Seasonal variations of sinking velocities in Austral diatom blooms: Lessons learned from COMICS

1 Seasonal variations of sinking velocities in Austral diatom blooms: Lessons learned from 1 COMICS 2 M. Villa-Alfageme1, N. Briggs2, E. Ceballos-Romero1, F. de Soto3, C. Manno4, S.L.C. Giering2 3 1 Dept. Física Aplicada II. Universidad de Sevilla. Av. Reina Mercedes 4A. 41012. Sevilla, Spain 4 2 Dept. Ocean Biogeochemistry and Ecosystems, National Oceanography Centre, Southampton, 5 UK 6 3 Dept. de Sistemas Físicos, Químicos y Naturales. Universidad Pablo de Olavide, Carretera de 7 Utrera, km 1, 41013 Sevilla, Spain 8 4 British Antarctica Survey, Cambridge, UK 9 10 ABSTRACT 11 The sinking velocity (SV) of organic particles is a critical driver of carbon transport to the deep 12 sea. Accurate determination of marine particle SV and their influencing factors is therefore a key 13 to better understanding of biological carbon storage in the ocean. We used two different 14 approaches to estimate average SVs of particles during a Southern Ocean spring bloom (North 15 of South Georgia): optical backscatter sensors on gliders (“large”, >50 µm diameter), and 16 radioactive pairs (234Th-238U and 210Po-210Pb). Our results were complemented with time-of flight 17 estimations of bulk SVs from deep sediment traps deployed at 1950 m. 18 Bulk SVs increased consistently with depth from 15 ± 1 m d-1 at 10 m to 50 ± 10 m d-1 at the 19 depth of export (Zp= 95 m) and from 96 ± 35 m d-1 at 150 m to 119 ± m d-1 at 450 m. Only the 20 fastest particles, mainly comprised by faecal pellets (FPs) and diatom aggregates, survived 21 remineralization and dominated carbon fluxes at deep depth. 22 The SV variability at the base of the Euphotic Zone was studied in relation to the stage of the 23 bloom by analysing three different moments of the spring diatom bloom in the region during the 24 years 2012, 2013 and 2017. The export efficiency (ExpEff), defined as the ratio POC flux 25 exported below the Euphotic Zone to the satellite derived surface NPP, was also evaluated. It 26 was found from the temporal series that ExpEff and SV vary throughout the diatom bloom as the 27 community structure progresses. A good correlation between both variables was observed 28 (ExpEff = (0.023 ± 0.006) SV, r = 0.82, p = 0.04). Showing that the variability in how efficiently 29 the carbon flux is exported out of the Euphotic Zone can be explained by the SV at which the 30 particles sink. Further investigations are required to analyse if this is a specific model of the 31 functioning of the BCP during the diatom bloom in North South Georgia or if it can be 32 extrapolated to other scenarios. 33 34 KEY WORDS: particle sinking velocity, 210Po, 234Th, gliders, deep sediment traps, 35 transfer efficiency, export efficiency, carbon attenuation. 36 37 2 1. INTRODUCTION 38 Sinking particles constitute one of the main mechanisms of the Biological Carbon Pump (BCP) 39 (Fowler and Knauer, 1986) and are a major pathway of carbon from the surface into the deep 40 ocean, exporting 5-20 Gt C every year (Henson et al., 2011). Sinking particles are a complex mix 41 of biogeochemical materials, and differ in their size, density, porosity, morphology, and hence 42 sinking velocity (SV). The magnitude of the downward flux by sinking particles depends mainly 43 on the production rate of the particles, the velocity and carbon content of these particles, and the 44 rates at which this material is fragmented or biologically consumed (i.e. the remineralization rate) 45 (De La Rocha and Passow, 2007). Accurate estimation of these rates is hence of great value for 46 advancing understanding of the global ocean carbon sink and its variability. 47 Below the Euphotic Zone, the variation of sinking carbon flux is determined by the balance 48 between particle remineralization rate and SV (Sarmiento and Gruber, 2006). How these two 49 rates vary and interact spatiotemporally is not fully understood (Cram et al., 2018; Henson et al., 50 2012; Le Moigne et al., 2013), partly because both rates are highly variable (Buesseler and Boyd, 51 2009; Cavan et al., 2017; De La Rocha and Passow, 2007; Iversen and Ploug, 2013; Villa52 Alfageme et al., 2016a; Wiedmann et al., 2014), their measurements are intrinsically complex, 53 and few data are available. Moreover, both rates are coupled: SV depend on size, composition 54 and shape, which are typically altered as particles are being remineralized. 55 SV is included in a range of models, including ecological (e.g. Yool et al., 2013), stochastic (e.g. 56 de Soto et al., 2018; Jokulsdottir and Archer, 2016), mechanistic (e.g. Cram et al., 2018; Devries 57 et al., 2014; DeVries and Weber, 2017) and regional Lagrangian models (e.g. Aumont et al., 58 2015). While some mechanistic flux models include different approaches to parameterize particle 59 SVs and remineralisation in order to more accurately describe dynamic biological processes (De 60 Soto et al., 2018; DeVries, 2014; Jokulsdottir and Archer, 2016; Omand et al., 2020), the 61 parameterization of SV in global ocean models usually does not reflect its temporal and global 62 variability (e.g. Aumont et al., 2015; DeVries and Weber, 2017). 63 It is not easy to obtain an accurate determination of in situ SV (e.g. Giering et al., 2020). SV can 64 be measured in the laboratory (e.g. Bach et al., 2012; Laurenceau-Cornec et al., 2020; van der 65 Jagt et al., 2018; Zetsche et al., 2020) and in the field using particle settling velocity traps (IRS66 ST) (Alonso-Gonzalez et al., 2010) or imaging analyses of particles collected on gel sediment 67 traps (McDonnell and Buesseler, 2012). A significant step forward in the evaluation of SV are 68 techniques based on underwater camera imaging (e.g. Giering et al., 2020; Guidi et al., 2008; 69 Iversen et al., 2010; Kiko et al., 2017). These methods have in common that the individual SV 70 of a set of single particles is tracked directly in field or laboratory settings. The bulk SV is then 71 obtained by either extrapolation or estimating particle distributions (Guidi et al., 2008). An 72 alternative is to use methods that estimate in situ bulk SV directly, such as gliders (Briggs et al.) 73 or radioactive pairs (Villa-Alfageme et al., 2014), which track the signal left by the particles as 74 they sink. 75 Here, we synthesise in situ bulk SV measurements during the peak and decline of the 76 phytoplankton bloom north of South Georgia, a naturally iron-fertilized region downstream of 77 3 the island. This work is part of the COMICS (Controls over Ocean Mesopelagic Carbon Storage) 78 project (Sanders et al., 2016). Average particle SV throughout the upper 500 m of the water 79 column were obtained from radioactive tracer pairs 210Po-210Pb and from plume-tracking on 80 glider-derived optical backscatter. In addition, we calculated the bulk SV of a pulse of particles 81 sinking from the base of the Euphotic Zone to a deep sediment trap located at 1950 m depth. We 82 discuss how SV related to export rates and how they vary with community structure, depth, stage 83 of the bloom or other factors. 