Overview of the project and objectives: Assessing the contribution of the different N substrates to the primary production process, such as the biogenic silica production and dissolution in the Antarctic sea-ice provides a means to understand the biogeochemical system functioning. In such a semi closed-type system, assess incorporation rates of HCO3-, NO3-, NH4+, SiOH4, BSi dissolution, nitrification, C-release in close-by ice-cores (3 ice-cores dedicated to (i) 13C-assimilation + 15NH4+ uptake rate, (ii) 13C-assimilation + 15NO3- uptake rate and nitrification, (iii) Biogenic silica production and dissolution via 30Si isotope tool) will allow improving the knowledge of system functioning. This is also closely linked to the thematic of iron availability since these experiments are done close to / on the Trace Metal site allowing us to hopefully propose a relatively complete image of biogeochemical activity and relationship with trace metals on this site. Methodology and sampling strategy: Most of the time we worked close to / directly on the Trace Metal site following precautions concerning TM sampling (clean suits etc.). When we worked close to the TM site, precautions were not such important because we don't need the same drastic precautions for our own sampling. We work together because we want to propose a set of data which helps to characterize the system of functioning in close relation with TM availability (for that, sampling location have to be as close as possible). 14C and 13C-incubation experiment intercalibration work were conducted on the Biosite (different place than TM site except for station 7) Incubation experiment samples are analyzed via (1) Elemental Analyzer - Isotope Ratio Mass Spectrometer (EA-IRMS) for carbon and nitrogen (VUB, Brussels, Belgium); (2) High Resolution Inductively Coupled Mass Spectrometer (HR-ICPMS) for silicon (RMCA, Brussels, Belgium).

Metadata record for data from AAS Project 3127 See the link below for public details on this project. Bacteria in marine environments have been found to be able to partially support growth by using light to generate energy in a non-photosynthetic process. This is possible due to a special protein called proteorhodopsin. It is hypothesised that formation of proteorhodopsin has evolved to cope with extreme lack of nutrients. The goal is to determine the significance of proteorhodopsins in the productivity of Southern Ocean microbial communities. This includes determination of proteorhodopsin distribution, presence in seawater and sea-ice samples using molecular techniques, and determination of how important environmental factors (light, nutrient availability, temperature) may drive its synthesis and activity. Taken from the 2009-2010 Progress Report Project objectives: 1. Determine incidence of proteorhodopsins in Southern Ocean water and sea-ice derived bacteria (Year 1) and other Antarctic aquatic environments (Year 2 and 3). 2. Determine whether proteorhodopsins contribute to food web energy budgets. 3. Determine how proteorhodopsin contributions are influenced by physicochemical features of the environment including light availability, temperature and nutrients. Progress against objectives: Proteorhodopsin is a light harvesting membrane protein that has been found recently to occur in 30-70% of marine bacterial cells. The role of this protein is uncertain but believed to be highly important in energy and nutrient budgets in food webs as it is capable of generating a proton gradient. Amongst a cultured set of Antarctic bacteria we have discovered many PR-producing species. These include many Antarctic lake species. Research is ongoing to determine affect of light on the physiology of these bacteria in particular the genome sequenced species Psychroflexus torquis, an extremely cold-adapted resident of Antarctic sea-ice. 1. Completed screen of Antarctic bacterial collection for proteorhodopsin (PR) genes using PCR-based approaches 2. Proteomic-based analysis of PR-bearing sea-ice species Psychroflexus torquis is currently ongoing 3. Light/dark defined growth-based experiments determining conditions leading to biomass enhancement are ongoing

Oceanographic processes in the subantarctic region contribute crucially to the physical and biogeochemical aspects of the global climate system. To explore and quantify these contributions, the Antarctic Cooperative Research Centre (CRC) organised the SAZ Project, a multidisciplinary, multiship investigation carried out south of Australia in the austral summer of 1997-1998. Taken from the abstracts of the referenced papers: In March 1998 we measured iron in the upper water column and conducted iron- and nutrient-enrichment bottle-incubation experiments in the open-ocean Subantarctic region southwest of Tasmania, Australia. In the Subtropical Convergence Zone (~42 degrees S, 142 degrees E), silicic acid concentrations were low (less than 1.5 micro-M) in the upper water column, whereas pronounced vertical gradients in dissolved iron concentration (0.12-0.84 nM) were observed, presumably reflecting the interleaving of Subtropical and Subantarctic waters, and mineral