Data for Boyd et al. (2023) 'Controls on polar Southern Ocean deep chlorophyll maxima: viewpoints from multiple observational platforms'

Brief description

This record presents data used in the paper 'Controls on polar Southern Ocean deep chlorophyll maxima: viewpoints from multiple observational platforms,' Philip W Boyd 𝘦𝘵. 𝘢𝘭., submitted to Global Biogeochemical Cycles, November 2023.

All methods for the following datasets are detailed and cross-referenced in the paper.

Data were collected from a range of methods, including:

• vertical profiles (from 1 m resolved profiling using sensors on a CTD rosette: temperature, salinity, chlorophyll fluorescence, transmissivity - all calibrated)

• vertical profiles (from discrete samples collected from CTD rosette or trace metal clean rosette, for nutrients, chlorophyll, POC, dissolved and particulate iron, active fluorescence, net primary productivity, biological iron uptake)

• tow-body sections (undulating tow body (Triaxus) for temperature, salinity, chlorophyll fluorescence, transmissivity (and the ratio of chlorophyll fluorescence, transmissivity)

• time-series observations from a robotic profiling float (BGC-ARGO) for temperature, salinity, chlorophyll fluorescence, and transmissivity).

Lineage

Maintenance and Update Frequency: notPlanned

Statement: The data presented here were obtained during the 42 day SOLACE (Southern Ocean Large Areal Carbon Export) voyage on the RV Investigator (IN2020_V08) to the Southern Ocean from 6 December 2020 to 15 January 2021, and include several months of observations from a robotic profiling BGC-ARGO float deployed at a polar site (55.48°S 138.5°E) during this voyage. SOLACE occupied three sites, one in the sub-Antarctic (47.1°S, 141.4°E) near the Southern Ocean Time Series (SOTS) station, and two south of the Polar Front (55.8°S 138.5°E (24 to 31 December 2020), 57.8°S 141.5°E (1 to 11 January), S-Figure 1). The results presented here focus only on the two polar sites, both sampled in a quasi-Lagrangian mode following the deployment of a holey-sock drogue at the mid-depth of the seasonal mixed layer. The drogue drift trajectory over 10 days at the 55.8°S site is presented in S-Figure 2. The trajectory was typical of the low advective regimes we sought on SOLACE, and was similar to that observed at the southern site.

At each polar site, vertical oceanographic profiles were obtained using a Sea-Bird SBE911-plus CTD unit (conductivity, temperature, and depth) that was linked to calibrated fluorometer (Chelsea Aqua-Tracker Mk3), oxygen (SBE 43 electrode), photosynthetically active radiation (i.e., PAR, Biospherical Laboratories) and transmissometer (Wetlabs C-Star 700 nm) sensors. Mixed layer depths were computed for each CTD profile using the mean of a density threshold and density gradient algorithm. For the threshold, we followed Boyer Montegut's 𝘦𝘵 𝘢𝘭. (2004) criteria (via Holte and Talley, 2009): a density difference of 0.03 kg m⁻³ referenced to the closest measurement to 10 dbar. For the mixed layer gradient, we followed Dong 𝘦𝘵 𝘢𝘭. (2007) (via Holte and Talley, 2009) where the gradient criterion was 0.0005 kg m⁻³ dbar⁻¹.

For CTD profiles of chlorophyll fluorescence, generally daytime values exhibiting NPQ (non-photochemical quenching) were interpolated between dark (i.e., nighttime or deep) values. For the continuous underway fluorescence measured while in the vicinity of each site, daytime values were interpolated because nighttime measurements occur close in space and time (Thomalla 𝘦𝘵 𝘢𝘭., 2018). However, for the CTD profiles, nighttime profiles were only used if they were obtained within 24 hours and 50 km of the relevant daytime profile. In this scenario, all nearby nighttime profiles were averaged to create a representative mean night profile. Daytime values from above the euphotic depth, and which were lower than the mean nighttime profile, were replaced with those from the mean nighttime profile, after Thomalla 𝘦𝘵 𝘢𝘭. (2018). The euphotic depth was calculated as the 𝘪𝘯 𝘴𝘪𝘵𝘶 depth were PAR is 1% of surface ocean values (Kirk 𝘦𝘵 𝘢𝘭., 1994). If there were no nearby nighttime profiles to interpolate over a given daytime cast, then all fluorescence values above the maximum value within the euphotic zone were assigned equal to that value, following Xing 𝘦𝘵 𝘢𝘭. (2012) and Biermann 𝘦𝘵 𝘢𝘭. (2015). Profiles were subsequently smoothed with a 5 m moving average to remove high frequency variability and the associated risk of over-correction (Xing 𝘦𝘵 𝘢𝘭., 2012).

