) to the open ocean is terrestrially-derived particulate aerosols, often assumed to be sourced from arid soil-based dust storms. However, portions of Fe-limited regions of the ocean are in the high latitudes, proximal to coasts that are strongly influenced by glacial weathering and associated sediment fluxes. In these waters, glacial dust storms could be an important intermittent source of relatively soluble Fe to phytoplankton. Additionally, recent work has suggested that anthropogenic sources, such as fossil fuel combustion, are important contributors of soluble aerosol Fe to the ocean, which is particularly relevant to Fe-limited waters downwind of industrial or developing nations. Here, we quantify the solubility of Fe in Alaskan glacial flour, Chinese loess, African dust, and oil fly ash; a suite of samples likely to represent aerosol sources to many Fe-limited waters. We have conducted a series of successive leaching experiments following the protocol of Sedwick et al. 2007 to measure Fe solubility. Additionally, we determined the speciation of Fe in these sediments by synchrotron-based X-ray absorption spectroscopy (XAS). We demonstrate that the particulates differ substantially in solubility by dust source, which can be partially explained by fundamental differences in the partitioning of Fe species. Sediments sourced in arid soils are least soluble, and Fe is primarily composed of particularly insoluble secondary pedogenic ferric phases. Glacial flour Fe is roughly an order of magnitude more soluble than arid soil Fe, likely due to the large fraction that exists as ferrous primary and secondary silicate minerals in the glacial flour matrix. Interestingly, oil fly ash Fe is an additional order of magnitude more soluble than that in glacial flour, which is due to the presence of extremely soluble ferric mineral phases. Our study emphasizes the potential importance of Fe aerosol source and related differences in speciation to Fe solubility in aerosol particulates. These results should be considered when determining the flux of bioavailable Fe to waters thought to be Fe-limited, particularly in the context of global and regional changes in climate and fossil fuel combustion. References Cited: Sedwick et al. (2007) Geochem Geophys Geosyst, Vol 8, Q10Q06
OS23E-1295
Estimation of Fe and Al aerosol solubility from a diffusive method based on particle size
Mineral aerosol deposition is recognized as the dominant source of iron to the open ocean. Solubility of this aerosol iron is highly uncertain. Previous solubility estimates range widely from 0.01-80%. Here we use the diffusion equation to solve for the solubility of different particle sizes. We derived a time and spatial varying solubility map for aerosol Fe and Al based on the aerosol lifetime in the atmosphere and aerosol deposition for different size bins from a global dust model (DEAD). Our solubility map corresponds well with the observations from the CLIVAR/CO2 cruise in the North Atlantic Ocean. But it does not explain the large solubility variation observed from the 2002 IOC cruise in the North Pacific Ocean. We examine the impacts of spatially variable, size-dependent solubility on marine biogeochemistry with the Biogeochemical Elemental Cycling (BEC) ocean model. Simulated values of dissolved Al and Fe were compared with observations under constant and variable aerosol solubility. Diffusion-based variable solubility considerably reduces the global RMS error of the simulations.
OS23E-1296
Modeling the Nd Oceanic Cycle Using a Fully Prognostic Dynamical/Biogeochemical Coupled Model
The "Nd paradox" refers to the decoupling of Nd isotopic composition (hereafter refered as epsNd) and Nd concentration in the ocean. While epsNd is commonly used as a conservative water-mass tracer in the ocean interior far from any lithogenic inputs, Nd concentration increases with depth and along the thermohaline circulation, resembling typical non-conservative patterns of nutrients that are affected by vertical scavenging. Also, Nd oceanic budgets inferred from traditional sources (aeolian dusts and riverine discharge) cannot account for both epsNd and Nd concentration (Tachikawa et al., 2003). Recently, "Boundary Exchange" (BE, i.e. process of boundary scavenging compensated by inputs from the sediments along the continental margin) has been proposed as an important source/sink term of Nd that could solve the "Nd paradox". Here, we simulate the oceanic epsNd and Nd concentrations using the coupled dynamical/biogeochemical ocean general circulation model NEMO/PISCES. A reversible scavenging model has been parameterized to reproduce vertical cycling and dissolved/particulate interactions. Sources taken into consideration are partial dissolution of sediment (source of the BE process), atmospheric dust and dissolved riverine discharge. Some discrepancies in modeled particle fields create too pronounced vertical gradients in Nd concentrations and may lead to an overestimated role of the reversible scavenging. However, results show that: 1) the reversible scavenging is a necessary component to reproduce both epsNd and Nd concentration distributions and thus explain the "Nd paradox", and 2) BE is the predominant Nd source to the ocean (1.0E+10 g(Nd)/year) with fluxes as large as 25 times those of other sources (4.0E+8 g(Nd)/year). Hence it likely represents the missing flux necessary to reconcile Nd and Nd concentration budgets. Nevertheless, dust inputs are necessary to constrain epsNd values in surface waters. Tachikawa, K., Athias, V., and Jeandel, C.: Neodymium budget in the ocean and paleoceanographic implications, Journ. Geophys.Res., 108, 3254 doi:3210.1029/1999JC000285, 2003.
