Changing biogeochemistry of the Southern Ocean and its ecosystem implications

The datasets published here apply only to unpublished nutrient data from the wintertime trans-Southern Ocean sections WOCE line IO6 (2017) and A12 (2019) and one summertime surface ocean A12 ammonium dataset. Nutrient concentrations are in units micromole per liter. Variables measured from the CTD (pressure, temperature and salinity) for the two wintertime datasets are provided at 1 m resolution.

Winter nutrient sampling was conducted aboard the R/V SA Agulhas II in 2017 (WC-17; 28 June – 13 July 2017) along WOCE line IO6 (Indian sector) and in 2019 (SCALE; 18 July – 12 August 2019) along WOCE line A12 (the GoodHope repeat hydrographic line; Atlantic sector). Seawater was collected at regular depth intervals in 12-L Niskin bottles attached to a CTD rosette. Samples for the analysis of nitrate, nitrite, dissolved silicon and phosphate concentrations were decanted into replicate 50 mL HDPE tubes that were copiously rinsed prior to filling. Duplicate tubes were immediately frozen at -20°C for later measurement of nitrate and dissolved silicon, whilst nitrite and phosphate samples that were to be measured shipboard within a few hours were stored in the fridge. Duplicate samples of unfiltered seawater (~40 mL) were also collected at each depth between the surface and 500 m for the analysis of ammonium concentrations in 50 mL HDPE Nalgene bottles that had been stored (“aged”) with orthophthaldialdehyde working reagent (OPA-WR) prior to sample collection. The OPA-WR was decanted just prior to sample collection and bottles were rinsed three times with sample seawater prior to filling.

Phosphate and nitrite concentrations were analysed manually according to the methods described by Grasshoff et al. (1983), with absorbance measured using a Thermo Scientific Genesis 30 Visible spectrophotometer. Aliquots of a certified reference material (CRM; JAMSTEC) were analysed with each sample run to ensure data quality. Nitrate+nitrite and dissolved silicon were measured in the Marine Biogeochemistry Lab at the University of Cape Town (MBL-UCT) using a Lachat Quick-Chem flow injection autoanalyser (Wolters, 2002;Egan, 2008). Standards of varying concentration were run after every ten samples to monitor instrument performance and allow for correction of any drift, and a CRM was measured at the beginning and end of each run to ensure measurement accuracy. The precision of the nitrate+nitrite, dissolved silicon, phosphate, and nitrite measurements was ± 0.4 µmol L-1, ± 0.2 µmol L-1, ± 0.06 µmol L-1, and ± 0.05 µmol L-1, respectively, and the detection limit was 0.1 µmol L-1, 0.2 µmol L-1, 0.05 µmol L-1, and 0.05 µmol L-1, respectively.

Ammonium concentrations were measured shipboard via the fluorometric method of Holmes et al. (1999) using a UV module in a Turner Designs Trilogy Fluorometer 7500-000. Standards were made daily using Type-1 ultrapure Milli-Q water, and samples and standards were measured in duplicate. Precision was ± 0.01 µmol L-1 and the detection limit was <0.02 µmol L-1. The matrix effect resulting from the calibration of seawater samples to Milli-Q standards was calculated according to the standard addition method (Saxberg and Kowalski, 1979). All samples were corrected for the matrix effect (Holmes et al., 1999), which was always <10% and typically <5%. No ammonium concentration data are available for the summer in the WOCE database. However, we collected triplicate samples of unfiltered seawater (~40 mL) from the underway system (~7 m intake depth) of the R/V SA Agulhas II every ~0.25 degrees of latitude during the 2018/2019 resupply voyage along WOCE line A12 between Cape Town and Antarctica (SANAE 58; 7 – 17 December 2018). These samples were measured shipboard as described above. Although not collected at the same time as the other summertime nutrient data from WOCE line A12, the SANAE 58 ammonium concentrations provide an indication of typical summertime conditions. 

N* is a quasi-conservative tracer used to track the changes in dissolved inorganic nitrogen (DIN) relative to phosphate, thereby providing information as to whether fixed nitrogen is being added to or lost from an ecosystem relative to phosphorus (Gruber and Sarmiento, 1997). N* was originally defined as N* (µmol L-1) = [NO3-] – 16 x [PO43-]. Here, we further derive N*DIN = [NO3-+NO2-+NH4+] – 16 x [PO43-] where DIN is the sum of nitrate, nitrite and ammonium, and 16 is the average stoichiometric N:P ratio observed during the autotrophic production and heterotrophic remineralisation of organic matter (Redfield et al., 1963;Anderson and Sarmiento, 1994). The tracer Si* was initially developed to track SAMW from its formation region into the lower latitude ocean (Sarmiento et al., 2004). It is also an indicator of the nutrient status of diatoms. Si* leverages the general observation that under favourable conditions, diatoms consume dissolved silicon and nitrate in a ratio of ~1:1 (Hutchins and Bruland, 1998;Takeda, 1998;Ragueneau et al., 2000), but that under conditions of limitation (e.g. of low iron), the ratio of dissolved silicon-to-nitrate uptake changes (e.g. Franck et al., 2000;Brzezinski et al., 2003). Si* was computed from measurements of nitrate and dissolved silicon concentrations following Sarmiento et al. (2004) as Si* (µmol L-1) = [Si(OH)4] – [NO3-].