Description: Here we demonstrate the use of reverse titration - competitive ligand exchange-adsorptive cathodic stripping voltammetry (RT-CLE-ACSV) for the analysis of iron (Fe) binding ligands in seawater. In contrast to the forward titration, which examines excess ligands in solution, RT-CLE-ACSV examines the existing Fe-ligand complexes by increasing the concentration of added (electroactive) ligand (1-nitroso-2-naphthol) and analysis of the proportion of Fe bound to the added ligand. The data manipulation allows the accurate characterisation of ligands at equal or lower concentrations than Fe in seawater, and disregards electrochemically inert dissolved Fe such as some colloidal phases. The method is thus superior to the forward titration in environments with high Fe and low ligand concentrations or high concentrations of inert Fe.We validated the technique using the siderophore ligand ferrioxamine B, and observed a stability constant K'Fe3+FoB of 0.74-4.37×1021mol-1, in agreement with previous results. We also successfully analysed samples from coastal waters and a deep ocean hydrothermal plume. Samples from these environments could not be analysed with confidence using the forward titration, highlighting the effectiveness of the RT-CLE-ACSV technique in waters with high concentrations of inert Fe. © 2013 Elsevier B.V.

Description: On voyages in the Iceland Basin in 2007 and 2009, we observed low (ca. 0.1nM) total dissolved iron concentrations dFe in surface waters (〈150m), which increased with depth to ca. 0.2-0.9nM. The surface water dFe was low due to low atmospheric Fe inputs combined with biological uptake, with Fe regeneration from microbial degradation of settling biogenic particles supplying dFe at depth. The organic ligand concentrations LT in the surface waters ranged between 0.4 and 0.5nM, with conditional stability constants (logK'FeL) between 22.6 and 22.7. Furthermore, LT was in excess of dFe throughout the water column, and dFe was therefore largely complexed by organic ligands (>99%). The ratio of LT/dFe was used to analyse trends in Fe speciation. Enhanced and variable LT/dFe ratios ranging between 1.6 and 5.8 were observed in surface waters; the ratio decreased with depth to a more constant LT/dFe ratio in deep waters. In the Iceland Basin and Rockall Trough, enhanced LT/dFe ratios in surface waters resulted from decreases in dFe, likely reflecting the conditions of Fe limitation of the phytoplankton community in the surface waters of the Iceland Basin and the high productivity in the Rockall Trough.Below the surface mixed layer, the observed increase in dFe resulted in a decrease of the LT/dFe ratios (1.2-2.6) with depth. This indicated that the Fe binding ligand sites became occupied and even almost saturated at enhanced dFe in the deeper waters. Furthermore, our results showed a quasi-steady state in deep waters between dissolved organic Fe ligands and dFe, reflecting a balance between Fe removal by scavenging and Fe supply by remineralisation of biogenic particles with stabilisation through ligands. © 2011 Elsevier Ltd.

Description: As part of the PREDICT Tamar Workshop, the toxicity of estuarine waters in the Tamar Estuary (southwest England) was assessed by integration of metal speciation determination with bioassays. High temporal resolution metal speciation analysis was undertaken in situ by deployment of a Voltammetric In situ Profiling (VIP) system. The VIP detects Cd (cadmium), Pb (lead) and Cu (copper) species smaller than 4 nm in size and this fraction is termed 'dynamic' and considered biologically available. Cadmium was mainly present in the dynamic form and constituted between 56 and 100 of the total dissolved concentration, which was determined subsequently in the laboratory in filtered discrete samples. In contrast, the dynamic Pb and Cu fractions were less important, with a much larger proportion of these metals associated with organic ligands and/or colloids (45-90 Pb and 46-85 Cu), which probably reduced the toxicological impact of these elements in this system. Static toxicity tests, based on the response of Crassostrea gigas larva exposed to discrete water samples showed a high level of toxicity (up to 100 abnormal development) at two stations in the Tamar, particularly during periods of the tidal cycle when the influence of more pristine coastal water was at its lowest. Competitive ligand-exchange Cu titrations showed that natural organic ligands reduced the free cupric ion concentration to levels that were unlikely to have been the sole cause of the observed toxicity. Nonetheless, it is probable that the combined effect of the metals determined in this work contributed significantly to the bioassay response. © 2011 Elsevier Ltd.

