Iron-rich Water Research Articles

A quantitative analysis of sources for summertime phytoplankton variability over 18 years in the South Shetland Islands (Antarctica) region

Eighteen years of summertime hydrographic and chlorophyll- a (Chl- a) data (∼2700 stations) from the South Shetland Islands (Antarctica) region show that a “bell-shaped” (unimodal) distribution of phytoplankton biomass results annually when plotted against the inshore to offshore gradient in surface salinity. The maximum for this unimodal Chl- a distribution corresponds with a shallow upper mixed layer (UML) in iron-rich waters that occurs at salinities ∼34. Methods of gradient analysis are used to distinguish sources of variability for bloom development among years. The control of phytoplankton biomass is resolved across the salinity gradient that separates the co-limiting conditions of deep UML depths and low-iron concentrations as opposing end-members. Chlorophyll-fluorescence yield data (a proxy for Fe-stress) showed that at salinities ∼34, phytoplankton biomass was unlikely to be limited by Fe. Instead, blooming at salinities ∼34 (1.3±1 mg Chl- a m −3) co-varied with shallow UML depths (41±19 m) that occurred as a function of higher UML temperature (1.5±0.5 °C) among years, and is evidence that atmospheric climate variability impacts summertime phytoplankton biomass and production in this Southern Ocean seascape.

Sediment diagenesis and porewater solute fluxes in acidic mine lakes: the impact of dissolved organic carbon additions

Sediment diagenesis through microbial sulfate reduction is considered a critical process in the pH amelioration of acidic mine lakes, but is often limited by the availability of organic carbon. Organic substrates are therefore frequently added to mine lake sediments to stimulate sulfate reduction. Dissolved organic carbon (DOC) was added to sediment collected from three mine lakes, one (in Germany) with typically high concentrations of Fe and SO4 and another two (in Australia) with unusually low concentrations of Fe and SO4. After the DOC additions caused the dissolved oxygen concentrations in the overlying waters to fall below 50 μmol L–1, the sediment porewater at all sites progressed through the expected anaerobic respiration sequence. The paucity of Fe and SO4 in the Australian lakes did not appear to constrain microbial iron and sulfate reduction. Indeed, the low Fe concentrations appeared to promote microbial sulfate reduction in the Australian sites. In the German site, there was little evidence of sulfide production in the porewater and no changes in porewater pH profiles. In contrast, the sediment porewater from the two Australian sites exhibited sulfide production and increased porewater pH. Bioremediation of acidic lakes must consider the need to treat iron-rich water before attempting pH amelioration.

Hickman Crater, Ophthalmia Range, Western Australia: evidence supporting a meteorite impact origin*

A newly discovered, morphologically well-preserved crater with a mean diameter of 260 m is reported from the Ophthalmia Range, Western Australia. The crater is located in hilly terrain ∼36 km north of Newman, and is situated in the Paleoproterozoic Woongarra Rhyolite and the overlying Boolgeeda Iron Formation. The morphometry of the crater is consistent with features characteristic of small meteorite impact craters. The rhyolite of the crater's rim exhibits widespread shatter features injected by veins of goethite bound by sharply defined zones of hydrous alteration. The alteration zones contain micro-fractures injected by goethite, which also fills cavities in the rhyolite. The goethite veins are interpreted in terms of forceful injection of aqueous iron-rich solutions, probably reflecting high-pressure hydrothermal activity by heated iron-rich ground water. None of these features are present in the Woongarra Rhyolite outside the immediate area of the crater. Petrography of the rhyolite indicates possible incipient intracrystalline dislocations in quartz. The Boolgeeda Iron Formation, which crops out only on the southern rim of the crater, displays brecciation and mega-brecciation superposed on fold structures typical of the banded iron-formations in the region. Geochemical analysis of two goethite veins discloses no siderophile element (Ni and PGE) anomalies, negating any contribution of material from an exploding meteorite. Instead, the strong iron-enrichment of the fractured rhyolite is attributed to a hydrothermal system affecting both the Boolgeeda Iron Formation and the Woongarra Rhyolite, and localised to the area of the crater. An absence of young fragmental volcanic material younger than the Woongarra Rhyolite is inconsistent with an explosive diatreme, leading us to a preferred interpretation in terms of an original impact crater about 80 m deep excavated by a ∼10 m-diameter projectile and accompanied by hydrothermal activity. A minor north–south asymmetry of the crater, and an abundance of ejecta north, up to about 300 m northwest and northeast of the crater, suggest high-angle impact from the south. A youthful age of the structure, probably Late Pleistocene (104–105 years old), is indicated by damming of the drainage of a south-southeast-flowing creek by the southern crater rim.

