Auxiliary Material for Paper [2005GB002505]
Dissolved Iron in the Tropical and Subtropical Atlantic Ocean
B.A. Bergquist(1,2), E.A. Boyle(1)
1) Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
2) Department of Geological Sciences, University of Michigan, 2534 C.C. Little Building, 1100 N. University Ave, Ann Arbor, MI 48109, USA
Global Biogeochem. Cycles, accepted

Expanded Sampling and Analytical Methods:
Sampling Methods
       The trace metal clean seawater samples collected in this study were collected using a variety of methods.  Many of the samples collected on the cruise were taken with the Moored In situ Trace Element Serial Sampler (MITESS) water sampler or with a single MITESS “ATE” (Automated Trace Element) module [Bell et al., 2002].  Each MITESS module opens and closes an acid-cleaned 500 ml polyethylene bottle while underwater in order to minimize chances for contamination.  Details of the different type of sampling schemes can be found in Bergquist [2004] and Boyle et al. [2005].
       For profile work, it is especially challenging to collect trace metal clean samples in the upper 30 m of the water column while maintaining good depth control.  Near surface techniques such as the “towed fish” device [Vink et al., 2000] or “pole” sampling only allow samples to be collected in the upper few meters of the water column, and trace metal clean Go-Flow collectors and MITESS might be suspect in shallow depths where ship contamination might occur.  Therefore, detailed shallow water profiles for Fe are rare [Bruland et al., 1994].  In this study, high density profile samples in the upper 200 m were collected on the July 2001 cruise using multiple “ATE/Vane” samplers and the same principle as the “towed fish” and “pole” sampling methods.  Five “ATE/Vane” devices were attached to a hydrowire running off a winch/crane that extended 3-4 meters off the side of the ship.  To reconstruct depths, the spacing between each “ATE/Vane” was measured, a depth recorder attached to the deepest sampler, and the wire angle measured for each deployment.  All samplers were lowered into the water, and the ship was moved forward at 1-2 knots.  Thus the samplers were being towed along the side of the ship and the “weather vanes” pointed the samplers upstream of the hydrowire.  The forward movement of the ship insured that the samplers collected water moving parallel to the ship’s path and not water that had contacted the ship.  After a minimum of 20 minutes of rinsing in seawater, the samplers opened, flushed for 10 minutes, and then closed.
       After sample collection, sealed sample bottles were taken into a class 100 clean air flow environment for filtration within 12-24 hours of collection in order to avoid Fe loss to bottle walls.  Splits of each sample were vacuum filtered through acid cleaned 0.4 micrometer Nuclepore® filters.  Prior to each filtration, acid cleaned filters and the filter rig were thoroughly rinsed with dilute trace metal clean HCl and then several aliquots of seawater.  The acid cleaned collection bottles were also rinsed several times with filtered seawater prior to the final sample collection.  Two to three separate aliquots of the filtrates were collected and the sequence noted on bottles.  The last aliquot is the sample usually measured for Fe concentration because it is considered to be least likely to have been contaminated during filtration because it was collected after the most flushing.  Random bottle contamination happened infrequently (less than 10%) and high values were checked against measurements of earlier filtrates.  Filtrates were acidified at sea in a class 100 clean environment to pH 2 by addition of triply distilled Vycor 6 N HCl in a ratio of 1 ml acid to 500 ml of seawater. 
       
Fe and Mn Measurement
       Iron, Mn, and Cr concentrations on filtrates were measured simultaneously by a modified version of the method by Wu and Boyle [1998], which utilizes isotope dilution followed by Mg(OH)2 co-precipitation and measurement by ICPMS [Bergquist, 2004; Boyle et al., 2005].  This paper focuses on the Fe data with limited Mn data (data available on-line).  Details of the method and the Mn and Cr data can be found in Bergquist [2004] and will be published in the future.  Briefly, the main differences of the new method used in this study are the use of a Fe-54 isotope spike and a GV Instruments (formerly Micromass) IsoProbe MC-ICPMS.  The IsoProbe incorporates a hexapole collision cell prior to the magnet that eliminates Ar(40)O(16)+ and Ar(40)N(16)+ interferences on masses 56 and 54, which allows samples to be measured in low mass resolution.  The multi-collection feature permits simultaneous collection of masses 52 (monitor Cr and correct for Cr interference on 54), 54, 55 (Mn), 56, and 57.  The largest interference correction for Fe is CaO+ on mass 56.  The CaO+ interference is monitored by measuring CaOH+ on mass 57, measuring the CaO/CaOH ratio on a trace metal clean Ca solution throughout the run, and correcting mass 56 for the CaO+ interference.  Mn and Cr concentrations are calculated by measuring a recovery efficiency (from spiked samples) compared to the Fe-54 spike and by measuring the relative ionization efficiency of Mn, Cr, and Fe in the plasma.  The recovery efficiency for Mn and Cr is calculated by measuring several samples with standard addition spikes throughout a run.  The recovery efficiency for Mn and Cr depends on the time delay between precipitation and centrifugation.  Therefore, each step in the precipitation procedure was timed and kept as constant as possible.  Typical recovery efficiencies for Mn and Cr are 50 +/- 7% and 60 +/- 8% (1 sigma standard deviation (SD)) respectively.  
