TYPE Original Research PUBLISHED 22 June 2023 DOI 10.3389/fmars.2023.1159884 Reduction in size of the OPEN ACCESS calcifying phytoplankton EDITED BY Pedro Cermeno, Institute of Marine Sciences (CSIC), Spain Calcidiscus leptoporus to REVIEWED BY Maria Vernet, environmental changes between University of California, San Diego, United States Nina Bednarsek, the Holocene and modern National Institute of Biology (NIB), Slovenia *CORRESPONDENCE Subantarctic Southern Ocean Andrés S. Rigual-Hernández arigual@usal.es Andrés S. Rigual-Hernández 1*, Gerald Langer 2, RECEIVED 06 February 2023 ACCEPTED 29 May 2023 Francisco Javier Sierro 1, Helen Bostock 3, PUBLISHED 22 June 2023 José Manuel Sánchez-Santos 4, Scott Davidson Nodder 5, CITATION Tom W. Trull 6,7, Anne Marie Ballegeer 8, Andrew D. Moy 7,9, Rigual-Hernández AS, Langer G, Sierro FJ, Bostock H, Sánchez-Santos JM, Ruth Eriksen 6,10, Laura Makowka 1, Thibauld M. Béjard 1, Nodder SD, Trull TW, Ballegeer AM, Moy AD, Eriksen R, Makowka L, Béjard TM, Francisco Henri Rigal-Muñoz 1, Alberto Hernández-Martı́n 1, Rigal-Muñoz FH, Hernández-Martı´n A, Marı́a Zorita-Viota 1 and José Abel Flores 1 Zorita-Viota M and Flores JA (2023) Reduction in size of the calcifying 1 Área de Paleontologı´a, Departamento de Geologı´a, Universidad de Salamanca, Salamanca, Spain, phytoplankton Calcidiscus leptoporus to 2 Institut de Ciència i Tecnologia Ambientals, Autonomous University of Barcelona, Bellaterra, Spain, environmental changes between the 3 School of Earth and Environmental Sciences, University of Queensland, Brisbane, QLD, Australia, Holocene and modern Subantarctic 4 Departamento de Estadı´stica, Universidad de Salamanca, Salamanca, Spain, 5 National Institute of Southern Ocean. Water and Atmospheric Research, Wellington, New Zealand, 6 Commonwealth Scientific and Industrial Front. Mar. Sci. 10:1159884. Research Organisation, Oceans and Atmosphere, Hobart, TAS, Australia, 7 Australian Antarctic Program doi: 10.3389/fmars.2023.1159884 Partnership, Institute of Marine and Antarctic Studies, University of Tasmania, Hobart, TAS, Australia, COPYRIGHT 8 Departamento de Didáctica de las Matemáticas y de las Ciencias Experimentales, Universidad de © 2023 Rigual-Hernández, Langer, Sierro, Salamanca, Salamanca, Spain, 9 Australian Antarctic Division, Kingston, TAS, Australia, 10 Institute for Bostock, Sánchez-Santos, Nodder, Trull, Marine and Antarctic Studies, University of Tasmania, Hobart, TAS, Australia Ballegeer, Moy, Eriksen, Makowka, Béjard, Rigal-Muñoz, Hernández-Martı´n, Zorita-Viota and Flores. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The The Subantarctic Zone of the Southern Ocean plays a disproportionally large role use, distribution or reproduction in other forums is permitted, provided the original on the Earth system. Model projections predict rapid environmental change in author(s) and the copyright owner(s) are the coming decades, including ocean acidification, warming, and changes in credited and that the original publication in nutrient supply which pose a serious risk for marine ecosystems. Yet despite the this journal is cited, in accordance with accepted academic practice. No use, importance of the Subantarctic Zone, annual and inter-annual time series are distribution or reproduction is permitted extremely rare, leading to important uncertainties about the current state of its which does not comply with these terms. ecosystems and hindering predictions of future response to climate change. Moreover, as the longest observational time series available are only a few decades long, it remains unknown whether marine pelagic ecosystems have already responded to ongoing environmental change during the industrial era. Here, we take advantage of multiple sampling efforts – monitoring of surface layer water properties together with sediment trap, seafloor sediment and sediment core sampling – to reconstruct the modern and pre-industrial state of the keystone calcifying phytoplankton Calcidiscus leptoporus, central to the global marine carbonate cycle. Morphometric measurements reveal that modern C. leptoporus coccoliths are 15% lighter and 25% smaller than those preserved in the underlying Holocene-aged sediments. The cumulative effect of multiple environmental factors appears responsible for the coccolith size Frontiers in Marine Science 01 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 variations since the Last Deglaciation, with warming and ocean acidification most likely playing a predominant role during the industrial era. Notably, extrapolation of