Arsenic distribution in the dissolved, colloidal and particulate size fraction of experimental solutions rich in dissolved organic matter and ferric iron

Due to the widespread contamination of groundwater resources with arsenic (As), controls on As mobility have to be identified. In this study we focused on the distribution of As in the dissolved, colloidal and particulate size fraction of experimental solutions rich in ferric iron, dissolved organic matter (DOM) and As(V). Size fractions between <5 kDa and >0.2 μm were separated by filtration and their elemental composition was analyzed. A steady-state particle size distribution with stable element concentration in the different size classes was attained within 24 h. The presence of DOM partly inhibited the formation of large Fe-(oxy)hydroxide aggregates, thus stabilized Fe in complexed and colloidal form, when initially adjusted molar Fe/C ratios in solution were <0.1. Dissolved As concentrations and the quantity of As bound to colloids (<0.2 μm) increased in the presence of DOM as well. At intermediate Fe/C ratios of 0.02–0.1, a strong correlation between As and Fe concentration occurred in all size fractions ( R 2 = 0.989). At Fe/C ratios <0.02, As was mainly present in the dissolved size fraction. These observations indicate that As mobility increased in the presence of DOM due to (I) competition between As and organic molecules for sorption sites on Fe particles; and (II) due to a higher amount of As bound to more abundant Fe colloids or complexes <0.2 μm in size. The amount of As contained in the colloidal size fractions also depended strongly on the initial size of the humic substance, which was larger for purified humic acids than for natural river or soil porewater samples. Arsenic in the particle size fraction >0.2 μm additionally decreased in the order of pH 4 ≫ 6 > 8. The presence of DOM likely increases the mobility of As in iron rich waters undergoing oxidation, a finding that has to be considered in the investigation of organic-rich terrestrial and aquatic environments.

Towards Regulatory Criteria for Discharging Iron-rich Mine Water into the Sea

Appropriate criteria are needed for regulating the discharge of mine water into the sea. Simply applying criteria developed for freshwater ecosystems to marine settings would be naive and highly inappropriate because marine organisms have very different mechanisms for coping with high concentrations of cations than do freshwater species. Furthermore, the hydrodynamics and geochemistry of marine waters mean that certain processes that are very important in determining the magnitude of mine water impacts in freshwater ecosystems are largely irrelevant in the oceans. This is particularly the case relative to benthic smothering by ochre, which does not occur when mine waters are discharged to the sea. Visual impacts of mine water discharge are still important in marine systems, due to the possibility of developing unsightly “slicks” of suspended ochre on the water surface (albeit these are actually innocuous in ecological terms). Avoidance of ochre slicks is a common concern wherever ferruginous mine waters are discharged to coastal waters. Compilation of data from a range of case studies indicates that no visible plume of ochre would be expected where the rate of iron release is less than about 200 kg/day. Maintenance of iron loadings below this critical threshold can be ensured by calculating a target maximum iron concentration (FeMAX, in mg/L) for the final effluent (which must be achieved by treatment if necessary) using the simple formula: FeMAX = 2,314.8/Q MAX, where Q MAX is the maximum anticipated flow rate in litres per second from the mined system (pumped or flowing by gravity).