       Replicate analysis of samples yield precisions of better than +/- 0.05 nmol/kg for Fe and +/- 0.15 nmol/kg for Mn.  Error bars reported in this study represent the 1 SD of replicate analysis of samples.  Procedural blanks for Fe ranged from 0.08 to 0.17 nmol/kg from run to run with typical precisions of +/- 0.03 nmol/kg (1 SD) for individual runs.  For Mn, procedural blanks ranged from 0.4 to 1.0 nmol/kg with typical precisions of +/- 0.1 nmol/kg (1 SD) for individual runs.  
       There are two components to the blank in the method described above: (1) a reagent blank and (2) an instrument blank.  Reagent blanks are assessed by processing a 50 microliter aliquot of a low-Fe seawater sample though the same procedure as the samples.  The 0.3 M HNO3 blank can be measured directly, and the blank associated with the NH4OH is negligible (doubling or tripling of NH4OH does not change blank).  Most of the Fe, Mn, and Cr blanks in this method are due to instrument blank (blanks released from the hardware of the instrument), and not due to reagents.  Because the samples and procedural blanks have different matrices, the reagent blank does not correctly characterize the instrument blank.  Therefore, multiple consistency samples are analyzed at the beginning, middle, and end of each analytical session.  Consistency samples are large-volume, in-house seawater samples that are repeatedly measured.  In order to ensure that these samples do not get contaminated over time, several (usually 3) are measured in each run (random contamination would hopefully not contaminate multiple consistency samples by the same amount of Fe and long term trends in the raw concentration measurements of these samples were not observed).  The concentrations of these samples were defined by comparing concentration measurements by other techniques and/or by other laboratories.  The reagent blank is then slightly adjusted (less than 0.15 nmol/kg) in order to bring the consistency samples into agreement with their defined concentration (after blank subtraction).  The adjusted reagent blank is considered the full procedural blank of the method (including both the reagent and instrument blank) and is subtracted from the raw concentration data to give blank-corrected concentrations. 
       Based on the reagent blank alone, samples can only be compared with confidence using their individual sample replication within a given analytical session (e.g., less than +/- 0.05 nmol/kg (1 SD) for Fe).  However, in order to compare samples from different analytical sessions (e.g., to better than 0.15 nmol/kg for Fe), the changes in the instrument blank must also be included.  Offsets (up to 0.15 nmol/kg for Fe) from analytical session to session due to changes in the instrument blank are corrected by using the consistency samples, which better mimic the matrix of an actual sample.  Therefore, we feel differences between samples in our extended data set are comparable to within the analytical session sample replication (+/- 0.05 nmol/kg (1 SD) for Fe).  Comparisons of our data to other published data sets within approximately 0.15 nmol/kg for Fe should be made with caution, as no inter-lab consistency sample was available at the time these measurements were made.  However, agreement between deep-water concentrations in DFe in the North Pacific measured in our lab [Boyle et al., 2005] and concentrations observed at a nearby station by Bruland et al. [1994] suggest that the concentrations of our consistency samples are reasonably well-defined.  A similar approach was used for the Mn and Cr data and the consistency samples were reproduced to within +/- 15%. 
       Random contamination during analysis remains a problem for Fe (approximately 10%), therefore samples were always analyzed in triplicate.  When at least two replicates agreed within expected reproducibility, the average of two or three replicates was taken as the sample concentration.  If no replicates agree or two replicates were high and one low, the sample was re-analyzed.  If all the sample replicates seem high based on “oceanographic consistency”, then a second filtrate was measured.  If the new result was lower and “consistent”, then the contaminated bottle data was discarded.  If both replicate filtrates are high or no replicate exists, then the sample data is flagged in the attached data tables with a question mark.  Data is omitted from figures and the discussion where a clear judgment could be made that the sample was contaminated.