OS23E-1297
Cobalt Biogeochemistry in the South Atlantic: A Full-Depth Zonal Ocean Section of Total Dissolved Cobalt, and Development of a High Throughput Cobalt ICP-MS Method
This study presents the first high-resolution full-depth zonal section of total dissolved cobalt from a recent cruise transecting the South Atlantic Ocean along approximately 11S. This section demonstrates that current electrochemical analytical techniques are capable of producing the high precision and high resolution datasets for total dissolved cobalt expected to be generated as a part of the international GEOTRACES Program. The micronutritive role of cobalt may affect community structure in different regions of the oceans, a compelling reason to include cobalt in the trace element analyses planned for the GEOTRACES Program. This cobalt section reveals an advective source of cobalt from the African coast near Namibia, which we propose to be due to the Benguela Current interacting with reducing shelf sediments. These high concentrations of cobalt were also observed within the oxygen minimum zone that extends across much of the South Atlantic basin in this section, and are likely indicative of redox cycling of cobalt in the water column. Nutrient-like vertical structure of cobalt was observed in the surface waters across the majority of the basin due to biological utilization, and the expected hybrid-type trend is observed at depth, with scavenging of cobalt below the nutricline. Deepwater concentrations of cobalt were around 50pM across the basin below 3000m. Analysis of the shelf-life of refrigerated filtered samples stored without acidification for electrochemical cobalt analysis demonstrated that those samples which were collected specifically within oxygen minimum zones may underestimate cobalt if not analyzed within a few weeks of collection. These results motivate our on-going development of a method to measure cobalt in acidified samples via inductively coupled plasma mass spectrometry (ICP-MS). The benefit of this technique would be twofold: acidification would extend the shelf-life of the samples significantly, and samples would be preserved identically to those intended for other ICP-MS total metal analyses such as iron, manganese, cadmium, and zinc. Initial method development indicates that UV-irradiation is required to destroy natural cobalt-binding ligands, which we proposed is due to incomplete dissociation of organic cobalt complexes at pH values typically used for sample acidification.
OS23E-1298
Atmospheric deposition fluxes to Monetary Bay
Atmospheric deposition has been widely recognized as a source of pollutants and nutrients to coastal ecosystems. Specifically, deposition includes nitrogen compounds, sulfur compounds, mercury, pesticides, phosphate, trace metals and other toxic compounds that can travel great distances. Sources of these components include both natural (volcanoes, mineral dust, forest fires) and anthropogenic (fossil fuels, chemical byproducts, incineration of waste) sources, which may contribute to harmful health and environmental impacts such as eutrophication, contaminated fish and harmful algal blooms. This study looks at the flux of aerosol deposition (TSP - total suspended particle load) to Monterey Bay, California. Samples are collected on a cascade impactor aerosol sampler (size fractions PM 2.5 and PM 10) every 48 hours continuously. Preliminary results indicate that the TSP for PM 10 ranged from 0.026 to 0.104 mg m-3 of air and for PM 2.5 from 0.014 to 0.046 mg m-3 of air. Using a deposition velocity of 2 cm s-1 for the large fraction (PM10 - PM 2.5) and a deposition velocity of 0.7 cm s-1 for the fine fraction (PM 2.5) deposition rates are 13 and 86 mg m-2 d-1 respectively.