84 85 2. SAMPLING AND METHODS 86 2.1. Site description 87 The region downstream of South Georgia was sampled in Nov/Dec 2017 on board RRS 88 Discovery (cruise DY086) as part of the UK COMICS programme (Sanders et al., 2016). Station 89 P3 (52º24 S, 40º04 W) (Figure 1), close to British Antarctic Survey’s mooring, was extensively 90 sampled three times in cycles of ~1 week each: P3A (15 November – 22 November, DOY 31991 326), P3B (29 November – 5 December, DOY 333-339) and P3C (9 December – 15 December, 92 DOY 343-349). P3 is located in the Northern South Georgia region, which is characterised by 93 annual large phytoplankton blooms fertilized by iron from the islands nearby (Robinson et al., 94 2016). The iron is introduced to the water column by the Subantarctic Circumpolar Current Front, 95 leading to the increased diatom productivity downstream of South Georgia (Jones et al., 2012; 96 Korb et al., 2012, 2010). In this region, the phytoplankton growth season is long, and blooms 97 usually persist for more than four months (Atkinson et al., 2001). Our cruise followed a rapidly98 changing diatom spring bloom from peak to decline. 99 100 2.2. Radioactive pair disequilibrium (210Po-210Pb and 234Th-238U) 101 2.2.1. Particle organic carbon (POC) export flux from radionuclide pairs 102 The methods to calculate POC downward flux from POC-210Po and POC-234Th are well 103 established (Ceballos-Romero et al., 2016; Le Moigne et al., 2013; Verdeny et al., 2009) and are 104 based on the disequilibrium in the activity concentration of the radioactive daughter (210Po and 105 234Th) and its radioactive parent (210Pb and 238U). A disequilibrium appears when the particle106 reactive radionuclide, either 210Po (T1/2 = 138 days) or 234Th (T1/2 = 24 days), is scavenged from 107 the water column by sinking particles. For 210Po and 234Th downward flux, the radioactive 108 disequilibrium is converted into flux using a one-box model (Savoye et al., 2006). Assuming 109 steady state and negligible advection effects downward flux, P, is obtained following, 110 𝑃 = 𝜆2∑(𝐴2 𝑡𝑜𝑡𝑎𝑙 − 𝐴1 𝑡𝑜𝑡𝑎𝑙) 𝑧=ℎ 𝑧=0 Δ𝑧 (1), 111 where 𝐴1 𝑡𝑜𝑡𝑎𝑙 is the total parent activity concentration for 238U or 210Pb; 𝐴2 𝑡𝑜𝑡𝑎𝑙 is the total 234Th 112 or 210Po activity concentration and λ2 is the decay constant of the daughter. The tracer flux P is 113 then converted into POC downward flux, POCflux, by applying the measured ratio (Q>53) of POC114 to-radionuclide (POC: 210Po or POC: 234Th) in large (>53 m) particles: 115 4 𝑃𝑂𝐶𝑓𝑙𝑢𝑥 = 𝑃 · 𝑄>53 (2). 116 Uncertainties in Q>53 for 210Po and 234Th were obtained directly from error propagation applied 117 to the counting uncertainty (using ±  of number of counts), volume and tracer uncertainties. 118 Both radionuclides present very different half-lives, that have implications specially in the 119 interpretation of results during a non-steady state situation. 120 121 2.2.2. Sinking velocity from radionuclide pairs 122 The methods to calculate SV using the radioactive pair disequilibrium are relatively novel (Villa123 Alfageme et al., 2014), and a full description of the method is presented elsewhere (Villa124 Alfageme et al., 2016a, 2014). In brief, as explained above, the disequilibrium 234Th-238U and 125 210Po-210Pb emerges when 234Th or 210Po is scavenged from the water column by particles sinking 126 at different velocities. This way, 210Po or 234Th flux, P, is generated by the sinking particles and 127 thus average particle SV at specific depths can be diagnosed from, 128 𝑃(𝐵𝑞 𝑚−2𝑠−1)=𝑆𝑉(𝑚 𝑠−1)· 𝐴2 𝑝𝑎𝑟𝑡(𝐵𝑞 𝑚−3) (3), 129 where 𝐴𝑇2 𝑝𝑎𝑟𝑡 is the activity concentration of 234Th or 210Po in the particulate fraction at that depth. 130 We have implemented a method applying an inverse model to Eq (1) to estimate which average 131 bulk SV is responsible for the observed parent-daughter disequilibrium. Eq. (1) is rearranged to 132 include , the fitting parameter in our inverse model, and it is combined with Eq. (3) as, 133 𝐴1 𝑡𝑜𝑡𝑎𝑙(𝑧)· 𝜆2−𝐴2 𝑡𝑜𝑡𝑎𝑙(𝑧)· 𝜆2− 𝜆𝑇ℎ 𝑑(𝛿·𝐴2 𝑡𝑜𝑡𝑎𝑙) 𝑑𝑧 = 0 (4), 134 where  corresponds to, 135 𝛿(𝑧)=𝑆𝑉 ·𝐴2 𝑝𝑎𝑟𝑡 𝜆2 1 𝐴2 𝑡𝑜𝑡𝑎𝑙 (5). 136 SV is obtained from fitting . A confidence interval for each fitting  of the inverse model was 137 estimated from the difference between modelled and measured 210Po concentrations, and 138 subsequent error propagation was used to calculate the uncertainties. 139 140 2.2.3. Sampling and 210Po and 210Pb analysis during COMICS I 141 Total 210Po and 210Pb activities were determined using the coprecipitation with iron (Villa142 Alfageme et al., 2016b). Three vertical profiles (one during each visit: P3A, P3B and P3C) of 5143 L seawater samples were collected from 10 to 1000 m using Niskin bottles attached to a CTD 144 rosette, with higher vertical sampling resolution in the upper 150 m of the water column (Figure 145 S1, Supplementary Material). Samples were immediately acidified and spiked with stable Pb and 146 209Po as chemical yield tracers and Fe3+ for subsequent coprecipitation of Fe(OH)3 by raising the 147 pH using NH4 OH. Iron hydroxide was collected by siphoning the remaining water and stored 148 for processing on land. The coprecipitate was digested with HNO3 and samples were re-dissolved 149 with 1 M HCl. 210Po and 209Po were selectively auto‐deposited onto silver discs during 12 h. The 150 silver discs were counted by alpha-spectrometry using passivated implanted planar silicon (PIPS) 151 5 alpha detectors (Canberra, USA). Solutions were replated. Samples were re-spiked with 209Po 152 and stored for 9–11 months for later determination of 210Pb via 210Po ingrowth. At that time, 153 samples were plated and counted once more by alpha-spectrometry. 210Pb and 210Po activities at 154 sampling time were calculated by applying in‐growth, decay, and recovery corrections (Ceballos155 Romero et al., 2016). 