aerosol input. Results of a bottle-incubation experiment performed at this location indicate that phytoplankton growth rates were limited by iron deficiency within the iron-poor layer of the euphotic zone. In the Subantarctic water mass (-46.8 degrees S, 142 degrees E), low concentrations of dissolved iron (0.05-0.11 nM) and silicic acid (less than 1 micro-M) were measured throughout the upper water column, and our experimental results indicate that algal growth was limited by iron deficiency. These observations suggest that availability of dissolved iron is a primary factor limiting phytoplankton growth over much of the Subantarctic Southern Ocean in the late summer and autumn. The importance of resource limitation in controlling bacterial growth in the high-nutrient, low-chlorophyll (HNLC) region of the Southern Ocean was experimentally determined during February and March 1998. Organic- and inorganic-nutrient enrichment experiments were performed between 42 degrees S and 55 degrees S along 141 degrees E. Bacterial abundance, mean cell volume, and [3H]thymidine and [3H]leucine incorporation were measured during 4- to 5-day incubations. Bacterial biomass, production, and rates of growth all responded to organic enrichments in three of the four experiments. These results indicate that bacterial growth was constrained primarily by the availability of dissolved organic matter. Bacterial growth in the subtropical front, subantarctic zone, and subantarctic front responded most favourably to additions of dissolved free amino acids or glucose plus ammonium. Bacterial growth in these regions may be limited by input of both organic matter and reduced nitrogen. Unlike similar experimental results in other HNLC regions (subarctic and equatorial Pacific), growth stimulation of bacteria in the Southern Ocean resulted in significant biomass accumulation, apparently by stimulating bacterial growth in excess of removal processes. Bacterial growth was relatively unchanged by additions of iron alone; however, additions of glucose plus iron resulted in substantial increases in rates of bacterial growth and biomass accumulation. These results imply that bacterial growth efficiency and nitrogen utilisation may be partly constrained by iron availability in the HNLC Southern Ocean. The download file also contains three excel spreadsheets of iron data from the project. The file Sedwick_A9706_Fe_data contains water-column dissolved Fe and total-dissolvable Fe data from cruise A9706, which is presented in Sedwick et al. (1999) and Sedwick et al. (2008). The files Sedwick_A9706_ProcessStn1_Exp_data and Sedwick_A9706_ProcessStn2_Exp_data present data from shipboard experiments conducted during cruise A9706 at Process Stations 1 and 2, respectively, as reported in Sedwick et al. (1999).

Oceanographic processes in the subantarctic region contribute crucially to the physical and biogeochemical aspects of the global climate system. To explore and quantify these contributions, the Antarctic Cooperative Research Centre (CRC) organised the SAZ Project, a multidisciplinary, multiship investigation carried out south of Australia in the austral summer of 1997-1998. Taken from the abstracts of the referenced paper: We developed and applied a one-dimensional (z) biophysical model to the Subantarctic Zone (SAZ) and the Polar Frontal Zone (PFZ) to simulate seasonal phosphate export production and resupply. The physical component of our model was capable of reproducing the observed seasonal amplitude of sea surface temperature and mixed layer depth. In the biological component of the model we used incident light, mixed layer depth, phosphate availability, and estimates of phytoplankton biomass from the Sea-viewing Wide Field-of-view Sensor to determine production and tuned the model to reproduce the observed seasonal cycle of phosphate. We carried out a series of sensitivity studies, taking into account uncertainties in both physical fields and biological formulations (including potential influence of iron limitation), which led to several robust conclusions (as represented by the ranges below). The major growing season contributed 66-76% of the annual export production in both regions. The simulated annual export production was significantly higher in the PZF (68-83 mmol P m-2) than in the SAZ (52-61 mmol P m-2) despite the PFZ's having lower seasonal nutrient depletion. The higher export production in the PFZ was due to its greater resupply of phosphate to the upper ocean during the September to March period (27-37 mmol P m-2) relative to that in the SAZ (8-15 mmol P m-2). Hence seasonal nutrient depletion was a better estimate of seasonal export production in the SAZ, as demonstrated by its higher ratio of seasonal depletion/export (64-78%) relative to that in the PFZ (34-47%). In the SAZ, vertical mixing was the dominant mechanism for supplying phosphate to the euphotic zone, whereas in the PFZ, vertical mixing supplied only 37% of the phosphate to the euphotic zone, whereas in the PFZ, vertical mixing supplied only 37% of the phosphate to the euphotic zone and horizontal transport supplied the remaining 63%.