Reprint of “Processes influencing dissolved iron distributions below the surface at the Atlantic Ocean–Celtic Sea shelf edge”

Shelf break systems are highly dynamic environments. However little is known about the influence that benthic interactions and water mass mixing may have on vertical distributions of iron in these systems. Dissolved Fe (< 0.4 μm) concentrations were measured in samples from nine vertical profiles across the upper slope (150–2950 m water depth) at the Atlantic Ocean–Celtic Sea shelf break. Dissolved iron concentrations varied between less than 0.2 and 5.4 nM, and the resulting detailed section showed evidence of a range of processes influencing the Fe distributions. The near sea floor data were interpreted in terms of release and removal processes. The concentrations of dissolved Fe present in near seabed waters were consistent with release of Fe from in situ remineralisation of particulate organic matter at two upper slope stations, and possibly release from pore water upon resuspension on shelf. Lateral transport of dissolved iron was evident from elevated Fe concentrations in an intermediate nepheloid layer and its advection along isopycnals. Surface waters at the shelf break also showed evidence of vertical mixing of deeper iron-rich waters. These waters contained macronutrients that sustained primary productivity in these otherwise nutrient-depleted surface waters. The data also suggest some degree of stabilisation of relatively high concentrations of iron, presumably through ligand association or as colloids. This study supports the view that lateral export of dissolved iron to the interior of the ocean from shelf and coastal zones and may have important implications for the global budget of oceanic iron.

Processes influencing dissolved iron distributions below the surface at the Atlantic Ocean–Celtic Sea shelf edge

Shelf break systems are highly dynamic environments. However little is known about the influence that benthic interactions and water mass mixing may have on vertical distributions of iron in these systems. Dissolved Fe (< 0.4 μm) concentrations were measured in samples from nine vertical profiles across the upper slope (150–2950 m water depth) at the Atlantic Ocean–Celtic Sea shelf break. Dissolved iron concentrations varied between less than 0.2 and 5.4 nM, and the resulting detailed section showed evidence of a range of processes influencing the Fe distributions. The near sea floor data were interpreted in terms of release and removal processes. The concentrations of dissolved Fe present in near seabed waters were consistent with release of Fe from in situ remineralisation of particulate organic matter at two upper slope stations, and possibly release from pore water upon resuspension on shelf. Lateral transport of dissolved iron was evident from elevated Fe concentrations in an intermediate nepheloid layer and its advection along isopycnals. Surface waters at the shelf break also showed evidence of vertical mixing of deeper iron-rich waters. These waters contained macronutrients that sustained primary productivity in these otherwise nutrient-depleted surface waters. The data also suggest some degree of stabilisation of relatively high concentrations of iron, presumably through ligand association or as colloids. This study supports the view that lateral export of dissolved iron to the interior of the ocean from shelf and coastal zones and may have important implications for the global budget of oceanic iron.

Analysis of environmental transcriptomes by DNA microarrays

In this work we investigated the correlations between global gene expression patterns and environmental parameters in natural ecosystems. We studied the preferential gene expression of the iron oxidizer bacterium Leptospirillum ferrooxidans to adapt its physiology to changes in the physicochemical parameters in its natural medium. Transcriptome analysis by DNA microarrays can proportionate an instant picture about the preferential gene expression between two different environmental samples. However, this type of analysis is very difficult and complex in natural ecosystems, mainly because of the broad biodiversity and multiple environmental parameters that may affect gene expression. The necessity of high-quality RNA preparations as well as complicated data analysis are also technological limitations. The low prokaryotic diversity of the extremely acidic and iron-rich waters of the Tinto River (Spain) ecosystem, where L. ferrooxidans is abundant, allows the opportunity to achieve global gene expression studies and to associate gene function with environmental parameters. We applied a total RNA amplification protocol validated previously for the amplification of the environmental transcriptome (meta-transcriptome). The meta-transcriptome of two sites from the Tinto River mainly differing in the salt and oxygen contents were amplified and analysed by a L. ferrooxidans DNA microarray. The results showed a clear preferential induction of genes involved in certain physicochemical parameters like: high salinity (ectAB, otsAB), low oxygen concentration (cydAB), iron uptake (fecA-exbBD-tonB), oxidative stress (carotenoid synthesis, oxyR, recG), potassium (kdpBAC) or phosphate concentrations (pstSCAB), etc. We conclude that specific gene expression patterns can be useful indicators for the physiological conditions in a defined ecosystem. Also, the upregulation of certain genes and operons reveals information about the environmental conditions (nutrient limitations, stresses, etc.).