156 Large (>53 µm) and small (<53 µm) particles were collected using stand-alone in situ pumps 157 (SAPS, Challenger Oceanic, UK). SAPS were deployed at 4 depths from 25 m to 450 m, filtering 158 on average 1500 L over 2 hours. Particles retained on the 53-µm pore size nylon mesh screens 159 were rinsed into a bottle using filtered seawater, homogenized and subdivided into six aliquots 160 using a splitter. One aliquot was filtered through QMA filters for analysis of 210Po and 210Pb. Any 161 zooplankton observed on the filters (by naked eye) were removed using forceps. The filters were 162 then spiked with 209Po and stable Pb, digested using a mixture of concentrated HNO3 (3 mL), 163 HCl (1 mL), and HF (0.5 mL), dried and re-dissolved with 1 M HCl. Samples were then analysed 164 using alpha-spectrometry as described for the water samples. One aliquot was filtered onto pre165 combusted and weighted GF/F filters for particulate organic carbon (POC), dried and analysed 166 using by mass evaluation after combustion (Le Moigne et al., 2013). 167 168 2.3. Sinking velocity from optical sensors on gliders during COMICS I 169 Three high-resolution underwater gliders surveyed a triangle with sides of ~ 12 km, centred on 170 52°45 S, 40°26 W, with its southernmost vertex 2 km north of the British Antarctic Survey’s P3 171 mooring, (52º24 S, 40º04 W) as part of the sister project GOCART (Gauging ocean Organic 172 Carbon fluxes using Autonomous Robotic Technologies, Henson et al., this issue). During 2017 173 austral spring, a Seaglider, belonging to CSIR (South Africa), was deployed on the DOY 292 174 and recovered on DOY 365. Two Slocum gliders (G2s type), belonging to NOC (UK), were 175 deployed on DOY 323 and DOY 326 and both recovered on DOY 44 from 2018. All gliders 176 measured temperature, salinity and dissolved oxygen concentration (Henson et al., this issue). 177 They were further fitted with a custom-made Wetlabs Environmental Characterization (ECO) 178 Triplet, measuring backscatter (at 532 and 700 nm) and chlorophyll fluorescence. The gliders 179 were programmed to profile between 0 and 1000 m, with each dive cycle taking ~5 h to complete 180 and covering an average horizontal distance of 4 km. 181 High-resolution vertical profiles of large and small particles were calculated following the 182 method by Briggs et al. (2011). Briefly, a minimum-maximum filter isolates spikes (i.e. ‘large 183 particles’) from a baseline signal (i.e. ‘small particles’). The size of ‘large particles’ was 184 estimated to be >420 µm in diameter (Henson et al., this issue). We then calculated the bulk SV 185 of these large particles using a modification of the plume tracking (Briggs et al., 2020a). Five 186 plumes were identified between DOY 323 and DOY 347 and tracked both visually and using a 187 statistical approach. Two of these plumes also showed a strong signal in the small particles, so 188 the plume tracking was performed for both particle size classes, providing SVs for both large and 189 small particles. Detailed descriptions of the method and POC flux profiles are provided by 190 Henson et al. (this issue) and Giering et al. (this issue). 191 6 192 2.4. Net primary productivity and export efficiency 193 The 8-day satellite-derived net primary production (NPP) data with a spatial resolution of 0.083 194 by 0.083º were obtained from the Oregon State University Ocean Productivity standard products 195 (http://www.science.oregonstate.edu/ocean.productivity/), wherein NPP was estimated by the 196 Vertically Generalized Production Model (VGPM) (Behrenfeld and Falkowski, 1997). Satellite197 derived NPP values agreed with glider-derived primary production estimates for our cruise 198 period (Henson et al., this issue), which was obtained by following the approach of Mignot et al. 199 (2018). 200 Export Efficiency (ExpEff) was defined as the ratio of POC export flux at the base of the export 201 zone (Zp), in our case derived from the radioactive pairs, to NPP. Due to the persistence of the 202 radioactive disequilibrium in the water for more than three weeks for 234Th-238U and more than 203 six weeks for 210Po-210Pb, NPP was calculated in two ways: by integrating the data from (i) the 204 three weeks previous to the sampling, and (ii) if during three weeks previous to the sampling, the 205 blooms, the integration would include only the days of the bloom. 206 The export depth was defined as the base of the productive layer (Zp), which is the layer in which 207 particles can be produced photosynthetically. Hence, Zp denotes the depth that separates euphotic 208 and mesopelagic zone. The estimates of Zp used here were the ones calculated for our cruise 209 (Giering et al., this issue). Briefly, Zp was calculated from the glider time-series (Henson et al., 210 this issue) following the latest recommendations by Buesseler et al. (2020) and using a 211 modification of the metric by Owens et al. (2015). For each glider profile, Zp was defined as the 212 deepest point at which Chl was higher than 10% of the maximum Chl concentration for that 213 profile. The resulting Zps were 92.5 m ± 8.1 m, 91.5 m ± 12.9 m and 88.5 m ± 10.4 m for P3A, 214 P3B and P3C, respectively. As the glider-derived high-resolution POC concentrations and fluxes 215 were binned in 10-m bins, a nominal value of 95 m (i.e. encompassing fluxes between 90 and 216 100 m) was thus used as export depth for all visits. 217 218 2.5. Time-of-flight estimation of bulk SV using a deep sediment trap during COMICS I 219 A sediment trap (McLane Parflux sediment trap, 0.5 m2 surface collecting area; McLane Labs, 220 Falmouth, MA, USA) was deployed on the mooring at P3 as part of the Ecosystems programme 221 and the Scotia Sea Open Ocean Laboratories (SCOOBIES) sustained observation programme at 222 the British Antarctic Survey. The trap contained a carousel of 21 receiving cups and was fitted 223 with a plastic baffle mounted on the opening to prevent the entrance of large organisms. 224 During COMICS I, the sediment trap was deployed at 1950 m and the carousel programmed to 225 rotate every 5 days over a 105-day period between mid-October 2017 and late January 2018, 226 covering the spring bloom period during our cruise. A complete description and discussion of 227 the mooring and observed POC flux at 1950 m is provided by Manno et al. (this issue). 228 7 Using a ‘time-of-flight approach’ (Armstrong et al., 2009; Mas et al., 2020), we estimated the 229 velocity at which temporal patterns in flux propagated between the Zp and the deep sediment trap 230 located at 1950 m depth. The range of SV of the particles that produced the peak of export can 231 be obtained from the time elapsed between the maximum in POC export flux and the moment 232 this peak is detected by the deep sediment trap. As detailed in section 3.1, in 2017 a peak in the 233 carbon was detected by the gliders at the Zp (DOY 319) and an intense peak in carbon export was 234 collected in the deep sediment trap at 1950 m 17 days later (DOY 336). To estimate the bulk SV 235 of these sinking particles, we used the distance between POC flux export peak (DOY 319 at 95 236 m) and deep-sediment trap fluxes (DOY 336 at 1950 m) and the time elapsed between them. 237 Thus, a time-of-flight approach was followed, using 17 days as elapsed time and the 1855 m 238 distance between both detections. 