Metadata record for data from ASAC Project 2784 See the link below for public details on this project. This project utilised an existing 55 year model reanalysis (SODA) - so no new models were developed. The methodologies/data used are described in the referenced publications. Modelling investigations of the shoaling of iron-rich upper circumpolar deep water (UCDW) and its role in the regulation of primary production at 60-65S. Taken from the project application: We intend to utilise a number of existing data sources to study the factors leading to spatiotemporal variability in the upwelling of iron-rich UCDW in the 60-65S zone, which, as discussed above, seems critical to regional ecosystem function, and the carbon and sulphur budgets of the SO. As sea-ice extent appears to have declined in the Southern Ocean since the 1950s (Curran et al., 2003) it will also be extremely interesting to examine whether this has had any affect on the upwelling of the UCDW. Given the restricted spatial domain of in situ field data in the Southern Ocean, satellite products provide us with one of the few means to investigate coherent variability over large spatial and temporal scales. This study takes advantage of our previous AAS funded work (Projects: 2584, 2319), where we have gained considerable experience in the coupling of biogeochemical and climate models and where we have already assembled satellite data sets on wind speed, sea-ice, SST, aerosols and chlorophyll-a concentration. This previous experience will allow us to examine the relationship between the physical forcings, the dynamics of the UCDW and the biological response on seasonal and interannual timescales over the period 1950-2000. The key scientific questions we seek to answer include: - What is the range of interannual and interdecadal variability in upwelling of the UCDW and how does this relate to variability in primary production? - Is there a connection between interannual/decadal variability in sea-ice extent and the strength or location of upwelling of UCDW and hence the character of regional primary production? - Is there a relation between the seasonal production of DMS and associated S-aerosols and the dynamics of UCDW? Details from previous years are available for download from the provided URL. Taken from the 2009-2010 Progress Report: Progress against objectives: This three-year project has been investigating the nexus between the large-scale meridional circulation patterns in the SO, in particular UCDW upwelling, and concomitant iron delivery to surface waters and the phytoplankton. Key Scientific Questions to be considered by the project What is the range of interannual and inter-decadal variability in upwelling of the UCDW and how does this relate to variability in primary production? This study initially focussed on the Australian region of the Southern Ocean (110-160 degrees S, 40-70 degrees E) and the physical oceanographic data for the project came from monthly Simple Ocean Data Assimilation (SODA) reanalysis data, which covers the period 1958-2007 over the global ocean. Decadal-scale trends in upper ocean structure and meridional circulation were analysed, including the upwelling of nutrient-rich UCDW, and these results were initially documented in presentation (3) below and will shortly be published in publication (1) listed below. The project identified UCDW in SODA using temperature and density criteria and, using this, a number of variables were developed to characterise UCDW and its upwelling: UCDW vertical velocity, temperature, density and salinity, UCDW top depth (the shallowest depth at which UCDW is found) and UCDW southern-most position. Climatological values were found for each of the 5-degree strips in the sector and, in addition, trends were found over the period 1958-2005. Later work involved comparing these results with those of two more Southern Ocean sectors - one in the Pacific (130-80 degrees W) and one in the Indian Ocean (20-60 degrees E). These results were presented at the AMOS conference in January 2010 (see Presentation (1) below) and are also the subject of a paper in the Proceedings of that conference (see Publication (2) below). It was found that during 1958-2005: (1) UCDW top depth varies seasonally, peaking in March, and displays considerable interannual variability; (2) Climatological properties for UCDW variables such as temperature, vertical velocity and upwelling depth vary between the three ocean sectors, as