OS23E-1299
Solubility of Metals in Aerosols Over the East China Sea: correlation Between Co and Fe
Our previous study found that the trace metal composition in the phytoplankton collected in the SouthEast-
Asia-Time-Series (SEATS) offshore station of the South China Sea was mainly dominated by extracellular
portion. The metals were most likely derived from anthropogenic aerosols containing highly soluble trace
metals (Ho et al. 2007). In this study, total and soluble concentrations of Al, Ti, V, Mn, Fe, Co, Ni, Cu, Zn, Cd,
Ba, Pb and P were determined in the aerosol samples collected in 2005 at an offshore station in the East
China Sea. The total concentrations were much higher during winter and spring seasons; relatively, the
solubilities for most elements were generally high all year long but varied significantly and randomly. In terms
of total concentrations, V and Ni are not associated with any other metals; in terms of solubility, Co, Ni, and V,
three indicative metals for the emission from fossil fuel burning, show strongest correlation with Fe and Al.
Principal component analysis suggests that the soluble Fe came from the same source or process for the
soluble Co. Since Co in aerosols mainly comes from coal burning, we hypothesize that soluble Fe are closely
linked to the emission of fossil fuel burning in the region, presumably from coal burning.
http://www.aslo.org/lo/toc/vol_52/issue_5/1776.pdf
OS23E-1300
Spatial Variation in Chemical Properties of Suspended Matters in the North Pacific Surface Water
Oceanic suspended particulate matters (SPM) play an important role in controlling trace chemical composition of seawater via sorption and desorption of trace elements and other substances in SPM, because of its longer residence time compared to falling particulate matters and its larger surface area-to- volume ratio. The physical and chemical properties of SPM reflect the characteristics of the water mass, including the interaction with biota and/or atmospheric input. In this study, using both bulk and single particle chemical analysis techniques, size and elemental compositions of the SPM collected in the Pacific area is characterized. SPM were obtained from surface seawater of the North Pacific during the several cruises in summer. The bulk elemental composition of particles for Si, Ca, Ti and Fe was determined by an X-ray fluorescence spectrometry. Individual particles were also analyzed using an electron probe X-ray micro analyzer and characterized by size and elemental composition. Results of bulk analysis show that Si and Ca were much more abundant than Ti and Fe in surface water SPM. Average concentrations of Si and Ca were 3.3 ± 4.0 μg L-1 and 4.5 ± 7.6 μg L-1, respectively. Their high concentrations were found in the high primary productive region in the Northeastern subarctic Pacific. Fragments of diatoms and coccolithophorids were directly reflected to the amount of Si and Ca in SPM. Average concentration of Ti and Fe as the indicator of mineral particles was 96 ± 89 ng L-1 and 180 4± 210 ng L-1, respectively. The average Fe/Ti ratio in SPM was about a quarter of the ratio of crustal composition (Fe/Ti=8.5). Results of single particle analysis show that Ti and Fe were enriched in aluminosilicate-type particles with average diameter of 1.6 ± 1.2 μm. While Fe/Al ratio in the particles was constant with particle diameter, Ti/Al ratio was increased with decreasing particle diameter. It was suggested that chemical composition of mineral particles was modified during transport or suspension in seawater.
OS23E-1301
Trace Metal Transport in the Western North Pacific: Relative Contributions from Alternative Sources
Atmospheric deposition of Asian dust has been suspected as a major source of some trace metals to the Western North Pacific. However, the surface and vertical profile distributions of some trace metals do not always follow the distribution of mineral dust tracers, suggesting alternative sources, including continental margin inputs, industrial aerosols, upwelling, and advective transport from seas neighboring the Pacific. To investigate the inputs from these sources, filtered water samples and water column particulate samples collected on the Intergovernmental Oceanographic Commission 2002 North Pacific expedition were analyzed for Fe, Mn, Co, Ni, Cu, Cd and Pb using ICP-MS. In addition to these trace metals, the particulate samples were also analyzed for REE and other elements (P, Si, etc.) as tracers. A low salinity/low temperature intrusion at 450m was observed off the coast of Japan. This water mass is distinguishable within the water column by high oxygen concentrations and enrichments in particulate REE and a Mn:Fe ratio greater than crustal abundance. The high oxygen values suggest recent entrainment of surface water, while the high REE values indicate dust loading from the Asian desert. However, the elevated Mn:Fe ratio above crustal abundance at this depth suggests that at least some material is being remineralized from the shelf during transport. The origin of this water mass is yet to be determined. In the Western Subarctic Gyre (WSG), dissolved Cd:P in vertical profiles suggest that deep water is being upwelled to near surface depths. While the vertical profiles of dissolved Ni, Cu, and Cd exhibit nutrient-like behavior, the surface concentrations are 3-50 times higher than typical open ocean values outside the WSG. In contrast, Co shows a surface maximum, suggesting an atmospheric source. The absence of mineral dust elements (Al, Ga) at the surface Co maximum suggests that industrial aerosols are the most likely source. Also within the WSG, particulate REE values indicate that continental margin material is being transported to the open ocean not only off the coast of the Kuril Islands, but also further north (50 N, 167 E).