239 240 2.6. NPP, deep sediment traps and sinking velocity during 2012 and 2013 241 To obtain additional information of the variables analysed for P3 during COMICS I spring 242 bloom, NPP, ExpEff and particle average SV were also evaluated for the years 2012 and 2013 243 using 234Th-238U and 210Po-210Pb disequilibrium values from previous cruises. In line with our 244 cruise, the Zp was assumed to be 95 m. All measurements were taken in the proximities of the P3 245 station. For 2012, 234Th-238U data were collected during cruise JR274 (Le Moigne et al. (2014)). 246 For 2013, 210Po-210Po and 234Th-238U data were collected during the Western Core Box survey 247 JCR291 (Ceballos-Romero et al. (2022)). For both cruises, average SV and ExpEff were obtained 248 as described in the methods section, together with the satellite derived NPP. As the austral bloom 249 ranges from October to February, for every bloom we included results from October to December 250 (DOY from 250 to 365) of the year and from January to March of following year (DOY from 1 251 to 85). 252 Deep sediment traps were also deployed by the British Antarctic Survey between 2012 and 2014. 253 The traps collected downward sinking material at ~1950 m with a periodicity of 1 month (Figure 254 3) from 1st December 2012 to 1st January 2014. According to the corresponding NPP profiles, 255 the material collected in January during the 2012-2013 deployment (Figure 3) corresponded to 256 the end of 2012 austral bloom, whereas material collected at the end of 2013 during the 2013257 2014 deployment (Figure 3) corresponded to material from the austral bloom of 2013. 258 259 3. RESULTS 260 3.1. NPP and POC fluxes at 1950 m depth during COMICS I 261 Our cruise captured the peak and decline of a spring bloom as shown by satellite-derived (Figure 262 2) and glider-derived NPP (Henson et al., this issue). Based on satellite-derived NPP estimates, 263 NPP started to increase in mid-September with an initial short-lived bloom (~1400 mg C m-2 d264 1) in early October. NPP steadily increased until the start of our cruise, where NPP peaked again 265 at ~1400 mg C m-2 d-1 (mid-November; P3A) before it steadily decreased during our cruise to 266 ~550 mg C m-2 d-1 (mid-December; P3C). After we left the site, a third and more pronounced 267 8 bloom occurred around mid-January to early February with satellite-derived NPP rates as high 268 as 2600 mg C m-2 d-1. Glider-derived NPP data shows the same temporal trend, though the 269 absolute magnitude of the three bloom peaks was ~1600, ~2000 and ~1900 mg C m-2 d-1 (Henson 270 et al., this issue). 271 The deep sediment trap (1950 m depth) appeared to have captured the first, short-lived bloom 272 (mid-September) in early November (DOY 311), with fluxes as high as 25 mg C m-2 d-1 (Figure 273 3) and mainly comprised of faecal pellets (60% of the total POC flux) and fresh single cell 274 diatoms (39%) (Manno et al., this issue). The second bloom (mid-November), which concurred 275 with P3A, appeared to have been captured by the deep sediment trap in early December. This 276 second deep-flux peak was comprised of fast sinking zooplankton faecal pellets (66% of the total 277 POC flux), lowly-sinking partially remineralized diatom aggregates and phytodetritus (28%), and 278 single-cell diatoms (5%) (Manno et al., this issue). 279 280 3.2. 210Po-derived POC fluxes during COMICS I 281 During both P3A and P3C, 210Po-derived POC fluxes increased with depth throughout the 282 productive layer, reaching a maximum at Zp (~95 m; i.e. ‘export flux’) with 360 ± 100 and 350 283 ± 100 mg C m2 d-1, respectively (Figure 4, Table 1). Below the Zp, 210Po-derived POC fluxes 284 decreased to minima of 200 mg C m2 d-1 and 0 mg C m2 d-1 at 350 m depth during P3A and P3C, 285 respectively. During P3B, 210Po-derived POC fluxes steadily increased from the surface to 250 286 m depth (maximum: 350 mg C m-2 d-1). 210Po-derived export flux during P3B was 153 ± 39 mg 287 C m2 d-1. Our 210Po-derived export flux estimates fall within the range of the glider-derived export 288 flux estimates (Table 1; Henson et al., this issue), though are on the lower end (mean glider289 derived POC export during P3A, P3B and P3C, respectively: 646, 409, 287 mg C m2 d-1; Giering 290 et al., this issue). As the glider fluxes were calibrated using independent flux measurements (e.g. 291 Marine Snow Catcher) (Giering et al., this issue), we speculate that the 210Po-derived POC fluxes 292 underestimated true fluxes owing to the non-steady state assumption. Although the steady-state 293 assumption is widely applied to most methods for obtaining fluxes or SV, it can introduce a 294 systematic bias (Ceballos-Romero et al., 2018; Giering et al., 2017). In addition, there are 295 temporal mismatches between both (1) flux estimates using gliders, marine snow catchers and 296 radiotracers, and (2) between the radiotracer in sinking particles (instantaneously collected to 297 estimate POC/210Po ratio) and water (210Po flux is associated to 210Po radioactive half-life of 138 298 days). The non-steady state conditions during our cruise likely introduced uncertainties in the 299 interpretation of radiotracer disequilibria as, when export increased at the beginning of the cruise, 300 the radioactive disequilibrium would not have appeared instantaneously in the water column, 301 rather it would have been delayed by a few days (Ceballos-Romero et al., 2018). Conversely, 302 when export decreased, the disequilibrium would have been maintained for a few days after. As 303 a result, during the bloom, as a general rule, the radiotracer approach underestimates POC flux 304 at the beginning and overestimates it at the end of the bloom (Ceballos-Romero et al., 2018). 305 An additional source of inaccuracy might overcome due to an under collection of 210Po from the 306 water in relation to the 209Po internal tracer used to quantify the extraction yield (Roca-Martí et 307 9 al., 2021). This can sometimes produce a small underestimation of 210Po concentrations that 308 would result in overestimated fluxes. As 210Po-derived POC fluxes here are in the lowest range 309 of all the used techniques during COMICS I, it is unlikely to expect overestimated 210Po-derived 310 POC fluxes due to under collection of 210Po in the sampling process. 