Rapid natural acid weathering, physical erosion, and debris-flow hazards in scar areas developed on hydrothermally-altered rocks along the Red River Valley near Questa, New Mexico, USA

In southern Rocky Mountains, catchments characterized by acidic, metalliferous waters that are relatively unaffected by human activity usually occur within areas that have active or historical mining activity. The US Geological Survey has utilized these mineralized but unmined catchments to constrain geochemical processes that control the surfaceand ground-water chemistry associated with near surface acid weathering as well as to estimate premining conditions. Study areas include the upper Animas River watershed, Lake City, Mt. Emmons, and Montezuma in Colorado and Questa in New Mexico. Although host-rock lithologies range from Precambrian gneisses to Cretaceous sedimentary units to Tertiary volcanic complexes, mineralization is Tertiary in age and associated with intermediate to felsic composition, porphyritic plutons. Pyrite is ubiquitous. Variability of metal concentrations in water is caused by two main factors: mineralogy and hydrology. Parameters that potentially affect water chemistry include: host-rock lithology, intensity of hydrothermal alteration, sulfide mineralogy and chemistry, gangue mineralogy, length of flow path, precipitation, evaporation, and redox conditions. Springs and headwater streams have pH values as low as 2.5, sulfate up to 3700 mg/L and high dissolved metal concentrations (for example: A1 up to 170 rag/L; Fe up to 250 mg/L; Cu up to 3.5 mg/L and Zn up to 14 mg/L). With the exception of evaporative waters, the lowest pH values and highest Fe and A1 concentrations occur in water draining the most intense hydrothermally altered areas consisting of the mineral assemblage quartz-sericite-pyrite. Stream beds tend to be coated with iron floc, and some reaches are underlain by ferricrete. When iron-rich ground water interacts with oxygenated waters in the stream or hyporheic zone, ferrous iron is oxidized to ferric iron, which is less soluble, leading to the precipitation of iron oxyhydroxides. Ground-water wells have been drilled and sampled in two unmined, alpine catchments to characterize constituent concentrations, to identify hydrogeochemical processes controlling constituent concentrations, to determine rock hydraulic properties, and to delineate flow paths. By using an integrated approach to investigating surface and ground waters in these acidic catchments a more complete understanding of the hydrogeologic framework is gained.

Assessing the performance of phosphorus-saturated ochre as a fertilizer and its environmental acceptability

. Flooding of abandoned coal mines often causes discharges of iron-rich drainage water into the environment. Treatment of these discharges results in the formation of ochre (hydrous iron oxides) for which no end-use has been identified. Ochre effectively adsorbs phosphate from solution and thus could be used for remediation of waste waters. The resulting P-enriched ochre could then potentially be recycled as a P fertilizer. Pot and field experiments were set up to assess performance and environmental acceptability of ochre in this role, using grass and barley as test crops, as well as birch and spruce tree seedlings. Soils and plant materials were analysed for total and available P, total metals and pH. Results showed that P-saturated ochre functioned as a slow-release P fertilizer, and in the short term was as effective as conventional P fertilizer in maintaining crop yields. It also raised soil pH, and did not pose any significant problem through introduction of potentially toxic trace metals into the soil.

