OS23D-1280
Isotopic Composition of Dissolved Iron in Different Oceanic Basins
The iron oceanic cycle is involved in the global carbon cycle, notably through primary production limitation in High Nutrient Low Chlorophyll (HNLC) areas. As for other elements, the study of iron isotopes will help better understanding its oceanic cycle, including tracing its sources to the ocean and studying its speciation (redox, organic and physical). Although several studies reported isotopic data in the marine environment (plankton tows, pore waters, aerosols, seafloor or marginal seas; i.e. Bergquist and Boyle, 2006; Severmann et al., 2006; de Jong et al., 2007), no dissolved iron isotopic data in seawater in the open ocean has been published so far because of the analytical challenge of this measurement due to the very low iron content of seawater combined to a concentrated salt matrix. This work will present the first dissolved iron isotopic composition data from the Southern Ocean (Kerguelen area and Atlantic sector) and from the Equatorial Pacific Ocean. Preliminary data, from a deep profile (30 to 4000 m depth) in the Atlantic Sector of the Southern Ocean, display small but significant variations around the crustal value (delta57Fe = -0.15 to +0.32 permil relative to IRMM-14). Complementary phases, such as suspended particles, phytoplankton, sediments, pore waters and aerosols will also be presented. These data will be discussed in terms of iron sources to the ocean and/or transfers between different iron reservoirs. Potential applications of this new tracer will be discussed.
OS23D-1281
High Resolution Dissolved Fe and Al along 95°E in the South Indian Ocean: Results from the CLIVAR I8S Repeat Hydrography Section
The distribution of dissolved Fe and Al in the upper 1000m of the South Indian Ocean was determined during a meridional transect extending from 66°S to 28°S along ~95°E during the CLIVAR I8S Repeat Hydrography section (Feb-Mar 2007). The dataset consists of 37, 12 depth profiles of dissolved (0.4 μm filtered) Fe and Al spaced at ~1 degree intervals obtained using a trace metal clean rosette system and analyzed at sea by Flow Injection Analysis. The data reveals several noteworthy features on the sources and transport processes of Fe and Al in the southern portion of the Indian Ocean and South Indian subtropical gyre. A region of elevated dissolved Fe (~0.9 nM), presumably associated with an Antarctic shelf input, occupies the lower 900m of the water column and extends from the shelf break near 66°S to ~61°S. Dissolved Al values show a similar pattern, although the subsurface enrichment (~1.2 nM) is only noticeable below 200m and is not as pronounced. Fe concentrations dropped significantly away from the shelf (< 0.25 nM in upper 200m), and reached a minimum extending to ~600m depth from the Subtropical Front to 28°S. Between 55°S and 45°S, another Fe subsurface maximum of ~0.6 nM was observed between 400-1000m depth. This area lies downstream of the Kerguelen Plateau, where elevated dissolved Fe values driven by shelf sediment resuspension have previously been reported. Dissolved Al, a good tracer of atmospheric inputs to the upper ocean, exhibited low values (<0.5nM) in the upper 100 m of the water column south of the Subtropical Front, consistent with the accepted low dust deposition over the Southern Ocean. Near 33°S, however, Al levels increased abruptly to ~7 nM. This sharp gradient constitutes the southern boundary of a region of elevated Al values that extends to 8°S in the CLIVAR I9N section and occupies the upper 200m of the water column. These elevated Al values are consistent with a dust source from the arid coast of western Australia given the mean southeasterly winds at these latitudes. At the core of this region near 20°S, we estimate a dust deposition flux of 0.95 g m-2 yr-1 using surface dissolved Al values as a proxy for eolian input. High resolution Fe and Al data from this section further illustrates the role of continental shelves in supplying Fe to the macronutrient rich waters of the Southern Ocean and reveals new patterns of dust deposition previously unrecognized in the South Indian subtropical gyre.
OS23D-1282
Distribution and Chemical Reactivity of Dissolved Iron in Surface Waters of Northwestern Pacific Marginal Seas
Spatial variation and chemical reactivity (based on differential ion exchange method) of dissolved iron in surface seawater from northwestern Pacific marginal seas, including the South China Sea (SCS), Western Philippine Sea (WPS), East China Sea (ECS), and Taiwan Strait (TWS), were determined. Total dissolved Fe concentration ranged from 0.2 nM in the oligotrophic SCS waters to 1-3 nM in offshore marginal sea (Taiwan Strait and northern SCS) and shelf (ECS) waters, and 10-50 nM in coastal waters off China and Taiwan. The cation exchangeable (Chelex-100 retained) fractions accounted for the majority of dissolved Fe in offshore water, and the dissolved inert (unexchangeable) Fe fractions made up 50-90% of dissolved Fe in coastal waters. The anion exchangeable (AGMP 1 retained, organic) Fe fractions were the smallest among the fractions determined, and only eminent in coastal waters, especially near river mouths. This study suggests that Fe was more active and formed strong Fe complexes, and removed quickly offshore. The distributions of Fe including all its different reactivity fractions revealed that Fe behavior is source dependent.
