Sampling and analytical procedures:
Incubation 1 was setup from unfiltered seawater collected aboard the R/V/I/B Nathanial B. Palmer using a SeaBird GEOTRACES style SBE32 rosette system deployed on a conducting Kevlar line with OceanTestEquipment, Inc. X-Niskin samplers modified for trace element sampling. Seawater was collected from three consecutive casts to 25-35 m depth at -62.332N, -64.647E and the seawater from each cast was homogenized in three acid-cleaned and Milli-Q (>18.2 MΩ cm) conditioned 50-L polypropylene carboys, then distributed into acid-cleaned and Milli-Q conditioned 4-L polycarbonate incubation bottles that were rinsed three times with the seawater prior to filling. Once filled, incubation bottles were spiked with their treatments in laminar flow hoods in a shipboard trace metal bubble, sealed with the caps wrapped in parafilm and placed in a 2 ºC temperature-controlled lit incubator van with 24 hour blue light (Hopkinson et al. 2007). Dark treatment bottles were placed in the same van but wrapped in heavy-duty black construction garbage bags.

Incubation 2 was setup from unfiltered seawater collected aboard the R/V/I/B Nathanial B. Palmer using a SeaBird GEOTRACES style SBE32 rosette system deployed on a conducting Kevlar line with OceanTestEquipment, Inc. X-Niskin samplers modified for trace element sampling. Seawater was collected from two consecutive casts to 25-35 m depth at -62.46N, -59.565E and the seawater from each cast was homogenized in three acid-cleaned and Milli-Q (>18.2 MΩ cm) conditioned 50-L polypropylene carboys, then distributed into acid-cleaned and Milli-Q conditioned 4-L polycarbonate incubation bottles that were rinsed three times with the seawater prior to filling. Once filled, incubation bottles were spiked with their treatments in laminar flow hoods in a shipboard trace metal bubble, sealed with the caps wrapped in parafilm and placed in a 2 ºC temperature-controlled lit incubator van with 24 hour blue light (Hopkinson et al. 2007). Dark treatment bottles were placed in the same van but wrapped in heavy-duty black construction garbage bags.

Incubation 3 was setup from a combination of filtered and unfiltered seawater collected aboard the R/V/I/B Nathanial B. Palmer using a SeaBird GEOTRACES style SBE32 rosette system deployed on a conducting Kevlar line with OceanTestEquipment, Inc. X-Niskin samplers modified for trace element sampling. Seawater was collected from one cast to 25-35 m depth at -62.46N, -59.565E, the location of Incubation 2, filtered through a 0.2 um Acropak membrane capsule filter (Pall) and stored in an acid-cleaned and Milli-Q (>18.2 MΩ cm) conditioned 50-L polypropylene carboy. Additional unfiltered seawater was collected from one cast to 25-35 m depth at -62.332N, -64.647E, the location of Incubation 1, and homogenized in two acid-cleaned and Milli-Q (>18.2 MΩ cm) conditioned 50-L polypropylene carboys. The unfiltered water from the Incubation 1 station was distributed into acid-cleaned and Milli-Q conditioned 4-L polycarbonate incubation bottles that were rinsed three times with the seawater prior to filling and used for one treatment of Incubation 3 (Q in the light, V in the dark). The filtered seawater from the Incubation 2 station was mixed in a 50:50 ratio with the unfiltered water from the Incubation 1 station in acid-cleaned and Milli-Q conditioned 4-L polycarbonate incubation bottles that were rinsed three times with the seawater prior to filling and used for the second treatment in Incubation 3 (R in the light, W in the dark). Once filled, incubation bottles were sealed with the caps wrapped in parafilm and placed in a 2 ºC temperature-controlled lit incubator van with 24 hour blue light (Hopkinson et al. 2007). Dark treatment bottles were placed in the same van but wrapped in heavy-duty black construction garbage bags.