Preconcentration and determination of rare-earth elements in iron-rich water samples by extraction chromatography and plasma source mass spectrometry (ICP-MS)

A 100-fold preconcentration procedure based on rare-earth elements (REEs) separation from water samples with an extraction chromatographic column has been developed. The separation of REEs from matrix elements (mainly Fe, alkaline and alkaline-earth elements) in water samples was performed loading the samples, previously acidified to pH 2.0 with HNO 3, in a 2 ml column preconditioned with 20 ml 0.01 M HNO 3. Subsequently, REEs were quantitatively eluted with 20 ml 7 M HNO 3. This solution was evaporated to dryness and the final residue was dissolved in 10 ml 2% HNO 3 containing 1 μg l −1 of cesium used as internal standard. The solution was directly analysed by inductively coupled plasma mass spectrometry (ICP-MS), using ultrasonic nebulization, obtaining quantification limits ranging from 0.05 to 0.10 ng l −1. The proposed method has been applied to granitic waters running through fracture fillings coated by iron and manganese oxy-hydroxides in the area of the Ratones (Cáceres, Spain) old uranium mine.

Effect of acid mine drainage on the chemical composition and fall velocity of fine organic particles

Allochthonous fine organic particles (0.45 μm–1 mm, FPOM) are a significant natural source for biotic production of alkalinity in acidic, post-mining waters. Hence, the depositional characteristics and the nature of FPOM are of interest to remediation. Our objectives were firstly, to determine interactions between allochthonous FPOM and water affected by mine drainage and secondly, to verify the consequence on particle fall velocity, a major factor in particle distribution in streams and lakes. In laboratory experiments we studied interactions between particles (shredded leaves) and two types of post-mining water (very acidic, pH 3.7; extremely acidic, pH 2.9). Particle fall velocity was assessed by gravity sedimentation analysis. Suspended particles released minerals (1 mass%) and organic compounds (10 mass%) within 25 h. Fe 3+ was quickly reduced to Fe 2+ in aerobic conditions by organic reductants. Most of the iron attached to the particles (88% in very acidic and 71% in extremely acidic water) when the particle concentration was high. The median particle fall velocity was in acidic, iron-rich water up to 3 times higher than in distilled water. We assume particle aggregation in the highly mineralised water to be the reason for the increase in particle fall velocity.

Siderophores and the Dissolution of Iron-Bearing Minerals in Marine Systems

### Scope of this review Metal ions have critical functions in biological processes and provided important biological feedbacks with the environment throughout earth history. For example, Fe, Ni, Mg, Mn, Mo, Cu, W, V, and Zn play an essential role as catalysts in key compounds involved in respiration, photosynthesis, nitrogen fixation, and many other enzymatic processes (da Silva and Williams 2001). It is likely that some of these metal bearing enzymes evolved early in the history of life. However, the availability of metal ions has changed dramatically in the last 3.8 billion years due to changes in atmospheric and marine chemistry (Canfield 1998; Anbar and Knoll 2002; Saito et al. 2003). Homeostasis of metal ions (i.e., the maintenance of an approximately constant intracellular concentration) became a major problem over geologic time scales and in contemporary environments. Iron is an essential nutrient for almost all known organisms due to its important role in important enzymatic processes. While iron is the fourth most abundant element in the Earth crust, its low bioavailability limits primary production in various terrestrial and marine environments. The limitation of primary production in important ecosystems has significant implications for the global carbon cycle and the world climate. This review focuses on geochemical aspects of biological iron acquisition in iron limited “high nutrient low chlorophyll” (HNLC) ocean regions. The purpose of this review is to discuss the effect of biogenic iron specific ligands, the so called siderophores, on the iron speciation, and the dissolution of iron-bearing minerals in the presence of siderophores in these marine systems. Important iron sources for algal growth in HNLC ocean regions are the upward mixing of iron rich subsurface waters to the euphotic zone and the atmospheric deposition of dust particles on the sea surface followed by the dissolution of iron from these particles into the surface …