311 312 3.3. Sinking velocities during COMICS I 313 210Poderived SV increased with depth for all profiles from ~ 15 m d-1 at the surface to >100 m 314 d-1 in the upper mesopelagic zone (Figure 5). In the upper 100 m, estimated SV were the same 315 for all three visits (within their uncertainties). For P3A and P3B, from approximately 15 m d-1 in 316 surface waters to 185 m d-1 at 150 m and 110 m d-1 at 250 m. During P3C, 210Poderived SV in 317 the upper 100 m were the same as during P3A and P3B considering their uncertainties and 318 compared to the standard deviation of the three values at each depth. However, at and below 150 319 m, the SVs during P3C were considerably slower (~50 m d-1) than those for P3A and P3B (110320 186 m d-1) (Figure 5; Table 1). 321 Glider-derived SV estimates were ~60 m d-1 at 140, 80 – 130 m d-1 at 250 m, and 130 m d-1 at 322 460 m (average values 70 ± 20 m d-1, 100 ± 25 m d-1, 120 ± 10 m d-1 at the depths 155 m, 250 323 and 490 m) (Figure 5; Table 1), with no clear differences observed between the visits (Table 1) 324 according to SV uncertainties and standard deviation. 325 Both approaches to estimate SV, 210Po-210Pb disequilibrium and the glider-derived backscatter, 326 provide bulk-average estimates. The main difference between the two methods is that, whereas 327 the glider derives SV estimates for particles of two sizes (~100-400 m for the small particles 328 peak and > 400 m for the large particles peak), the 210Po-210Pb disequilibrium is not associated 329 to any particle size and thus represents all sinking particles. Moreover, while the glider-based 330 method takes advantage of non-steady-state conditions (i.e. plume tracking), the radionuclide 331 approach assumes steady state (as for the flux calculations). Despite these differences, the SV 332 estimates from the gliders and from the radionuclides agree well with each other (within 333 uncertainties; Figure 5a). 334 For the depth-averaged profile, we binned the depth 10 m, 50 m, 75-100 m (nominal: 90 m), 140335 175 m (nominal: 150 m), and 350 – 520 m (nominal: 442 m), and obtained the maximum, 336 minimum and average values of the SVbinned estimates for these depths (including both 337 radionuclideand glider-derived values). 338 The SVbinned increased from 50 ± 10 m d-1 at Zp to 100 ± 50 m d-1 (150 m) (Figure 5b). Minimum 339 and maximum SVbinned estimates in the dataset also increased with depth: for minima from 35 m 340 d-1 at Zp (95 m) to 110 m d-1 (440 m), and for maxima from 70 d-1 at Zp to 130 m d-1 at 440 m. 341 The depth-averaged vertical profile of both methods showed a rapid linear increase in SVbinned in 342 the upper 150 m at a rate of 0.6 [m d-1] m-1 (linear regression: R2 = 0.99; t = 11.5, p = 0.001, n = 343 4). Below 150, average SVbinned continued to increase albeit at a slower rate of 0.1 [m d-1] m-1 344 (linear regression: R2 = 0.96; t = 4.4, p = 0.05; n = 3). When the increase is examined using the 345 individual SV data (i.e. Figure 5a), the increase in SV in the upper 150 m is statistically 346 16 Figure 1. Map of North South Georgia and new primary production, NPP (mg m-2 d-1) during 557 P3A, P3B and P3C periods 558 Figure 2. Values of SV (m d-1) vs DOY during three different blooms measured at P3 in 2012 559 (blue), 2013 (green) and 2017 (red) at the base of the Euphotic Zone (Zp). SV values are derived 560 from radioactive disequilibrium (210Po-210Pb and 234Th-238U) and gliders datasets. NPP obtained 561 from satellite images is also included during the three bloom periods. The grey hatch corresponds 562 to COMICS I occupation. 563 Figure 3. POC downward flux collected every five days using a rotating sediment trap moored 564 at 1950 m during 2017, provided by (figure adapted from Manno et al. (this issue)). The grey 565 hatch corresponds to COMICS I occupation. The sediment trap was deployed on the mooring in 566 P3, as part of the Ecosystems programme at the British Antarctic Survey (BAS) and the Scotia 567 Sea Open Ocean Laboratories (SCOOBIES) sustained observation programme at the BAS. 568 Results from 2012 and 2013 are included, together with NPP obtained from satellite images 569 during the 2017 bloom period. 570 Figure 4. POC downward flux in mg m-2 d-1 derived from 210Po-210Pb disequilibrium measured 571 during P3A (15th Nov – 22th Nov), P3B (29th Nov – 5th Dec) and P3C (9th Dec – 15th 572 Dec).Uncertainties are obtained from standard deviation and error propagation and correspond 573 to 1 . 574 Figure 5. Average sinking velocities (SV) in m d-1 depth profiles. 5a) shows SVs obtained from 575 210Po-210Pb disequilibrium and from gliders. 5b) shows average (full circles), maximum and 576 minimum (grey lines outlining the hatch) sinking velocities depth profiles calculated using the 577 data points from the two methods (showed in Figure 5a) at each depth. Broken lines correspond 578 to the linear fitting of the average values from surface to 150 m and from 150 to 450 m. 579 Figure 6. Sinking velocities derived from 210Po-210Pb disequilibrium (Bloom of years 2013 and 580 2017) and 234Th-234Th disequilibrium (bloom of year 2012) versus Export Efficiency calculated 581 as the ratio POC flux at the base of the Euphotic Zone (Zp). The linear fitting of the complete 582 dataset is included 583 584 REFERENCES: 585 Alonso-Gonzalez, I.J., Aristegui, J., Lee, C., Sanchez-Vidal, A., Calafat, A., Fabres, J., Sangra, 586 P., Masque, P., Hernandez-Guerra, A., Benitez-Barrios, V., 2010. Role of slowly settling 587 particles in the ocean carbon cycle. Geophys. Res. Lett. 37. https://doi.org/Artn 588 L13608Doi 10.1029/2010gl043827 589 Armstrong, R.A., Peterson, M.L., Lee, C., Wakeham, S.G., 2009. Settling velocity spectra and 590 the ballast ratio hypothesis. Deep. Res. Part II Top. Stud. Oceanogr. 56, 1470–1478. 591 https://doi.org/10.1016/j.dsr2.2008.11.032 592 17 Atkinson, A., Whitehouse, M.J., Priddle, J., Cripps, G.C., Ward, P., Brandon, M.A., 2001. 593 South Georgia, Antarctica: A productive, cold water, pelagic ecosystem. Mar. Ecol. Prog. 594 Ser. 216, 279–308. https://doi.org/10.3354/meps216279 595 Aumont, O., Éthé, C., Tagliabue, A., Bopp, L., Gehlen, M., Ethé, C., Tagliabue, A., Bopp, L., 596 Gehlen, M., Éthé, C., Tagliabue, A., Bopp, L., Gehlen, M., 2015. PISCES-v2: an ocean 597 biogeochemical model for carbon and ecosystem studies. Geosci. Model Dev. Discuss. 8, 598 2465–2513. https://doi.org/10.5194/gmd-8-2465-2015 599 Bach, L.T., Riebesell, U., Sett, S., Febiri, S., Rzepka, P., Schulz, K.G., 2012. An approach for 600 particle sinking velocity measurements in the 3-400 μm size range and considerations on 601 the effect of temperature on sinking rates. Mar. Biol. 159, 1853–1864. 