OS23D-1283
Iron(II) Intercomparison in Baja California Sub-Oxic Zone
An iron(II) method intercomparison was performed on a cruise in sub-oxic waters off the coast of Baja California in June/July 2008. Iron(II) is an important bioavailable form of iron that is scarce in oxygenated seawater due to its short half-life, but significant concentrations of Fe(II) have been documented in low oxygen oceanic environments. Two different iron(II) analytical techniques, both based on the Waterville Analytical FeLume flow injection analysis system with luminol, were used to measure in-situ iron(II) at three stations over the sub-oxic zone. Measured iron(II) values at each station compared well between the two methods, with an r2 value of 0.94. The use of a calibration curve vs. the standard addition method was also compared by measuring 100, 500 and 1000 pM iron(II) standards with each method. A calibration curve was determined to be preferred in measuring iron(II) because of consistent underestimation of iron(II) concentrations due to rapid decay during standard additions. When decay rates were accounted for, iron(II) measurements agreed well between methods, suggesting that differences among users and reagents are slight. The greatest in-situ iron(II) concentrations were found near the top of the suboxic water column, coinciding with a nitrate minimum at approximately 125 m. Due to the general instability of iron(II) in seawater, the presence of iron(II) implies a biological reduction pathway of iron possibly in connection with the nitrogen cycle at these depths. Maximum iron(II) concentrations also correlated with the secondary deep chlorophyll maximum in the upper suboxic zone, a region which previous studies have shown to be dominated by a unique population of Prochlorococcus. This may be indicative of production or utilization of iron(II) by these organisms.
OS23D-1284
Correlating Remotely-Sensed Ocean Productivity Variations to Volcanic Ash Deposition Events
Similar to wet and dry atmospheric deposition of mineral dust, ash deposition from large volcanic eruptions has been hypothesized to release sufficient iron (Fe) to stimulate surface ocean primary production, providing a plausible link between intense volcanism and abrupt climate changes in the past. There is, however, very little information on the atmospheric deposition flux of volcanic ash or the subsequent seawater dissolution of ash associated Fe in the surface ocean, although a few recent laboratory studies have demonstrated that substantial amounts of macro- and micro-nutrients can be released from ash within 1-2 h. One way to test the idea that volcanic ash can naturally enrich the ocean acting as a source of micronutrients for marine organisms is by monitoring chlorophyll-a, an accepted proxy for phytoplankton concentration. Chlorophyll-a is accurately estimated from routinely-collected ocean color sensor imagery. Through the study of a readily-available MODIS data set, we carried out a systematic investigation of localized anomalous chlorophyll-a responses that could be correlated to recent volcanic ash deposition events. Previous works have shown that phytoplankton response to artificial and natural iron fertilization is most apparent in high-nutrient low-chlorophyll (HNLC) regions. The focus of this study is on the known HNLC region of the Northern Pacific Ocean which was the chosen site for earlier large-scale Fe fertilization experiments (SEEDS and SERIES), and is also conveniently near to Kamchatka, Japan, and Alaska volcanoes. MODIS images corresponding to periods before and after ash deposition for 25 eruptions were systematically investigated for chlorophyll response. Our findings thus far have not shown a reliably detectable effect in relation to ash deposition. We provide a discussion on the possible reasons for these findings.
OS23D-1285
Colloidal Size Spectra of Fe in Seawater and River Water as Revealed by Flow Field-Flow Fractionation
The size spectra of colloidal Fe between 0.5 and 40 nm was determined in coastal seawater from the Gulf of Mexico, and in river waters from the lower Mississippi, Atchafalaya and Pearl Rivers, using on-line coupling of flow field-flow fractionation to UV-absorbance, fluorescence and refractive index detectors, and to high resolution ICP-MS. In the seawater, the colloidal fraction of Fe was mostly associated with small (~ 0.5-4 nm) humic-type fluorescent organic matter. The size spectra of protein-type fluorescence indicated in-situ formation of larger (~ 3-15 nm) organic matter, but only with a minor fraction of the Fe associated. In the rivers, colloidal Fe was partitioned between small (~ 0.5-4 nm) humic-type fluorescent organic matter and a population of larger (> 4 nm) Fe rich colloids. The average size of these Fe rich colloids, and the fraction of dissolved Fe found in the 0.5-40 nm size region, varied with DOC-content and water chemistry. For example, in the high-DOC Pearl River, most of the Fe-rich colloids were ~ 4-30 nm in size, and ~ 50 % of the total dissolved Fe was detected in the 0.5-40 nm size region. In the lower-DOC and more alkaline Mississippi river, the Fe-rich colloid population had a larger average size, and a smaller fraction (~ 20 %) of the total dissolved Fe was found in the 0.5-40 nm size region. Tracer experiments using stable 57Fe, or radioactive 59Fe should reveal the mechanisms by which Fe associates with different types of natural colloidal matter, and the role of organic matter on the formation and size of Fe rich colloids.