OS23E-1302
Aerosol Chemical and Physical Properties Observed over Puerto Rico in the Tropical North Atlantic Ocean
Tropospheric aerosols that originate in Africa and are transported over the Atlantic Ocean have potential impacts over the Caribbean region. To investigate aerosol properties over this region, air sampling was conducted at San Juan Cape (18.46°N, 66.12°W), Puerto Rico during the summer months in 2006. Aerosol samples were collected by both commercial PM2.5 sampler and in-house fabricated TSP sampler. Analyses of aerosols were made through the use of the following instrumental methods: (1)Ion Chromatography for the determinations of water-soluble cations (sodium, ammonium, potassium, magnesium and calcium) and anions (fluoride, acetate, propionate, methanesulfonate, chloride, nitrate, succinate, malonate, sulfate and oxalate); (2)Inductively Coupled Plasma Mass Spectrometry for the concentrations of selected trace elements (Al, Fe, Mn, Sc, Cd, Pb, Sb, Ni, Co, Cr, Cu, Zn and V); (3)Scanning Electron Microscopy for individual aerosol particle characterization. Crustal enrichment factors were calculated to determine the strength of crustal source. Preliminary results indicate that sodium (22 - 99 μg m- 3) and ammonium (1.1 - 50 μg m-3) were the major cations and chloride (1.5 - 99 μg m-3) and sulfate (35 μg m-3) were the dominant anions. Malonate (3.8 - 6.9 μg m- 3) was the most abundant organic anion. Atmospheric concentrations of iron ranged 0.30 - 3.3 ng m- 3. The elements, Sc, Cd, Pb, Sb, Ni, Co, Cr, Cu, Zn and V, were enriched by factors of 600 to 40,000 relative to their natural abundance in crustal soil. Principal components analysis indicates six assemblages of fifteen types of aerosol particles, dominated by Si rich particles.
OS23E-1303
Speciation of Suspended Particles By Individual Particle Analysis In The Japan Sea And The Western Tropical Pacific Ocean
Trace elements such as Fe, Cd, Ni, Cu, Zn and Co are called gbioactive trace metalh. Reports have been made on bioactive trace metals distribution in sea water for various ocean: the North Pacific (Bruland, 1980; Boyle et al., 1981; Bruland et al., 1994; Ezoe, 2004;), the North Atlantic (Boyle et al., 1981; Bruland and Franks, 1983) and the South China Sea (Wen et al., 2006). The most of bioactive trace metals are taken up by marine organisms such as phytoplankton and bacteria. Consumption and decomposition of particulate matter sinking from surface waters return the bioactive trace metals to solution. On the other hand, some suspended particulate matters come from terrestrial sources transported to the ocean by rivers and by winds in particulate forms, and by rivers in dissolved forms. The bulk composition of suspended particulate matter in the various oceans is well known, whereas, the speciation of elements in suspended particle still remains poorly known. Individual particulate analysis can provide detailed information about the source, formation, transport and reactions of suspended particulate matter. The purpose of this work (1) the determination of dissolved bioactive trace metals (Fe, Co, Ni, Cu, Zn and Cd) in the Japan Sea and the western tropical Pacific Ocean by using the commercial PAPC type chelating resin solid phase extraction with ICP-MS method, (2) investigation for source of bioactive trace metals by the speciation of suspended particles by individual particulate analysis.
OS23E-1304
Aerosol Sampling and Analysis for the GEOTRACES Program
The GEOTRACES Science Plan emphasizes the importance of atmospheric deposition on the budgets and biogeochemistry of trace elements and isotopes in the world's oceans. With funding from the National Science Foundation, an aerosol and rainfall sampling program is being developed for use on future GEOTRACES cruises. This includes preparation and testing of dual high-volume TISCH 5170-VBL aerosol samplers for inorganic trace elements and isotopes, major ions, organic material, and isotopes of nitrogen and oxygen. A third 5170-VBL aerosol sampler is equipped with a 5-stage Sierra-style slotted impactor to collect size-fractionated aerosols for chemical measurements. The aerosol samplers will be operated using wind speed and wind sector control to avoid contamination from ship's exhaust. Duplicate automated rain samplers have also been developed to collect unfiltered and filtered rain samples. Rainfall will be filtered immediately (during collection) to avoid re-adsorption artifacts. Two intercalibration experiments are planned where aerosol and rainfall subsamples will be distributed to the community for testing and validation of analytical methods. The first experiment is being conducted in early September 2008 on the roof at RSMAS/University of Miami. Results from the GEOTRACES aerosol samplers will be compared to a multi-channel aerosol sampling system (using 47mm PCTE filters), and with ongoing aerosol collections at RSMAS. The second experiment is planned for the atmospheric sampling tower at Bellows AFB (Oahu, HI) in summer 2009. Details of the sampling equipment and sample collection methods will be discussed, along with preliminary results from the first intercalibration experiment. Community input will be solicited for planning the second intercalibration experiment.