602 https://doi.org/10.1007/s00227-012-1945-2 603 Behrenfeld, M.J., Falkowski, P.G., 1997. A consumer’s guide to phytoplankton primary 604 productivity models. Limnol. Oceanogr. 42, 1479–1491. 605 Berelson, W.M., 2002. Particle settling rates increase with depth in the ocean. Deep. Res. Part 606 Ii-Topical Stud. Oceanogr. 49, 237–251. 607 Briggs, N., Dall’Olmo, G., Claustre, H., 2020a. Major role of particle fragmentation in 608 regulating biological sequestration of CO2 by the oceans. Science (80-. ). 367, 791–793. 609 https://doi.org/10.1126/science.aay1790 610 Briggs, N., Henson, S.A., Villa-Alfageme, M., Claustre, H., Carvalho, P., Giering, S.L.C., 611 2020b. Consistent Increase in Particle Sinking Velocity with Depth Found across >30 612 High-Latitude Phytoplankton Blooms using Biogeochemical Argo, in: Ocean Sciences 613 Meeting 2020. San Diego. pp. 17-21 June 2020. 614 Briggs, N., Perry, M.J., Cetinic, I., Lee, C., D’Asaro, E., Gray, A.M., Rehm, E., 2011. High615 resolution observations of aggregate flux during a sub-polar North Atlantic spring bloom. 616 Deep Sea Res. Part I Oceanogr. Res. Pap. 58, 1031–1039. 617 Buesseler, K.O., Boyd, P.W., 2009. Shedding light on processes that control particle export and 618 flux attenuation in the twilight zone of the open ocean. Limnol. Oceanogr. 54, 1210–1232. 619 Buesseler, K.O., Boyd, P.W., Black, E.E., Siegel, D.A., 2020. Metrics that matter for assessing 620 the ocean biological carbon pump. Proc. Natl. Acad. Sci. U. S. A. 117, 9679–9687. 621 https://doi.org/10.1073/pnas.1918114117 622 Buesseler, K.O., Lamborg, C.H., Boyd, P.W., Lam, P.J., Trull, T.W., Bidigare, R.R., Bishop, 623 J.K.B., Casciotti, K.L., Dehairs, F., Elskens, M., Honda, M., Karl, D.M., Siegel, D.A., 624 Silver, M.W., Steinberg, D.K., Valdes, J., Van Mooy, B., Wilson, S., 2007. Revisiting 625 Carbon Flux Through the Ocean’s Twilight Zone. Science (80-. ). 316, 567–570. 626 https://doi.org/10.1126/science.1137959 627 Burd, A.B., Jackson, G.A., 2009. Particle aggregation. Ann. Rev. Mar. Sci. 1, 65–90. 628 https://doi.org/10.1146/annurev.marine.010908.163904 629 Cavan, E.L., Laurenceau-Cornec, E.C., Bressac, M., Boyd, P.W., 2019. Exploring the ecology 630 of the mesopelagic biological pump. Prog. Oceanogr. 176, 102125. 631 https://doi.org/10.1016/j.pocean.2019.102125 632 18 Cavan, E.L., Le Moigne, F.A.C., Poulton, A.J., Tarling, G.A., Ward, P., Daniels, C.J., Fragoso, 633 G.M., Sanders, R.J., 2015. Attenuation of particulate organic carbon flux in the Scotia Sea, 634 Southern Ocean, is controlled by zooplankton fecal pellets. Geophys. Res. Lett. 42, 821– 635 830. https://doi.org/10.1002/2014GL062744 636 Cavan, E.L., Trimmer, M., Shelley, F., Sanders, R., 2017. Remineralization of particulate 637 organic carbon in an ocean oxygen minimum zone. Nat. Commun. 8, 14847. 638 https://doi.org/10.1038/ncomms14847 639 Ceballos-Romero, E., Buesseler, K.O., Villa-Alfageme, M., 2022. Revisiting five decades of 640 234Th data: a comprehensive global oceanic compilation. Earth Syst. Sci. Data 14, 2639– 641 2679. https://doi.org/10.5194/essd-14-2639-2022 642 Ceballos-Romero, E., De Soto, F., Le Moigne, F.A.C., García-Tenorio, R., Villa-Alfageme, M., 643 2018. 234Th-Derived Particle Fluxes and Seasonal Variability: When Is the SS 644 Assumption Reliable? Insights From a Novel Approach for Carbon Flux Simulation. 645 Geophys. Res. Lett. 45. https://doi.org/10.1029/2018GL079968 646 Ceballos-Romero, E., Le Moigne, F.A.C., Henson, S., Marsay, C.M., Sanders, R.J., García647 Tenorio, R., Villa-Alfageme, M., 2016. Influence of bloom dynamics on Particle Export 648 Efficiency in the North Atlantic: a comparative study of radioanalytical techniques and 649 sediment traps. Mar. Chem. 186, 198–210. https://doi.org/10.1016/j.marchem.2016.10.001 650 Cook, K., Belcher, A., Bondyale Juez, D., Stowasser, G., Fielding, S., Saunders, R.A., Elsafi, 651 M.A., Wolf, G.A., Blackbird, S.J., Tarling, G.A., Mayor, D.J., 2023. Carbon budgets of 652 Scotia Sea mesopelagic zooplankton and micronekton communities: the importance of 653 lipids. Deep Sea Res. Part II Top. Stud. Oceanogr. This issue. 654 Cram, J.A., Weber, T., Leung, S.W., McDonnell, A.M.P., Liang, J.H., Deutsch, C., 2018. The 655 Role of Particle Size, Ballast, Temperature, and Oxygen in the Sinking Flux to the Deep 656 Sea. Global Biogeochem. Cycles 32, 858–876. https://doi.org/10.1029/2017GB005710 657 De La Rocha, C.L., Passow, U., 2007. Factors influencing the sinking of POC and the 658 efficiency of the biological carbon pump. Deep. Res. Part II Top. Stud. Oceanogr. 54, 659 639–658. https://doi.org/10.1016/j.dsr2.2007.01.004 660 de Soto, F., Ceballos-Romero, E., Villa-Alfageme, M., 2018. A stochastic model for particle 661 flux attenuation in ocean waters: application to radioactive pairs disequilibria. 662 Cosmochim. Geochim. Acta 239, 136–158. https://doi.org/10.1016/j.gca.2018.07.031 663 De Soto, F., Ceballos-Romero, E., Villa-Alfageme, M., 2018. A microscopic simulation of 664 particle flux in ocean waters: Application to radioactive pair disequilibrium. Geochim. 665 Cosmochim. Acta 239. https://doi.org/10.1016/j.gca.2018.07.031 666 DeVries, T., 2014. The oceanic anthropogenic CO2 sink: Storage, air-sea fluxes, and transports 667 over the industrial era. Global Biogeochem. Cycles 631–647. 668 https://doi.org/10.1002/2013GB004739 669 Devries, T., Liang, J.H., Deutsch, C., 2014. A mechanistic particle flux model applied to the 670 oceanic phosphorus cycle. Biogeosciences 11, 5381–5398. https://doi.org/10.5194/bg-11671 5381-2014 672 DeVries, T., Weber, T., 2017. The export and fate of organic matter in the ocean: New 673 19 constraints from combining satellite and oceanographic tracer observations. Global 674 Biogeochem. Cycles 31, 535–555. https://doi.org/10.1002/2016GB005551 675 Giering, S.L.C., Cavan, E.L., Basedow, S.L., Briggs, N., Burd, A.B., Darroch, L.J., Guidi, L., 676 Irisson, J.O., Iversen, M.H., Kiko, R., Lindsay, D., Marcolin, C.R., McDonnell, A.M.P., 677 Möller, K.O., Passow, U., Thomalla, S., Trull, T.W., Waite, A.M., 2020. Sinking Organic 678 Particles in the Ocean—Flux Estimates From in situ Optical Devices. Front. Mar. Sci. 6. 679 https://doi.org/10.3389/fmars.2019.00834 680 Giering, S.L.C., Sanders, R., Blackbird, S., Briggs, N., Carvalho, F., East, H., Espinola, B., 681 Henson, S.A., Kiriakoulakis, K., Iversen, M.H., Lampitt, R.S., Pabortsava, K., Pebody, C., 682 Peel, K., Preece, C., Saw, K., Villa-Alfageme, M., Wolff, G.A., 2023. Vertical imbalance 683 in organic carbon budgets is indicative of a missing vertical transfer during a 684 phytoplankton bloom near South Georgia (COMICS). Deep Sea Res. Part II Top. Stud. 685 Oceanogr. 