Discrete chlorophyll samples were used fit to a linear regression against the corresponding in-situ measurements, both underway and on the CTD. These relationships were used to correct between instruments, such that all measurements are corrected toward the shipboard fluorometer. For CTD values, this relationship was computed independently at SOTS and the two combined Southern sites with site-specific correction factors used. For underway values a single correction factor was computed, as all discrete underway sampling was done in transit, between sites.

The CTD and associated instruments were mounted within the frame of a 24 bottle (12 L) rosette sampler. The CTD sensor package was calibrated after Kwong 𝘦𝘵 𝘢𝘭. (2020). Seawater was sampled from the rosette at selected depths for nutrients and rate measurements (including iron uptake, see later). Dissolved macronutrients were analysed following procedures in Rees 𝘦𝘵 𝘢𝘭. (2018). Seawater samples for trace metal and isotope determination were collected using acid-cleaned, Teflon-coated, externally-sprung, 12 L Niskin bottles attached to an autonomous rosette equipped with a Sea-Bird SBE911-plus CTD unit following methods detailed in Ellwood 𝘦𝘵 𝘢𝘭. (2020a, 2020b). Particulate trace metal samples were collected 𝘪𝘯 𝘴𝘪𝘵𝘶 onto acid-leached 0.2-µm PVDF (142 mm diameter) filters (Sterlitech) using six large-volume dual-head pumps (McLane Research Laboratories), deployed at various water depths (Ellwood 𝘦𝘵 𝘢𝘭., 2020a, b). Elemental analysis for dissolved and particulate trace metals followed procedures in Ellwood 𝘦𝘵 𝘢𝘭. (2020b). Discrete particulate organic carbon (POC) samples were analysed following Trull 𝘦𝘵 𝘢𝘭. (2018). The POC data were used to calibrate the CTD transmissometer (S-Figure 3). 𝘐𝘯 𝘴𝘪𝘵𝘶 values from the instrument were averaged across all depths within 5 m (above or below) of the Niskin bottles from which the discrete POC samples were obtained. Beam attenuation, c (m⁻¹) was computed from the transmissometer as

c = -(1/.25) * ln(trans/100)

where trans is the beam transmittance (%). The beam attenuation from particulates (cp) is estimated by subtracting out the attenuation due to the intrinsic properties of seawater (csw),

cp = c - csw

where csw is set using the minimum value measured by the sensor at depth, assuming particle-free seawater. An estimate of the proportion of detrital (i.e., non-phytoplankton) POC was obtained by assuming a carbon to chlorophyll ratio of 30 (g:g) in living phytoplankton (Strutton 𝘦𝘵 𝘢𝘭., 2023), such that

Detrital fraction = (POC - 30*chlorophyll).

Water from the trace metal rosette was also obtained for biological and biogeochemical metrics. Extracted chlorophyll (chl), photochemical efficiency (Fᵥ/Fₘ), and the functional absorption cross section (σPSII; nm² reaction centre (RC)⁻¹) of photosystem II (PSII) were measured after Boyd 𝘦𝘵 𝘢𝘭. (2022). Biogenic silica (BSi) was determined by measuring silicic acid spectrophotometrically after converting BSi to silicic acid through leaching with 0.1 M sodium hydroxide at 85°C for 2.25 h (Paasche 1973). Samples from multiple depths across the seasonal mixed layer and DCM/DBM were incubated in shipboard temperature-controlled (± 0.5 °C) seawater incubators, with light depths mimicked using a range of neutral density screening. Daily rates of net primary production (NPP) and iron uptake were measured for 0.2-2 μm, 2-20 μm, > 20 µm size fractions and the community (> 0.2 μm) following procedures in Boyd 𝘦𝘵 𝘢𝘭. (2022). NPP was calculated from non-titanium-washed filters, as titanium decreased carbon (C) uptake rates by ~15%. However, iron (Fe) and Fe:C uptake rates were calculated using titanium-washed Fe and C samples, and so are intracellular. Six light depths were chosen to provide coverage across the mixed layer and within the underlying DCM. The 1% 𝘐₀

Notes

Credit
This research was supported by a grant of sea time on RV Investigator from the CSIRO Marine National Facility (https://ror.org/01mae9353