Scenario for undirected progression of extremely acidic streams in the Lusatian post-mining landscape, Germany

The environment in former open-pit brown coal dumps differs from natural conditions. This paper gives the most likely scenario of the long-term progression of streams planned in the re-deposited lowlands in Lusatia, Germany. The scenario is based on observations of streams in old abandoned mines. The stream's progression is controlled by non-cohesive virgin soils, extremely acidic iron-rich water, specific stream and bank vegetation, and incomplete in-stream decay of plant litter and wood. The scenario differs from progression of natural streams. It is a sequence from an open-land stream to a woodland stream and finally to a wetland stream. Key processes responsible for these transitions are strong deposition of iron oxyhydroxides, incomplete decay of leaves and wood, and reduced vitality of riparian trees. Uncertainties of the scenario and management conflicts that may emerge from undirected stream progression are briefly discussed.

Distribution of Fe in waters and bottom sediments of a small estuarine catchment, Pumicestone Region, southeast Queensland, Australia

Dissolved and extractable iron concentrations in surface water, groundwater and bottom sediments were determined for Halls Creek, a small subtropical tidally influenced creek. Dissolved iron concentrations were much higher in fresh surface waters and groundwater compared to the estuarine water. In bottom sediments, iron minerals were determined by X-ray diffraction (XRD); of these, hematite (up to 11%) has formed by precipitation from iron-rich water in the freshwater section of the catchment. Pyrite was only identified in the estuarine reach and demonstrated several morphologies [identified by scanning electron microscopy (SEM)] including loosely and closely packed framboids, and the euhedral form. The forms of pyrite found in bottom sediments indicate in situ production and recrystallisation. In surface waters, pyrite was detected in suspended sediment; due to oxygen concentrations well above 50 μmol/l, it was concluded that framboids do not form in the water column, but are within resuspended bottom sediments or eroded from creek banks. The persistence of framboids in suspended sediments, where oxygen levels are relatively high, could be due to their silica and clay-rich coatings, which prevent a rapid oxidation of the pyrite. In addition to identifying processes of formation and transport of pyrite, this study has environmental significance, as this mineral is a potential source of bioavailable forms of iron, which can be a major nutrient supporting algal growth.

Formation of mineral and thermal waters of some artesian basins in Russia

Artesian basins contain the largest mineral water resources of the world. There are several types of mineral therapeutic water: sulfate, chloride, radon-rich, iron-rich waters, etc. Artesian basins occupy very large areas in Russia. However, genesis of water and brines is still not very clear. This is one of the most important hydrogeological problems that is being attempted to solve for many years. Most of the Russian hydrogeologists traditionally consider that these waters are of sedimentary origin. However, higher concentrations of bromine, iodine, iron, radon and other balneologically active components can be of different origin, for example, of infiltration or juvenile water. As an example, two areas will be considered – West-Siberian basin and East-European artesian area.

Nutrient and Plankton Dynamics in the NE and NW Gyres of the Subarctic Pacific Ocean

The western subarctic gyre (WSG) and the eastern Alaska Grye (AG) on each side of the subarctic North Pacific, have many similarities. In both gyres, macronutrients are generally high and chl is low, and hence both gyres are High Nitrate, Low Chlorophyll (HNLC) regions. Despite the general similarities between these two gyres, there are many important differences. The time series station established at Stn KNOT on the southwest edge of the WSG and two in situ mesoscale iron enrichment experiments at each of the gyres has provided more information on iron concentrations, the dual role of iron and silicate limitation and seasonal cycles in the gyres. There is more seasonality in many parameters at Stn KNOT than at Stn P. There is an increase in Chl and primary productivity at Stn KNOT in May followed by increased iron limitation in summer. Low DIC:NO3 ratios and high Si:NO3 ratios in the WSG, indicate lower calcification and higher diatom production than at Stn P. The sources of iron for these areas are still not clear, but horizontal transport of iron rich coastal water and vertical transport could be important sources at certain times of the year in addition to dust input. Satellite images show that chl-rich coastal waters occasionally extend to the vicinity of Stn KNOT and therefore Stn KNOT may not always represent conditions in the main part of the WSG. This review focuses mainly on a comparison of Stn KNOT and Stn P, two time series stations on the edge of two very large gyres. At present, we have a limited understanding of the temporal and spatial variability within each of these large gyres.