209, 105277. https://doi.org/10.1016/j.dsr2.2023.105277 686 Giering, S.L.C., Sanders, R., Lampitt, R., Anderson, T.R., Tamburini, C., Boutrif, M., Zubkov, 687 M. V, Marsay, C.M., Henson, S.A., Saw, K., Cook, K., Mayor, D.J., 2014. Reconciliation 688 of the carbon budget in the ocean’s twilight zone. Nature 507, 480–483. 689 Giering, S.L.C., Sanders, R., Martin, A.P., Henson, S.A., Riley, J.S., Marsay, C.M., Johns, 690 D.G., 2017. Particle flux in the oceans: Challenging the steady state assumption. Global 691 Biogeochem. Cycles 31, 159–171. https://doi.org/10.1002/2016GB005424 692 Gloege, L., McKinley, G.A., Mouw, C.B., Ciochetto, A.B., 2017. Global evaluation of 693 particulate organic carbon flux parameterizations and implications for atmospheric pCO2. 694 Global Biogeochem. Cycles 31, 1192–1215. https://doi.org/10.1002/2016GB005535 695 Guidi, L., Jackson, G.A., Stemmann, L., Carlos Miquel, J., Picheral, M., Gorsky, G., 2008. 696 Relationship between particle size distribution and flux in the mesopelagic zone. Deep Sea 697 Res. Part I Oceanogr. Res. Pap. 1364–1374. https://doi.org/10.1016/j.dsr.2008.05.014 698 Henson, S.A., Briggs, N., Carvalho, F., Manno, C., Mignot, A., Thomalla, S., 2023. A seasonal 699 transition in biological carbon pump efficiency in the northern Scotia Sea, Southern 700 Ocean. Deep Sea Res. Part II This issue. 701 Henson, S.A., Sanders, R., Madsen, E., 2012. Global patterns in efficiency of particulate 702 organic carbon export and transfer to the deep ocean. Global Biogeochem. Cycles 26, n/a703 n/a. https://doi.org/10.1029/2011GB004099 704 Henson, S.A., Sanders, R., Madsen, E. estimate of the strength of the ocean’s biological carbon 705 pump, Morris, P.J., Le Moigne, F., Quartly, G.D., 2011. A reduced estimate of the 706 strength of the ocean’s biological carbon pump. Geophys. Res. Lett. 38, n/a-n/a. 707 https://doi.org/10.1029/2011GL046735 708 Iversen, M.H., Lampitt, R.S., 2020. Size does not matter after all: No evidence for a size709 sinking relationship for marine snow. Prog. Oceanogr. 189, 102445. 710 https://doi.org/10.1016/j.pocean.2020.102445 711 Iversen, M.H., Nowald, N., Ploug, H., Jackson, G.A., Fischer, G., 2010. High resolution 712 profiles of vertical particulate organic matter export off Cape Blanc, Mauritania: 713 Degradation processes and ballasting effects. Deep Sea Res. Part A. Oceanogr. Res. Pap. 714 20 771–784. https://doi.org/10.1016/j.dsr.2010.03.007 715 Iversen, M.H., Ploug, H., 2013. Temperature effects on carbon-specific respiration rate and 716 sinking velocity of diatom aggregates‚Äìpotential implications for deep ocean export 717 processes. Biogeosciences 10, 4073–4085. 718 Jokulsdottir, T., Archer, D., 2016. A stochastic, Lagrangian model of sinking biogenic 719 aggregates in the ocean (SLAMS 1.0): model formulation, validation and sensitivity. 720 Geosci. Model Dev. 9, 1455–1476. https://doi.org/10.5194/gmd-9-1455-2016 721 Jones, E.M., Bakker, D.C.E., Venables, H.J., Watson, A.J., 2012. Dynamic seasonal cycling of 722 inorganic carbon downstream of South Georgia, Southern Ocean. Deep. Res. Part II Top. 723 Stud. Oceanogr. 59–60, 25–35. https://doi.org/10.1016/j.dsr2.2011.08.001 724 Kiko, R., Biastoch, A., Brandt, P., Cravatte, S., Hauss, H., Hummels, R., Kriest, I., Marin, F., 725 McDonnell, A.M.P., Oschlies, A., Picheral, M., Schwarzkopf, F.U., Thurnherr, A.M., 726 Stemmann, L., 2017. Biological and physical influences on marine snowfall at the equator. 727 Nat. Geosci. 10, 852–858. https://doi.org/10.1038/ngeo3042 728 Korb, R.E., Whitehouse, M.J., Gordon, M., Ward, P., Poulton, A.J., 2010. Summer 729 microplankton community structure across the Scotia Sea: Implications for biological 730 carbon export. Biogeosciences 7, 343–356. https://doi.org/10.5194/bg-7-343-2010 731 Korb, R.E., Whitehouse, M.J., Ward, P., Gordon, M., Venables, H.J., Poulton, A.J., 2012. 732 Regional and seasonal differences in microplankton biomass, productivity, and structure 733 across the Scotia Sea: Implications for the export of biogenic carbon. Deep. Res. Part Ii734 Topical Stud. Oceanogr. 59, 67–77. https://doi.org/Doi 10.1016/J.Dsr2.2011.06.006 735 Kriest, I., Oschlies, A., 2008. On the treatment of particulate organic matter sinking in large736 scale models of marine biogeochemical cycles. Biogeosciences 5, 55–72. 737 Laurenceau-Cornec, E.C., Le Moigne, F.A.C., Gallinari, M., Moriceau, B., Toullec, J., Iversen, 738 M.H., Engel, A., De La Rocha, C.L., 2020. New guidelines for the application of Stokes’ 739 models to the sinking velocity of marine aggregates. Limnol. Oceanogr. 65, 1264–1285. 740 https://doi.org/10.1002/lno.11388 741 Le Moigne, F.A.C., Pabortsava, K., Marcinko, C.L.J., Martin, P., Sanders, R.J., 2014. Where is 742 mineral ballast important for surface export of particulate organic carbon in the ocean? 743 Geophys. Res. Lett. 41, 8460–8468. https://doi.org/10.1002/2014GL061678 744 Le Moigne, F.A.C., Villa-Alfageme, M., Sanders, R.J.J., Marsay, C., Henson, S., García745 Tenorio, R., 2013. Export of organic carbon and biominerals derived from 234Th and 746 210Po at the Porcupine Abyssal Plain. Deep. Res. Part I Oceanogr. Res. Pap. 72, 88–101. 747 https://doi.org/10.1016/j.dsr.2012.10.010 748 Le Moigne, F.A.C.C., Henson, S.A., Cavan, E., Georges, C., Pabortsava, K., Achterberg, E.P., 749 Ceballos‐Romero, E., Zubkov, M., Sanders, R.J., Ceballos-Romero, E., Zubkov, M., 750 Sanders, R.J., Ceballos‐Romero, E., Zubkov, M., Sanders, R.J., 2016. What causes the 751 inverse relationship between primary production and export efficiency in the Southern 752 Ocean? Geophys. Res. Lett. 43, 4457–4466. https://doi.org/10.1002/2016GL068480 753 Lee, C., Peterson, M.L., Wakeham, S.G., Armstrong, R.A., Cochran, J.K., Miquel, J.C., 754 Fowler, S.W., Hirschberg, D., Beck, A., Xue, J.H., 2009. Particulate organic matter and 755 21 ballast fluxes measured using time-series and settling velocity sediment traps in the 756 northwestern Mediterranean Sea. Deep. Res. Part Ii-Topical Stud. Oceanogr. 56, 1420– 757 1436. https://doi.org/Doi 10.1016/J.Dsr2.2008.11.029 758 Manno, C., Stowasser, G., Enderlein, P., Fielding, S., Tarling, G.A., 2015. The contribution of 759 zooplankton faecal pellets to deep-carbon transport in the Scotia Sea (Southern Ocean). 760 Biogeosciences 12, 1955–1965. https://doi.org/10.5194/bg-12-1955-2015 761 Manno, C., Stowasser, G., Fielding, S., Apeland, B., Tarling, G.A., 2023. Multiple peaks in 762 deep carbon export during the spring bloom in the Scotia Sea (Southern Ocean) are driven 763 by different biological pathways. Deep Sea Res. Part II This issue. 764 Marsay, C.M., Sanders, R.J., Henson, S.A., Pabortsava, K., Achterberg, E.P., Lampitt, R.S., 765 2015. Attenuation of sinking particulate organic carbon flux through the mesopelagic 766 ocean. Proc. Natl. Acad. Sci. U. S. A. 112, 1089–1094. 