our results suggests a future reduction in cell and coccolith size which will have a negative impact on the efficiency of the biological pump in the Southern Ocean through a reduction of carbonate ballasting. Lastly, our results tentatively suggest that C. leptoporus coccolith size could be used as a palaeo-proxy for growth rate. Future culture experiments will be needed to test this hypothesis. KEYWORDS Southern Ocean, coccolithophores, Calcidiscus leptoporus, ocean acidification, environmental change, sediment trap experiment, industrial era, Holocene 1 Introduction systems than at lower latitudes, Southern Ocean ecosystems have been proposed as bellwethers for prospective impacts of ocean Coccolithophores, the most prolific group of marine calcifying acidification on marine systems at mid- and low latitudes (Fabry phytoplankton, play an important, yet complex role in the global et al., 2009). The bulk of evidence suggests that the capacity of many carbon cycle and climate system (Milliman, 1993; Honjo et al., calcifying marine organism to build their shells will be impaired by 2008). On the one hand, they represent a key component of the enhanced oceanic uptake of anthropogenic CO2 (a process known biological pump taking up CO2 via photosynthesis, and then as ocean acidification) with potentially profound consequences for subsequent transport to depth of a fraction of the produced the global carbon cycle (Caldeira and Wickett, 2003; Orr et al., 2005; organic carbon (Ziveri et al., 2007; Balch, 2018). Moreover, their Pörtner et al., 2005; Raven et al., 2005). Coccolithophores are no dense calcite platelets (termed coccoliths) may even accelerate the exception to this trend. Laboratory and mesocosm experiments sinking of other organic matter via ballasting (e.g. (e.g. Armstrong have demonstrated that ongoing ocean acidification will most likely et al., 2002; Ziveri et al., 2007; Armstrong et al., 2009). On the other result in an overall decrease of calcification and the cellular PIC/ hand, the precipitation of carbonate coccolith calcite counteracts POC ratios in two of the most abundant coccolithophore species: the oceanic sequestration of atmospheric CO2 by elevating the Emiliania huxleyi and Gephyrocapsa oceanica (Meyer and Riebesell, aqueous CO2 by ~0.6 mole per mole of biogenic carbonate 2015). However, the response to ocean acidification is species- and precipitated (Frankignoulle and Gattuso, 1993; Frankignoulle strain-specific, both in E. huxleyi and C. leptoporus (Langer et al., et al., 1994), the so called carbonate counter pump. Additional 2006; Langer et al., 2009; Fiorini et al., 2011; Langer et al., 2011; potential impact on climate comes from the their role in the Langer et al., 2011; Diner et al., 2015). However, despite this generation of sulphur species and their possible role in cloud variability a general trend can be identified, both in E. huxleyi formation (Jackson and Gabric, 2022). Therefore, changes in the and in C. leptoporus (Diner et al., 2015; Meyer and Riebesell, 2015). distribution, abundance and/or composition of coccolithophore When considering the range of seawater carbonate chemistry communities in the global ocean could have profound impacts on relevant to the discussion of ocean acidification it is clear that C. the ocean carbon cycle with feedbacks to climate. This study looks leptoporus is also negatively impacted by acidification (Langer and at how environmental changes from the Late Deglaciation to the Bode, 2011; Diner et al., 2015). Notably, despite the obvious modern have impacted on Calcidiscus leptoporus, one of the most importance of these less abundant coccolithophore species, such important calcium carbonate vectors to the ocean interior on the as C. leptoporus, as carbonate vectors to the deep ocean, many Subantarctic Zone (SAZ), and how this may help to predict the uncertainties regarding their possible response to ongoing future response of this biogeochemical important species to environmental change in the Southern Ocean remain. A major ongoing environmental change in the Southern Ocean. reason for these uncertainties is the fact that our knowledge about The Southern Ocean also plays a significant role in the uptake of the susceptibility of coccolithophores to ocean acidification is CO2 due to the moderate alkalinity of its surface waters and to the largely based on laboratory studies. Field studies are a necessary increased solubility of CO2 at low temperatures, making it and complementary line of research that needs to be explored in particularly susceptible