Methodology:

Water samples were taken from 10 L Niskin bottles mounted on the CTD rosette during cruise JC211. Only selected water samples were analysed for the variables in this dataset; NaN denotes samples where particular analyses were not performed.

Nutrient samples were taken over the upper 500 m of each CTD cast at intervals that were larger in the deeper parts of each profile and smaller in surface waters, with 5 ± 1 depths sampled per profile. Samples were taken in acid cleaned 60 ml HDPE bottles and frozen at -20°C for subsequent analysis on shore at Plymouth Marine Laboratory. Samples were thawed using International GO-SHIP protocols (Becker et al., 2020) to allow redissolution of any secondary silicate precipitates that formed during freezing. Samples were analysed using a SEAL analytical AAIII segmented flow colorimetric autoanalyser for nitrate+nitrite, nitrite, phosphate, silicic acid and ammonium, following Woodward and Rees (2001). Sample handling and analytical protocols followed the International GO-SHIP protocols (Becker et al., 2020) as closely as possible.

Water samples for particulate organic carbon (POC) and particulate nitrogen (PN) analysis were taken from the CTD at two depths (200 m and 400 m). 250 ml water was collected directly into a HDPE bottle and immediately frozen at -20 °C. The water samples were filtered onto pre-combusted (450 °C, 16 h) 25 mm glass fibre filters (GF/F; nominal pore size 0.7 µm) and rinsed with water purified using a Millipore Milli-Q lab water system. Samples were air-dried and fumed for 24 h with 37 % HCl in a desiccator before finally being oven-dried at 50 °C for 24 h. Filters and filter blanks were placed in sterile tin capsules, and POC and PN were measured on a CE440 Elemental Analyser (Exeter Analytical Limited, Coventry, UK), calibrated using an acetanilide calibration standard with a known % C and % N of 71.09 % and 10.36 % respectively. Standards were interspersed regularly between samples to measure and correct for drift.

Water samples for particulate silica (SiO2) analysis were taken from the CTD at three depths (surface at either 5 m or 10 m, the chlorophyll a maximum which varied between 10 m and 50 m, and 100 m). Bottle material was filtered onto 25 mm, 0.4 µm polycarbonate filters and rinsed with Milli-Q water before drying at 50 °C for 24 h. Material on the filters was solubilised via an alkaline extraction method (Hatton et al., 2019) carried out at the Bristol Isotope Group laboratory. Sample material was digested in Teflon tubes with 0.2M NaOH at 100 °C for 40 min. This was followed by neutralisation with 6M HCl. Particulate silica concentrations were measured colorimetrically by spectrophotometry (heteropoly blue method; Strickland and Parsons, 1972) using a Hach DR3900 spectrophotometer set at a wavelength of 815 nm calibrated against a NIST silica standard.

Water samples for phytoplankton analysis were taken from the CTD at four depths corresponding to near surface (5 or 10 m), the chlorophyll a maximum (25 m for all stations used in this analysis other than station H2, which was 10 m), 100 m and 200 m. 200 ml water was collected directly into an amber glass bottle at each depth and was fixed immediately with 4 ml Lugol's iodine solution. Bottles were gently mixed and stored in the dark at 4 °C until later analysis. A sub-sample of phytoplankton samples from the surface (5 m) and chlorophyll a maximum were analysed for phytoplankton and microzooplankton (heterotrophic organisms 20-200 µm in size). Cell identification and enumeration were carried out using standard Utermöhl methodology (Karlson et al., 2010). Subsamples of 50 ml were settled for 24 hours and examined under an inverted light microscope, with taxa identified to species or genera where possible, or by group or size otherwise. Where chains were encounte...(4)

Data collection:

10 L OceanTest Equipment model 110B Niskin bottles mounted on the CTD frame were used to take discrete water samples for chemical and biological analysis.

Nutrient samples were analysed using a SEAL analytical AAIII segmented flow colorimetric autoanalyser for nitrate+nitrite, nitrite, phosphate, silicic acid and ammonium, following Woodward and Rees (2001).

Particulate organic carbon (POC) and particulate nitrogen were measured on a CE440 Elemental Analyser (Exeter Analytical Limited, Coventry, UK).

Particulate silica concentrations were measured colorimetrically by spectrophotometry (heteropoly blue method; Strickland and Parsons, 1972) using a Hach DR3900 spectrophotometer set at a wavelength of 815 nm calibrated against a NIST silica standard.

Phytoplankton cell identification and enumeration were carried out using standard Utermohl methodology (Karlson et al., 2010).

A Sea-Bird Scientific SBE 911plus Conductivity-Temperature-Depth (CTD) system was used to obtain profiles of temperature, salinity, dissolved oxygen (SBE 43 dissolved oxygen sensor), fluorescence (Chelsea Technologies AquaTracka Mk III) and photosynthetically active radiation (PAR; Biospherical QCP-2350-HP). A Sea-Bird Scientific SBE 35 deep ocean standards thermometer was mounted on the CTD frame, taking a measurement each time a bottle was closed.

Discrete water samples for salinity and dissolved oxygen calibration of the CTD data were collected from a subset of Niskin bottles. Salinity samples were analysed on a Guildline Autosal 8400B salinometer against IAPSO standard seawater batch P164.

Dissolved oxygen samples were collected and analysed on board following GO-SHIP procedures (Langdon, 2010), using a Metrohm automatic titration system with an amperometric system to determine the endpoint of the Winkler titration.

Data quality:

Nutrient calibration standards were prepared with low-nutrient seawater and the analyses were quality controlled and checked using Certified Reference Materials (CRMs) for nutrients in seawater from KANSO Ltd. (Japan). Raw data were further corrected to ambient ocean salinity and pH. Samples were analysed in duplicate and standard deviation was generally better than 0.02 µmol L-1 for nitrate+nitrite, 0.01 µmol L-1 for nitrite, 0.02 µmol L-1 for phosphate, 0.06 µmol L-1 for silicic acid and 0.05 µmol L-1 for ammonium.

Analytical precision was better than 1.0 % for particulate organic carbon (POC) and 1.1 % for particulate nitrogen.

Analytical precision for particulate silica was better than 3 %, and full replicates of sample aliquots taken through the dissolution process separately agreed within 8 %. Laboratory blanks were below the detection limit (0.01 mg L-1).

The standard deviation in practical salinity differences between the bottle and sensor values was less than 0.0015.

Dissolved oxygen samples had residual errors, and differences between downcasts and upcasts, generally within 1 µmol kg-1.

Chlorophyll fluorescence data from the CTD had the factory calibration applied, but no calibration against in situ samples.

NaN is used to indicate any bottle samples where a particular parameter was not measured.

The following variables have WOCE data quality flags of 2 (acceptable measurement), 3 (questionable measurement), 4 (bad measurement), 5 (not reported), 9 (not sampled):

- Salinity from sample measurement

- Dissolved oxygen from sample measurement

Manual flagging was not performed for the salinity samples; the flags merely indicate the presence (or absence) or samples. For dissolved oxygen samples, the use of these flags is described in more detail in the cruise report (Abrahamsen, 2021).