IC-ICP-MS and IC-ICP-HEX-MS determination of arsenic speciation in surface and groundwaters: preservation and analytical issues

Understanding the biogeochemical behaviour of arsenic in the weathering and shallow subsurface environment depends critically upon determining the nature and distribution of the chemical species present in natural waters. To this end, coupled ion chromatography-inductively coupled plasma mass spectrometry (IC-ICP-MS) is widely used, though species fractionation during ultrasonic nebulization, and matrix-dependent ionization in the plasma are analytical issues that need to be addressed. Hexapole collision cell technology is shown to be effective in suppressing chloride-based polyatomic interferences. Irrespective of the analytical technique used, As(III)/As(V) ratios of natural waters may change substantially during storage due to (1) differential adsorption of arsenic species on hydrated ferric oxides (HFOs); and (2) microbial activity. A wide range of apparently contradictory speciation changes observed by various workers can be rationalized in terms of the differences of microbial consortia present in different water samples. Arsenic speciation in certain water types can be stabilized for days or even weeks by combined filtration, acidification and refrigeration whilst the addition of EDTA and the use of 0.1 mm filters is indicated for iron-rich waters and waters with high activities of redox-active bacteria, respectively. Although the use of hydrochloric acid has been reported elsewhere as resulting in the apparent oxidation of As(III), we show that for certain water types it acts as an extremely effective preservative of arsenic speciation.

Investigation on stability and preservation of arsenic species in iron rich water samples

Important for the accurate and reproducible determination of inorganic redox forms of arsenic in iron-rich waters is their conservation prior to analysis. Species and trace element analysis methods are commonly laboratory based. Stabilisation of samples is necessary for subsequent laboratory analysis in order to preserve the information about the system from which the samples were taken. Pre-treatment procedures based on complexation of metal ions and moderate acidification of the samples are presented. The concentration of the stabilisation agent and the storage temperature were optimised. The most satisfactory results were obtained with 0.01 mol l −1 phosphoric acid and a storage at 6 °C.

Iron in the western Pacific: a riverine or hydrothermal source for iron in the Equatorial Undercurrent?

Upwelling of the (comparatively) iron-rich waters of the Equatorial Undercurrent (EUC) may control new production in the central and eastern equatorial Pacific. In an attempt to determine the source of this iron, we measured total dissolvable iron (Fe TD) concentrations in the Bismarck Sea on three cruises of R.V. Franklin as part of the Australian Joint Global Ocean Flux Study (JGOFS) program in 1993 and the Tropical River Ocean Processes In Coastal Settings (TROPICS) program in 1997 and 2000. Data in the Bismarck Sea were collected during the SE tradewind season at a time of a strong El Niño and severe drought in Papua New Guinea (PNG) (FR07/97), and during the NW monsoon season (FR01/00). Our results show that the New Guinea Coastal Undercurrent (NGCU) and the EUC in the western equatorial Pacific are enriched in iron that does not seem to be supplied directly from the Sepik River nor from known deep hydrothermal sources in the Bismarck Sea. The concentration of Fe TD in the NGCU does increase as it flows along the north coast of PNG and we propose that the Fe is mainly supplied by sediments deposited on the narrow continental shelf by the Sepik and other rivers. There is indirect evidence for an additional supply of iron via the New Ireland Coastal Undercurrent.