do trends (1958-2005) in the UCDW variables; (3) UCDW vertical velocity (ie. upwelling) appears to be increasing with time in most intermediate and deep waters of the three ocean sectors; (4) UCDW temperature is increasing in intermediate waters in the Pacific sector, at all depths in the Indian sector and at shallow depths in the Australian sector, but is decreasing in intermediate and deep waters in the Australian sector; (5) UCDW southern-most position is moving south in the Australian and Pacific sectors; (6) UCDW is upwelling closer to the surface in the Australian and Indian sectors and, in the case of the Australian sector, this translates into an increase in the number of times that UCDW can be detected in the mixed layer (a finding of possible importance for primary production); (7) UCDW trends in the Australian sector do not appear to be affected by trends in the winds, but by forcings acting on longer than decadal time-scales. This is not the case, however, for the other two sectors, leading to the speculation that these variables may be affected by the re-entry into UCDW of recirculated waters from the Indian and Pacific Oceans, which may themselves be affected by winds. (8) The Australian sector of the SO has been shown to have its own unique characteristics, distinct from either the Pacific or Indian sectors. More recent work has involved looking at the initial Australian sector considered above, over the period of the high resolution satellite data capture era (1997-2007), with the aim of using satellite data on chlorophyll a (chl a), sea-ice concentration and photosynthetically active radiation (PAR), as well as modelled data for primary production (PP), in addition to the reanalysis data, to look at factors that influence chl a and PP over that time period. Initial work was presented at the AMOS conference in January 2009 (see Presentation (2) below) and final work is reported in Publication (3) listed below, which is almost ready for submission. It was found that in the Australian sector during 1997-2007: (1) The most important controls on chl a in spring are sea-ice concentration and PAR in the southern-most zones (and mixed layer depth, SST, stratification and PAR in zones further north); (2) The situation is more complex in summer, especially in the southern-most zones (the areas of highest production, excluding the most northerly zone near Tasmania). In particular, in the 60-65 degrees S zone in summer, a variety of inter-acting controls affect chl a (and PP), including SST, stratification and UCDW top depth; (3) The number of times that UCDW is detected in the mixed layer is decreasing in summer during 1997-2007; (4) It is difficult to identify trends that are statistically significant over such a short time period and trends that are found are often opposite in sign to those for 1958-2005 and up to an order of magnitude larger. Thus care needs to be taken with trends found for chl a, PP and hydrodynamic variables over the short period of the satellite era, since there is a large range of such ten-year trends in the period 1958-2005. Is there a connection between interannual/decadal variability in sea-ice extent and the strength or location of upwelling of UCDW and hence the character of regional primary production? Given that UCDW upwells south of the Polar Front and no further south than the Southern Boundary of the ACC (approximately 65 degrees S in this sector), then UCDW, as identified here in its pure form, is not able to affect the 65-70 degrees S zone (although this is possible in its modified form, which is not studied here). It was found that, for the period 1997-2007 in the Australian sector of the SO, the southern-most position of UCDW is not correlated with sea-ice concentration, but that there are weak (90% level) correlations in 60-65 degrees S between UCDW top depth and sea-ice concentration in autumn (positive), the temperature of UCDW and sea-ice concentration in summer (positive) and northward Ekman transport and sea-ice concentration in summer (negative). It was found that, for 1997-2007 in the Australian sector of the SO, sea-ice concentration has a significant (inverse) relationship with chl a and PP in 60-70 degrees S in spring and 65-70 degrees S in summer. In addition, UCDW top depth and northward Ekman transport (ie. how quickly the UCDW nutrients are transported northwards and away from the zone) have a minor effect on chl a in 60-65 degrees S in summer.