767 https://doi.org/10.1073/pnas.1415311112 768 Martin, Knauer, G.A., Karl, D.M., Broenkow, W.W., 1987. Vertex - Carbon Cycling in the 769 Northeast Pacific. Deep. Res. Part a-Oceanographic Res. Pap. 34, 267–285. 770 https://doi.org/Doi 10.1016/0198-0149(87)90086-0 771 Mas, J.L., Martín, J., Pham, M.K., Chamizo, E., Miquel, J.C., Osvath, I., Povinec, P.P., 772 Eriksson, M., Villa-Alfageme, M., 2020. Analysis of a major Aeolian dust input event and 773 its impact on element fluxes and inventories at the DYFAMED site (Northwestern 774 Mediterranean). Mar. Chem. 223. https://doi.org/10.1016/j.marchem.2020.103792 775 McDonnell, A.M.P., Buesseler, K.O., 2012. A new method for the estimation of sinking 776 particle fluxes from measurements of the particle size distribution, average sinking 777 velocity, and carbon content. Limnol. Oceanogr. Methods 10, 329–346. 778 https://doi.org/10.4319/lom.2012.10.329 779 McDonnell, A.M.P., Buesseler, K.O., 2010. Variability in the average sinking velocity of 780 marine particles. Limnol. Oceanogr. 55, 2085–2096. https://doi.org/Doi 781 10.4319/Lo.2010.55.5.2085 782 Mignot, A., Ferrari, R., Claustre, H., 2018. Floats with bio-optical sensors reveal what 783 processes trigger the North Atlantic bloom. Nat. Commun. 9, 190. 784 https://doi.org/10.1038/s41467-017-02143-6 785 Omand, M.M., Govindarajan, R., He, J., Mahadevan, A., 2020. Sinking flux of particulate 786 organic matter in the oceans: Sensitivity to particle characteristics. Sci. Rep. 10, 5582. 787 https://doi.org/10.1038/s41598-020-60424-5 788 Owens, S.A., Pike, S., Buesseler, K.O., 2015. Thorium-234 as a tracer of particle dynamics and 789 upper ocean export in the Atlantic Ocean. Deep. Res. Part II Top. Stud. Oceanogr. 116, 790 42–59. https://doi.org/10.1016/j.dsr2.2014.11.010 791 Rayne, R.R.-P., Giering, S.L.C., Hartmann, M., Sanders, R.D., Evans, C., 2023. Temporal 792 shifts in prokaryotic metabolism in response to organic carbon dynamics in the 793 mesopelagic ocean during an export event in the Southern Ocean. Deep Sea Res. Part II 794 Top. Stud. Oceanogr. This issue. 795 Rembauville, M., Manno, C., Tarling, G.A., Blain, S., Salter, I., 2016. Strong contribution of 796 22 diatom resting spores to deep-sea carbon transfer in naturally iron-fertilized waters 797 downstream of South Georgia. Deep. Res. Part I Oceanogr. Res. Pap. 115, 22–35. 798 https://doi.org/10.1016/j.dsr.2016.05.002 799 Robinson, J., Popova, E.E., Srokosz, M.A., Yool, A., 2016. A tale of three islands: 800 Downstream natural iron fertilization in the Southern Ocean. J. Geophys. Res. Ocean. 121, 801 3350–3371. https://doi.org/10.1002/2015JC011319 802 Roca-Martí, M., Biogeochemistry, M., Puigcorbé, V., Castrillejo, M., Casacuberta, N., Garcia803 Orellana, J., Kirk Cochran, J., Masqué, P., 2021. Quantifying 210 Po/ 210 Pb 804 Disequilibrium in Seawater: A Comparison of Two Precipitation Methods With Differing 805 Results. Front. Mar. Sci. https://doi.org/10.3389/fmars.2021.684484 806 Sanders, R.J., Henson, S.A., Martin, A.P., Anderson, T.R., Bernardello, R., Enderlein, P., 807 Fielding, S., Giering, S.L.C., Hartmann, M., Iversen, M., Khatiwala, S., Lam, P., Lampitt, 808 R., Mayor, D.J., Moore, M.C., Murphy, E., Painter, S.C., Poulton, A.J., Saw, K., 809 Stowasser, G., Tarling, G.A., Torres-Valdes, S., Trimmer, M., Wolff, G.A., Yool, A., 810 Zubkov, M., 2016. Controls over Ocean Mesopelagic Interior Carbon Storage (COMICS): 811 Fieldwork, Synthesis, and Modeling Efforts. Front. Mar. Sci. 3. 812 https://doi.org/10.3389/fmars.2016.00136 813 Sarmiento, J.L., Gruber, N., 2006. Ocean biogeochemical dynamics. Princeton University 814 Press. 815 Savoye, N., Benitez-Nelson, C., Burd, A.B., Cochran, J.K., Charette, M., Buesseler, K.O., 816 Jackson, G.A., Roy-Barman, M., Schmidt, S., Elskens, M., 2006. Th-234 sorption and 817 export models in the water column: A review. Mar. Chem. 100, 234–249. 818 https://doi.org/Doi 10.1016/J.Marchem.2005.10.014 819 Sternberg, R.W., Berhane, I., Ogston, A.S., 1999. Measurement of size and settling velocity of 820 suspended aggregates on the northern California continental shelf. Mar. Geol. 154, 43–53. 821 Turner, J.T., 2015. Zooplankton fecal pellets, marine snow, phytodetritus and the ocean’s 822 biological pump. Prog. Oceanogr. 130, 205–248. 823 https://doi.org/10.1016/j.pocean.2014.08.005 824 van der Jagt, H., Friese, C., Stuut, J.B.W., Fischer, G., Iversen, M.H., 2018. The ballasting 825 effect of Saharan dust deposition on aggregate dynamics and carbon export: Aggregation, 826 settling, and scavenging potential of marine snow. Limnol. Oceanogr. 63, 1386–1394. 827 https://doi.org/10.1002/lno.10779 828 Verdeny, E., Masque, P., Garcia-Orellana, J., Hanfland, C., Cochran, J.K., Stewart, G.M., 829 2009. POC export from ocean surface waters by means of Th-234/U-238 and Po-210/Pb830 210 disequilibria: A review of the use of two radiotracer pairs. Deep. Res. Part Ii-Topical 831 Stud. Oceanogr. 56, 1502–1518. https://doi.org/10.1016/j.dsr2.2008.12.018 832 Villa-Alfageme, M., De Soto, F., Le Moigne, F.A.C., Giering, S.L.C., Sanders, R., García833 Tenorio, R., 2014. Observations and modeling of slow-sinking particles in the twilight 834 zone. Global Biogeochem. Cycles 28. https://doi.org/10.1002/2014GB004981 835 Villa-Alfageme, M., de Soto, F.C., Ceballos, E., Giering, S.L.C., Le Moigne, F.A.C., Henson, 836 S., Mas, J.L., Sanders, R.J., 2016a. Geographical, seasonal, and depth variation in sinking 837 23 particle speeds in the North Atlantic. Geophys. Res. Lett. 43. 838 https://doi.org/10.1002/2016GL069233 839 Villa-Alfageme, M., Mas, J.L., Hurtado-Bermudez, S., Masqué, P., 2016b. Rapid determination 840 of 210Pb and 210Po in water and application to marine samples. Talanta 160. 841 https://doi.org/10.1016/j.talanta.2016.06.051 842 Wiedmann, I., Reigstad, M., Sundfjord, A., Basedow, S., 2014. Potential drivers of sinking 843 particle’s size spectra and vertical flux of particulate organic carbon (POC): Turbulence, 844 phytoplankton, and zooplankton. J. Geophys. Res. Ocean. 119, 6900–6917. 845 Yool, A., Popova, E.E., Anderson, T.R., 2013. MEDUSA-2.0: An intermediate complexity 846 biogeochemical model of the marine carbon cycle for climate change and ocean 847 acidification studies. Geosci. Model Dev. 6, 1767–1811. https://doi.org/10.5194/gmd-6848 1767-2013 849 Zetsche, E.M., Larsson, A.I., Iversen, M.H., Ploug, H., 2020. Flow and diffusion around and 850 within diatom aggregates: Effects of aggregate composition and shape. Limnol. Oceanogr. 851 65, 1818–1833. https://doi.org/10.1002/lno.11420 852 853