For all incubations, samples for macronutrients were filtered through sequential 5 um and 0.4 um acid-cleaned polycarbonate track etched filters (Whatman Nuclepore) on Teflon filtration rigs (Savillex) and the filtrate collected in 50 mL Falcon tubes that had been rinsed with distilled water (DIW), soaked overnight in 10% hydrocholoric acid (HCl, Fisher, Trace Metal Grade), rinsed three times with DIW again, dried and rinsed three times with sample prior to filling. Samples were analyzed shipboard, typically within 24 hours, for nitrate+nitrite, phosphate, silicate, and occasionally nitrite. Until analyzed shipboard, samples were stored sealed at 4 ºC in the dark and, following analyses, samples were frozen at -20 ºC and shipped back to the University of South Florida for laboratory-based analyses of nitrite and ammonium, and in some cases again for nitrate+nitrite, phosphate and silicate.

Analytical methodology was based on established methods (Parsons 1984; Gordon et al. 1993) as described for the Lachat 8500 QuickChem in the Lachat QuickChem methods manuals and for the Technicon AAII in the CARIACO methods manual. All labware were either glass or high density polyethylene and were cleaned by an initial distilled water (DIW) rinse, followed by an overnight 10% hydrochloric acid (Fisher, Trace Metal Grade) soak, then rinsed three times with DIW and three times with solvent, analyte or sample prior to fill. Dedicated glassware was used for reagents and standards to avoid cross contamination issues. All reagents were made in high purity Milli-Q (>18 MΩ cm) water.

An artificial seawater matrix was used as the carrier for the analyses on the Lachat QuickChem 8500 at sea and on the Technicon AAII in the lab. 4L batches of artificial seawater were made by dissolving pre-weighed salts (128.50 g NaCl, 28.50 g MgSO4*7H2O, 0.6714 g NaHCO3) in Milli-Q to match Southern Ocean seawater salinity and adjusted to match Southern Ocean seawater pH with 10% hydrochloric acid.

Five-point standard curves were analyzed in duplicate at the beginning and end of each run with duplicate reagent blanks, and quality control checks every seventh sample. Quality control check consisted of either an international reference sample or a quality control sample. The certified nutrient reference samples used, lots CC and CD, were purchased from Kanso Technos in Osaka, Japan. The quality control sample was made from a 20-L homogenized surface Southern Ocean filtered and autoclaved seawater sample. The midpoint standard from the calibration curve was also analyzed every fourteenth sample to check for drift during the runs.

Detection limits for all five parameters on the two instruments used were determined from three times the standard deviation of replicate artificial seawater blanks (n>6). On the Lachat QuickChem 8500, limits of detection were 0.01 uM for nitrate+nitrite, 0.02 uM for phosphate, 0.03 uM for silicate, 0.05 uM for nitrite and 0.5 uM for ammonium. On the Technicon AAII, limits of detection were 0.06 uM for nitrate+nitrite, 0.02 uM for phosphate, 0.2 uM for silicate, 0.01 uM for nitrite, and 0.05 uM for ammonium.

Sample analyses for macronutrients were performed by William Abbott (USF).

Gordon, L.I., Jennings, J., J.C., Ross, A.A. and Krest, J.M. 1993. A suggested protocol for continuous flow automated analysis of seawater nutrients (phosphate, nitrate, nitrite and silicic acid) in the WOCE Hydrographic Program and the Joint Global Ocean Fluxes Study, Methods Manual WHPO 91-1. WOCE Hydrographic Program Office.

Hopkinson, B. M., B. G. Mitchell, R. A. Reynolds, H. Wang, K. E. Selph, C. I. Measures, C. D. Hewes, O. Holm-Hansen, and K. A. Barbeau. 2007. Iron limitation across chlorophyll gradients in the southern Drake Passage: Phytoplankton responses to iron addition and photosynthetic indicators of iron stress. Limnology and Oceanography 52: 2540-2554.

Parsons, T.R., Maita, Y. and Lalli, C.M. 1984. A manual of chemical and biological methods for seawater analysis. Pergammon Press, Oxford, 173 pp.