Antarctic sea ice is known to store key micronutrients, such as iron, as well as a suite of less studied trace metals in winter which are rapidly released in spring. This stimulates ice edge phytoplankton blooms which drive the biological removal of climatically-important gases like carbon dioxide. By linking the distribution of iron and other trace elements to the cycles of carbon, nitrogen and silicon in the sea ice zone in spring, this project will identify their biogeochemical roles in the seasonal ice zone and how this may change with predicted climate-driven perturbations. All sampling bottles and equipment were decontaminated using trace metal clean techniques. Care was taken at each site to select level ice with homogeneous snow thickness. At all the stations, the same sampling procedure has been used : Firstly, snow was collected using acid cleaned low density polyethylene (LDPE) shovels and transferred into acid-cleaned 3.8 l LDPE containers (Nalgene). Snow collected is analysed for temperature, salinity, nutrients, unfiltered and filtered metals. Snow thickness is recorded with a ruler. Ice cores were collected using a noncontaminating, electropolished, stainless steel sea ice corer (140 mm internal diameter, Lichtert Industry, Belgium) driven by an electric power drill. Ice cores were collected about 10 cm away from each other to minimise between-core heterogeneity. A first core is dedicated to the temperature, salinity and Chlorophyll a (Chla). To record temperature, a temperature probe (Testo, plus or minus 0.1 degrees C accuracy) was inserted in holes freshly drilled along the core every 5 to 10 cm, depending on its length. Bulk salinity was measured for melted ice sections and for brines using a YSI incorporated Model 30 conductivity meter. Chla is processed on board using a 10 AU fluorometer (turner Designs, Sunnyvale California). All those data, read from the screens instruments are directly inserted in the spreadsheet 'Notebook SIPEX-2' in the '4051 Lannuzel' folder. The total length of this core is cut in sections of 7 cm. The second core is dedicated to the POC/PON (Particulate Organic Carbon/ Particulate Organic Nitrogen), DOC (Dissolved Organic Carbon) and nutrients. Six sections of 7 cm are taken from this core. The six sections were chosen so that two top, two intermediate and two basal sections. Two cores are taken for the trace metal analysis. Those cores are directly triple bagged in plastic bags (the inner one is milli-Q washed) and frozen at -20 degrees C until analysis at the laboratory. Brine samples were collected by drainage from "sack holes". Brines and under ice seawater (~1 m deep) were collected in 1 l Nalgene LDPE bottles using an insulated peristaltic pump and acid cleaned C-flex tubing (Cole Palmer). All samples were then transported to the ship as quickly as possible to prevent further freezing. Those samples are used to analyse unfiltered and filtered metals, Chla, POC/PON, nutrients and DOC. Filtration for filtered metals is done on board using a peristaltic pump and a 0.2 micron cartridge filter. All the unfiltered and filtered metals collected are acidified (2 ppt HCl seastar) and stored at room temperature until analysis at the laboratory. Nutrients, DOC and filters for POC/PON are frozen at -20 degrees C until analysis. Chla filtrations and analysis are done on board. Auxiliary cores/brines/underlying seawater were also collected for Caitlin Gionfriddo (caitlingio@gmail.com, Uni. Melbourne) for total mercury (Hg) and methyl-Hg. Also included in this dataset are typed field notes.

Oceanographic processes in the subantarctic region contribute crucially to the physical and biogeochemical aspects of the global climate system. To explore and quantify these contributions, the Antarctic Cooperative Research Centre (CRC) organised the SAZ Project, a multidisciplinary, multiship investigation carried out south of Australia in the austral summer of 1997-1998. Taken from the abstracts of the referenced papers: The development of a semi-automated batch HG-AFS method for the shipboard determination of As(III), As(V),MMA and DMA is described. Procedures in the analytical sequence including addition of NaBH4 to samples, cooling and heating the U-trap used for pre-concentration and separation of the arsines, and logging the AFS output are automated. Overall control of the automated tasks into a logical analytical sequence is achieved using a commercially available data acquisition and control package, workbenchmac(TM). Further modifications required for the method to be adapted to shipboard use, including the use of a hydrogen generator, are also detailed. This method shows a number of advantages over a previously reported manual HG-AFS method including, shorter sample throughput time, increased precision and most significantly, ease of use under shipboard conditions. The semi-automated method was operated on the RSV Aurora Australis during a Southern Ocean voyage in March 1998. Arsenic measurements from a surface transect between 42 and 55 degrees S along 141 degrees 30 minutes E, are presented. Application of the method to more routine laboratory use is also discussed. Distribution of the arsenic species total inorganic arsenic [As(V+III)], arsenite [As(III)], monomethyl arsenic(MMA), andd dimethyl arsenic (DMA) was studied in the Subantarctic Zone (SAZ) of the Southern Ocean, south of Australia, during the austral autumn (March 1998). As (V) was the dominant arsenic species in both vertical profiles and surface waters along the meridional transect 42-55 degrees S, 141 degrees 30' E. It was also the only species observed at depths greater than 600 m. Concentrations of the reduced arsenic species (As(III), MMA, and DMA) were low in these waters compared with other oceanic sites with similar concentrations of chlorophyl a. As(III) concentrations could not be reliably quantified at any sites (less than 0.04 nM). The greatest conversion of As(V) to "biological" species was found at the surface in the Subtropical Convergence Zone(2.5%) and decreased heading southward to 1% in the Polar Front (PF). While the decline in methyl arsenic concentrations was broadly associated with water temperature and measures of biological production, slightly different trends were found in the SAZ and PF. North of the Subantarctic Front (SAF), methyl arsenic concentrations were well correlated with water temperature, while south of the front, no such relation existed. In addition, the ratio DMA/MMA increased south of the SAF, associated with a change in the microalgal community composition. Low water temperature, phosphate replete conditions, and low biological productivity in the Southern Ocean all contribute to the concentrations of biologically produced arsenic species in this region being among the lowest reported for oceanic waters.