Description: Siderophore type chelates were detected in nutrient enriched, incubated seawater collected from different biogeographical regions of the Atlantic Ocean. Seawater was enriched with glucose and ammonium, glycine (as a source of carbon and nitrogen) or chitin and ammonium at different concentrations and was incubated for up to 3-4. days in the dark. Siderophore type chelates were detected using high performance liquid chromatography coupled to inductively coupled plasma mass spectrometry (HPLC-ICP-MS) after complexation with Ga. Samples were subsequently analysed by HPLC-electrospray ionisation mass spectrometry (HPLC-ESI-MS) in order to confirm the identity of the known siderophores, and to obtain the pseudo-molecular ions of unknown siderophore type chelates. A total of 22 different siderophore type chelates were resolved in the HPLC-ICP-MS chromatograms. Ten different siderophore type chelates were identified by HPLC-ESI-MS, 3 of which had not previously been identified in nutrient enriched seawater incubations. The concentration and diversity of siderophore type chelates was highest in seawater amended with glucose. The concentrations and diversity of siderophore type chelates also varied with biogeographical area in the Atlantic Ocean, with the North Atlantic Sub-tropical Gyre yielding highest concentrations in incubations, and the South Atlantic Sub-tropical Gyre and Western Tropical Atlantic yielding the highest diversity

Description: The distribution of size fractionated dissolved iron (DFe, 〈0.2 μm) species was determined in the upper water column (0-150 m) of the Canary Basin (25-32°N and 18-24°W) on a research cruise in October 2002. A DFe concentration gradient resulting from a decrease in both soluble iron (SFe, 〈0.02 μm) and colloidal iron (CFe, 0.02-0.2 μm) was shown to extend from the coast of North West Africa into the oligotrophic gyre (varying from ∼1 nM in the shelf region to 0.15 nM in the most off shore waters). At the time of this study, the dominant dissolved Fe input to the region was deduced to be the advection of shelf and upwelled waters rather than Saharan dust deposition.SFe and CFe fractions had mean concentrations (± one standard deviation) of 0.25 ± 0.11 and 0.21 ± 0.16 nM, respectively (n = 58). Colloidal iron formed a highly variable fraction of DFe (ca. 0-80, mean of 42) in the region but was less variable in the low iron, oligotrophic intermediate waters (0.18 ± 0.06 nM, 31.7°N, 22.0°W, 0-1300 m depth). The high variability found at the most productive near-shelf stations was driven by biological processing and mixing of different water masses. In contrast, less variability between SFe and CFe at the remote off shore stations suggested that vertical variations in the water column were controlled more by chemical partitioning and vertical particle fluxes with evidence of preferential biological uptake and/or removal of SFe in the most remote surface waters. © 2010.

Description: Iron (Fe) binding phases in two hydrothermal plumes in the Southern Ocean were studied using a novel voltammetric technique. This approach, reverse titration–competitive ligand exchange–adsorptive cathodic stripping voltammetry, showed that on average 30±21% of dissolved Fe in the hydrothermal plumes was stabilised by chemically labile binding to ligands. The conditional stability constant (log K′FeL) of the observed complexes was 20.61±0.54 (mean±1 SD) for the two vent sites, intermediate between previous measurements of deep ocean ligands (21.4–23; Kondo et al., 2012) and dissolved weak estuarine ligands (〈20; Gerringa et al., 2007). Our results indicate that approximately 7.5% of all hydrothermal Fe was stabilised by complexation with ligands. Furthermore, 47±26% of the dissolved Fe in the plume existed in the colloidal size range (0.02–0.2 µm). Our data suggests that a portion (∼7.5%) of hydrothermal Fe is sufficiently stabilised in the dissolved size fraction (〈0.2 µm) to make an important impact on deep ocean Fe distributions. Lateral deep ocean currents transport this hydrothermal Fe as lenses of enhanced Fe concentrations away from mid ocean ridge spreading centres and back arc basins.

Description: High quality carbonate chemistry measurements are required in order to fully understand the dynamics of the oceanic carbonate system. Seawater pH data with good spatial and temporal coverage are particularly critical to apprehend ocean acidification phenomena and their consequences. There is a growing need for autonomous in situ instruments that measure pH on remote platforms. Our aim is to develop an accurate and precise autonomous in situ pH sensor for long term deployment on remote platforms. The widely used spectrophotometric pH technique is capable of the required high-quality measurements. We report a key step towards the miniaturization of a colorimetric pH sensor with the successful implementation of a simple microfluidic design with low reagent consumption. The system is particularly adapted to shipboard deployment: high quality data was obtained over a period of more than a month during a shipboard deployment in northwest European shelf waters, and less than 30 mL of indicator was consumed. The system featured a short term precision of 0.001 pH (n = 20) and an accuracy within the range of a certified Tris buffer (0.004 pH). The quality of the pH system measurements have been checked using various approaches: measurements of certified Tris buffer, measurement of certified seawater for DIC and TA, comparison of measured pH against calculated pH from pCO2, DIC and TA during the cruise in northwest European shelf waters. All showed that our measurements were of high quality. The measurements were made close to in situ temperature (+0.2 ̊C) in a sampling chamber which had a continuous flow of the ship’s underway seawater supply. The optical set up was robust and relatively small due to the use of an USB mini-spectrometer, a custom made polymeric flow cell and an LED light source. The use of a three wavelength LED with detection that integrated power across the whole of each LED output spectrum indicated that low wavelength resolution detectors can be used instead of the current USB mini spectrophotometer. Artefacts due to the polychromatic light source and inhomogeneity in the absorption cell are shown to have a negligible impact on the data quality. The next step in the miniaturization of the sensor will be the incorporation of a photodiode as detector to replace the spectrophotometer.