OS23D-1286
Distribution of iron isotopes in dissolved and particulate iron from the anoxic Gotland Basin in the the Baltic Sea
Fe isotope ratios are a potentially useful tool to determine the sources of marine dissolved and particulate Fe in the present day ocean and the ocean's redox state of the past. However, to date there are only a few measurements of Fe isotope ratios available from the marine environment. Almost no data exists for dissolved Fe, yet knowledge of the mechanisms and magnitude of Fe isotope fractionation in the marine Fe cycle are essential, if Fe isotope ratios are to be applied to adress either of the above issues. We have analyzed Fe isotope ratios in dissolved and particulate Fe across a depth profile from the anoxic Gotland Basin in the Baltic Sea, sampled during two cruises in 2005. Particulate Fe was measured on MC-ICP-MS using standard bracketing and show δ56Fe values slightly depleted in the heavy isotopes with little variation down the profile. Dissolved Fe was measured with a double spike after co-precipitation with Mg(OH)2. Anoxic samples taken from bottle-casts and filtered in a clean bench may show large offsets in dissolved Fe isotope ratios in comparison to pump-CTD in-line filtered samples, suggesting the likely introduction of artifacts during filtration in contact with air. Particles generally show much higher δ56Fe values than the dissolved phase. Among pump CTD samples, the small number of dissolved Fe samples collected during the first cruise in July at the beginning of the cyanobacteria bloom season show much lower δ56Fe values than those sampled in October, after the bloom. Since neither set of pump CTD samples was exposed to the atmosphere during filtration, an artifact is not likely. Instead, the apparent seasonal difference could be the result of a growing proportion of Fe remineralized from sinking particles of relatively high δ56Fe towards the end of the surface plankton bloom. Prior to and early during the bloom, dissolved Fe at depth may predominatly originate from the basin margin sediments after suboxic early-diagenetic remineralization, for which low δ56Fe values are commonly reported. The apparent seasonal dynamics in Fe isotope ratios may provide useful information regarding the modern marine Fe cycle, but will complicate the use of Fe isotopes as a paleo-redox proxy.
OS23D-1287
Role of Natural Organic Matter in Regulating the Partitioning of Fe(II, III) in Seawater
Iron (Fe) is an essential micronutrient and plays an important role in controlling ocean productivity and carbon cycling. Fe has been shown to be mostly complexed with dissolved organic matter in seawater. However, the interaction of Fe with natural organic matter and how the quality and quantity of organic matter affect the chemical speciation of Fe in seawater remain poorly understood. Controlled laboratory experiments have been conducted to examine the partitioning of Fe(II, III) between dissolved, colloidal and particulate phases using radiotracers, model organic compounds, and ultrafiltration. In natural seawater, Fe is mostly partitioned in the colloidal and particulate phases, resulting in a logKd value of 7.3 and a logKc of 6.1, respectively. On average, about 25% of dissolved Fe-55 was found in the <1 kDa fraction, 56% in the 1-10 kDa small colloidal fraction, 11% in the 10-100 kDa medium colloidal fraction, and 8% in the large colloidal fraction (100 kDa-0.4 μm). In experimental treatments with extracellular polysaccharides, the partitioning of Fe(II, III) changed from mostly in the colloidal fraction to mostly in the <1 kDa truly dissolved fraction, with on average 67% in the <1 kDa, 13% in the 1-10 kDa, 4% in the 10-100 kDa, and 6% in the 100 kDa-0.4 μm fraction. The increase in Fe(II, III) solubility in the presence of extracellular polysaccharides is hypothesized to be the result of Fe reduction from Fe(III), a more particle-reactive form, to Fe(II), a more soluble form, during its interactions with natural organic matter in seawater. This experimental result has important implications for the biogeochemical cycling of Fe(II, III) and other redox sensitive trace elements in the ocean. While the complexation of Fe with DOM could depress the bioavailability of Fe in seawater, the resultant Fe reduction may significantly enhance its solubility and bioavailability to marine organisms.