Water samples for dissolved trace metal measurements were collected from the surface (15m) down to the 1000m using an autonomous intelligent rosette system (General Ocanics, USA) specially adapted for trace metal work and deployed on a Dyneema rope. The rosette was equipped with 12x10-L Niskin-1010X bottles specially modified for trace metal water sampling. This system has been successfully deployed on the RSV Aurora Australis during voyages au0703 and au0806. Care was taken to avoid any contamination from the ship and the operating personnel. Water samplers were processed aboard under an ISO class 5 trace-metal-clean laminar flow bench in to a trace-metal-clean laboratory container on the ship's trawl deck. All transfer tubes, filtering devices and sample containers were rinsed liberally with sample before final collection. Samples were then drawn through C-Flex tubing (Cole Parmer) and filtered in-line through 0.2 micron pore-size acid-washed capsules (Pall Supor membrane, Acropak 200). Regular sampling depths were as follows: 1000m, 750m, 500m, 300m, 200m, 150m, 125m, 100m, 75m, 50m, 30m, 15m. Samples were analysed within a minute of filtration. Iron(II) was detected with the luminol method combining the experimental set-up of Hansard et al. (2009) with the chemistry as described by Croot and Laan (2002). Samples were not acidified prior to analysis and were pumped directly into the flow cell without an injection valve. Care was taken to maintain a stable light field during measurements as the luminol reagent was found to be extremely sensitive to changes in light intensity. Photons from the reaction of luminol with iron(II) were counted with a Hamamatsu photomultiplier tube housed in a light-tight box. The signal was recorded using FloZ software (GlobalFIA) and the data for each run is stored in a separate file. There is a folder for each profile that contains all the files (automatically generated by the software), which are numbered. The file numbers (e.g. sample1, sample2,...) correspond to the runs as noted in the lab book (see scans). P.L. Croot, P. Laan (2002). Analytica Chimica Acta 466: 261-273. S.P. Hansard et al. (2009). Deep-Sea Res. I 56: 1117-1129.

Sampling was conducted according to GEOTRACES protocols. Samples for trace element analyses, including dissolved iron (dFe), were filtered through acid-cleaned 0.2 um cartridge filters (Pall Acropak) under constant airflow from several ISO class 5 HEPA units. All plastic ware was acid-cleaned prior to use, following GEOTRACES protocols. Samples were collected into low-density polyethylene (LDPE) bottles, acidified immediately to pH 1.7 with Seastar Baseline hydrochloric acid (HCl), double-bagged and stored at room temperature until analysis on shore. Samples for dFe analysis were pre-concentrated offline (factor 40) on a SeaFAST S2 pico (ESI, Elemental Scientific, USA) flow injection system with a Nobias Chelate-PA1 column. Samples were eluted from the column in 10% distilled nitric acid (HNO3), with calibration based on the method of standard additions in seawater (made using multi-element standards in a 10% HNO3 matrix, rather than an HCl matrix). Pre-concentrated samples were analysed using Sector Field Inductively Coupled Plasma Mass Spectrometry (SF-ICP-MS, Thermo Fisher Scientific, Inc.). Data were blank-corrected by subtracting an average acidified milli-Q blank that was treated similarly to the samples. The dFe detection limit for a given analysis run on the SeaFAST/SF-ICP-MS was calculated as 3 x standard deviation of the milli-Q blank on that run. Detection limits ranged from 0.016 to 0.067 nmol kg-1, with a median of 0.026 nmol kg-1 (n=12). GEOTRACES reference materials were analyzed along with samples and results were in good agreement with consensus values: SAFe D1 was measured at 0.69 +/- 0.05 nmol kg-1 (n=7; consensus value = 0.67 +/- 0.04 nmol kg-1) and GD was measured at 1.02 +/- 0.01 nmol kg-1 (n=6; consensus value = 1.00 +/- 0.1 nmol kg-1). Comments regarding the data spreadsheet: NaN = no sample dFe QC flags: 1 = high confidence in data quality 2 = detection limit 3 = low confidence in data quality detection limits: dFe data that were below the daily detection limit were replaced with the respective detection limit. They are flagged with the number 2 in the dFe QC flag column.