Description: In addition to ocean acidification due to a gradual anthropogenic CO2 uptake, strong seasonal variations in the carbonate system occur in the Arctic Ocean as a result of physical and biological processes. Understanding this seasonal variability is critical for predicting the onset of calcium carbonate mineral (Ω) undersaturation with increasing atmospheric CO2 concentrations. However, these variations are currently poorly understood because of a lack of winter data due to the challenging field conditions in this season. Here we report observations over an annual cycle of the carbonate system of surface waters in the Atlantic gateway to the Arctic Ocean, covering the region between Svalbard and mainland Norway. Dissolved inorganic carbon (DIC) concentrations ranged from 2137–2148 μmol kg− 1 in winter to 1986–2094 μmol kg− 1 in summer, and total alkalinity (TA) concentrations between 2312–2341 μmol kg− 1 in winter and 2199–2317 μmol kg− 1 in summer. This resulted in an increase in TA:DIC ratios from 1.077–1.090 in winter to 1.106–1.112 in summer, mainly due to the biological uptake of CO2 during spring and summer. Similarly, a significant seasonal variability was observed in Ω (0.4–0.9), with lowest saturation states in winter (Ωaragonite ~ 1.8–2.1) and highest in spring and summer (Ωaragonite ≈ 2.4). Analysis of the biogeochemical and physical processes that impact aragonite saturation states (Ωar) showed biological production to be the most important factor driving seasonal variability in Ωar in this area, accounting for 45–70% of the difference between winter and summer values. Future changes in these processes may alter the seasonal cycle of the carbonate system in both amplitude and timing, and further observations are required to determine the progress of ocean acidification in the Atlantic waters entering the Arctic Ocean.

Description: The Southern Ocean is an important biogeochemical region on a global scale, in which mineralising phytoplankton play a role in cycling energy, carbon and nutrients. Mineralising phytoplankton with cells 2-20. μm in diameter (nannoplankton) are poorly enumerated by traditional preservation and microscopy techniques, yet may fulfil an important role in the Southern Ocean. Here we define the spatial and temporal biogeography for these mineralising nannoplankton assessed by scanning electron microscopy in conjunction with an array of biological, physical, and chemical variables during two cruises to the Scotia Sea region of the Southern Ocean. The cruises encompassed two seasons, austral summer (January-February 2008) and austral autumn (March-April 2009).The biogeography of the three most numerous mineralising nannoplankton groups, the coccolithophore Emiliania huxleyi, the smaller (〈10μm) species of the diatom genus Fragilariopsis, and chrysophytes of the genus Tetraparma (mostly Tetraparma pelagica) were found to be related to the boundaries of the major circumpolar fronts. E. huxleyi abundances were relatively high in the northern water masses (maximum of 650cellsml -1), while T. pelagica abundances were high in the southern water masses (maximum of 1910cellsml -1). Small Fragilariopsis spp. abundances were also highest in the southern water masses (maximum of 1820cellsml -1), but this group was present throughout the Scotia Sea.Multivariate statistical analysis found that the most influential environmental variables controlling mineralising nannoplankton biogeography were sea surface temperature and silicate concentration. Estimates of biomass indicated that the Scotia Sea mineralising nannoplankton community formed a substantial part of the total phytoplankton community, particularly south of the Southern Antarctic Circumpolar Current Front (SACCF) during the austral autumn, where mineralising nannoplankton biomass reached 36 of the total phytoplankton biomass. The results that are obtained suggest that traditional microscopic surveys of large Southern Ocean phytoplankton may underestimate total biomass by excluding key mineralising nannoplankton groups. Greater appreciation of the ecological significance of mineralising nannoplankton in the Southern Ocean will improve our understanding of the relationships between environmental parameters, primary production, and the biological carbon pump in this ecosystem. © 2011 Elsevier Ltd.

Description: We report a high performance autonomous analytical system based on the vanadomolybdate method for the determination of soluble reactive phosphorus in seawater. The system combines a microfluidic chip manufactured from tinted poly (methyl methaaylate) (PMMA), a custom made syringe pump, embedded control electronics and on-board calibration standards. This "lab-on-a-chip" analytical system was successfully deployed and cross-compared with reference analytical methods in coastal (south west England) and open ocean waters (tropical North Atlantic). The results of the miniaturized system compared well with a reference bench-operated phosphate auto-analyser and showed no significant differences in the analytical results (student's t-test at 95% confidence level). The optical technology used, comprising of tinted PMMA and polished fluidic channels, has allowed an improvement of two orders of magnitude of the limit of detection (52 nM) compared to currently available portable systems based on this method. The system has a wide linear dynamic range 0.1-60 mu M, and a good precision (13.6% at 0.4 mu M, n=4). The analytical results were corrected for silicate interferences at 0.7 mu M, and the measurement frequency was configurable with a sampling throughput of up to 20 samples per hour. This portable micro-analytical system has a low reagent requirement (340 mu L per sample) and power consumption (756J per sample), and has allowed accurate high resolution measurements of soluble reactive phosphorus in seawater