OS23E-1305
Model Simulations of Dust Source Variability Using Surface Reflectance as a Proxy for Soil Erodibility and Its Impacts on Dust Transport and Deposition
Dust source locations and strengths are strongly sensitive to soil erodibility. It is important for dust simulations with correct soil erodibility. To test the effects of different soil erodibility on dust generation, transport, and deposition to the ocean, we compare the model simulation results by using two soil erodibilities, one based on TOPO method (Ginoux et al., 2001), and the other based on modified soil erodibility (Alf, et al., 2005) normalized by surface reflectance of MODIS satellite data. The results are intensively compared against observations. The results calculated by MODIS satellite data show that dust may be lower in southern hemisphere and higher in North African source compared to the results of TOPO. The results show that soil reflectivity measured by MODIS satellite is a useful proxy for soil erodibility, and the results are generic and not model-sensitive.
OS23E-1306
Atmospheric Contributions to Excess Nitrate Development in the Subtropical North Atlantic: Oceanographic Viewpoint
The main thermocline of the subtropical North Atlantic is known to have nitrate concentrations well in excess of expectations based on phosphate concentrations and Redfield elemental ratio assumptions. This excess N has generally been thought to accumulate primarily due to surface layer nitrogen fixation, and export of that N to the global ocean is thought to balance N sinks due to marine denitrification. Understanding the actual mechanisms and controls on excess N development is thus critical for understanding the global marine N budget. Here we demonstrate that excess N can be introduced by two additional mechanisms, each with contributions similar to nitrogen fixation. The first is atmospheric deposition of high N:P nutrients with an estimated addition rate of 3.0± 0.9 ×1011 mol excess N yr-1 (38 percent of the total). Second is the export to and remineralization of high N:P dissolved organic matter in the main thermocline, with a rate of 2.2±1.1 ×1011 mol excess N yr-1 (28 percent of the total). The rate at which nitrogen fixation adds excess N, calculated by mass balance of the excess N field, is 2.6 ± 2.2 ×1011 mol excess N yr-1 (33 percent of the total). If atmospheric deposition to the North Atlantic does in fact contribute to excess N development there, the increasing flux of anthropogenic N to the North Atlantic should appear as increasing concentrations of excess N in the main thermocline.
OS23E-1307
Modeling Aggregation and Combined Sedimentation of Dust and Organic Particles in the Atlantic Ocean
Mineral particles such as dust grains are often the primary source of dissolved marine micro-nutrients, and processes linking sedimentation of organic and mineral material may become important for nutrient cycling in the ocean. We introduced aggregation and disaggregation processes into a biogeochemical ocean circulation model in order to estimate the relative importance of aggregate formation for the sedimentation of organic and lithogenic particles. Dust deposition rates, which are mainly restricted to the tropical and subtropical North Atlantic, are provided by an Atmospheric General Circulation model. The results of the coupled model show seasonal cycles of primary production which are typical for the Atlantic Ocean. The organic aggregates formed in the mixed layer interact intensively with the pool of suspended eolian dust, and attached to the fast sinking organic material, mineral particles rapidly sink through the upper ocean into the deeper layers. In mid-depth regions, aggregation by differential settling and the release of dust particles by aggregate decay are the dominant processes which control the profiles of suspended dust. The horizontal distribution of the annual deep sea dust flux compares well with that of the atmospheric dust deposition rate, which suggests that oceanic large scale advection only plays a minor role in the spreading of eolian dust particles in the tropical and subtropical Atlantic. Observed deep sea fluxes of organic and lithogenic material, and dust concentrations recalculated from observed SPM-aluminium profiles were used to verify the model results.