Fast repetition rate fluorometer (FRRF) study of sea ice algae in low iron conditions. Algae were grown in an ice tank and the measurements were made at the end with a Chelsea Insrtuments FRRF. Materials and Methods (see the download document for original formatting and formulas) 1. Ice tank incubation The polar pennate diatom Fragilariopsis cylindrus, isolated from Antarctic pack ice in 2015 (Davis station, East Antarctica) was incubated in a purpose designed ice tank (Island Research, Tasmania). The ice tank, which was contructed of titanium to minimise dissolved Fe, was placed into a freezer (–20 degrees C), and the ice thickness and temperature gradient controlled by interaction between a basal heater and the adjustable ambient freezer temperature (see Kennedy et al., 2012). This enabled an ice thickness of approximately 5.5 cm to be maintained during the experiment. The diatom F. cylindrus was incubated in Aquil media (Price et al. 1989) buffered with ethylenediaminetetraacetic acid (EDTA) at 150 micro mol photons m−2 s−1 (PAR), a salinity of 35, and a Fe concentration of 400 nM, where the concentration of total inorganic forms of Fe (Fe') was 1.54 nM, this being continuously supplied to the medium and the exact value calculated using the software Visual MINTEQ, ver. 3.1 (https://vminteq.lwr.kth.se). Before a freezing cycle started, the seawater temperature was maintained at 2.5 degrees C, and a sample was obtained to assess the original physiological state of the algae (Day−5, hereafter). After obtaining the sample, the seawater temperature was set to −1.8 degrees C to initiate ice formation in the ice tank. Once ice had formed at Day−2, the under-ice seawater was partially replaced with ultrapure water to reduce the salinity down to 35, because the salinity had increased (to approximately 38) as a result of brine rejection from the ice. After a 2-day acclimation to the new salinity, ice samples were obtained every 5 days for 20 days (i.e., Days 0, 5, 10, 15, and 20). To minimize the heterogeneity among ice cores, ice samples were randomly collected from the tank chamber with a trace metal-free hand drill (2 cm in diameter) from randomly annotated grids on the ice surface, following normal random sampling numbers generated by the software R (https://www.r-project.org/). To assess the effects of melting and high light exposure, the ice was melted at 2.5 degrees C for 2 days. After the ice had completely melted, the seawater was exposed to a high light level, which was adjusted to represent the likely summer light intensity at the surface in ice-edge regions (800 micro mol photons m−2 s−1; MODIS Aqua), Seawater samples were obtained both after the melting and light exposure events (Melt and Light, respectively, hereafter). 2. Fast repetition rate (FRR) fluorometry To monitor the photophysiology of F. cylindrus during the freezing and melting processes, variable chlorophyll a fluorescence (ChlF) measurements were conducted using a bench-top Fast Repetition Rate fluorometer (FRRf) (FastOcean Act2Run Systems, Chelsea Technologies) with Act2Run software (Chelsea Technologies). Ice samples were directly thawed at 2 degrees C in the dark for 30 min, and the slushily melted ice samples were placed in a quartz tube and their flouresence (ChlF) was measured. A single turnover protocol was applied for the ChlF measurements; 100 flashlets with 1 micro second duration at a wavelength 450 nm and 2 micro second intervals for excitation of reaction centres of photosystem II (PSII, hereafter), and 20 flahlets with 1 μs duration and 100 micro second intervals for relaxation. Eighteen light steps were applied to generate a rapid light curve (RLC) from 0 to 1800 μmol photons m−2 s−1, taking less than 5 min to complete one RLC. At each light step (~15 s), at least five induction and relaxation curves were averaged to obtain ChlF yields, described in Table, after calibrating the ChlF yields with filtered seawater. According to the models proposed by Kolber et al. (1998), photosynthetic parameters of chlorophyll a (chl a) induction and relaxation curves were calculated based on the ChlF yields as shown in Table. Electron transport rate though the reaction centres of PSII (RCII) (ETRRCII) was calculated as per the equation detailed in the download document.