to ocean acidification (Orr et al., 2005; order to better predict the fate of coccolithophores under Fabry et al., 2009). Models predict that Southern Ocean waters climate change. (south of 60°S) will become undersaturated with respect to The subantarctic region is a globally important region for the aragonite as early as 2030, while calcite undersaturation (the most productivity of coccolithophores, termed the Great Calcite Belt stable polymorph of calcium carbonate) is expected to begin by the (Balch et al., 2011; Rosengard et al., 2015). Although more recent year 2095 (McNeil and Matear, 2008). Since these carbonate observations have found important spatial differences in saturation thresholds will be crossed sooner in high latitude coccolithophore productivity across sectors of across the SAZ Frontiers in Marine Science 02 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 (Trull et al., 2018). Southern Ocean coccolithophore communities of New Zealand. SOTS station (Trull et al., 2010; Shadwick et al., are largely dominated by the cosmopolitan species Emiliania 2022) is located at 46° 56’ S and 142° 15’ E on the abyssal plain south huxleyi. However, owing to the small size of E. huxleyi of Tasmania and can be considered representative of a large swath coccospheres and coccoliths, bigger species (such as Calcidiscus of the of the central SubAntarctic Zone (SAZ) (Trull et al., 2001; leptoporus and Helicosphaera carteri) account for a larger fraction of Rigual-Herná ndez et al., 2020b). The SOTS observatory is equipped the calcium carbonate production and export to the deep ocean with a sediment trap at 1000 m depth as well as instruments (Young and Ziveri, 2000; Baumann et al., 2004; Daniels et al., 2016; measuring multiple physical, chemical and biological parameters Rigual-Herná ndez et al., 2020c). Notably, despite the obvious of both the atmosphere and the mixed layer (e.g. Eriksen et al., 2018; importance of these less abundant coccolithophore species as Shadwick et al., 2022). SAM was deployed in subantarctic waters at carbonate vectors to the deep ocean, it is unclear how they will 46° 40’ S and 178° 30’ E between 2000 and 2012 with a sediment respond to ongoing environmental change. trap at 1500 m and although it has a series of instruments Comparison of materials collected from sediment traps measuring oceanographic parameters at depth it has no (representative of the industrial era) and seafloor sediments information for surface waters (Nodder et al., 2016). (representative of pre-industrial times) have been used to evaluate Here we present C. leptoporus data covering one annual cycle the response of key calcifying microplankton to recent for both sites. For the SOTS site samples were retrieved between environmental change (e.g. Jonkers et al., 2019; Bé jard et al., August 2011 and July 2012 with a resolution of 16 days (21 samples 2023). Moy et al. (2009) demonstrated that modern populations covering a total 336 days) at 1000 m depth. For the SAM moored of the planktonic foraminifera Globigerina bulloides are about 30- sediment trap, samples were retrieved between November 2011 and 35% lighter than those from the pre-industrial era in the SAZ and October 2012 with a resolution of ~8 days (25 samples covering a attributed this change to a physiological response of G. bulloides to total of 191 days). Gaps over the collection interval were quasi- ongoing human-induced ocean acidification. In contrast, a similar evenly distributed throughout the time series. study looking at the morphometrics of E. huxleyi coccoliths from pre-industrial and modern sediment traps tentatively suggested that 2.1.2 Seafloor sediments this species has not experienced substantial changes in their degree A total of 23 sea floor sediment samples from the subtropical of calcification (Rigual-Herná ndez et al., 2020a). However, E. and subantarctic zones south of Tasmania (Indian Ocean) and huxleyi is known to have eco-physiological features different from southeast of Aotearoa New Zealand (Southwest Pacific Ocean) were other coccolithophore species (Walker et al., 2018). Thus, it is of processed and analyzed for C. leptoporus coccolith morphometric critical importance to assess how important carbonate vectors to the characterization. Sample collection devices include gravity corers ocean interior, such as C. leptoporus, are responding and will (topmost sample) and multicorers (Table 1). Eighteen out of the 23 respond to natural and anthropogenic driven environmental samples were retrieved from settings above the calcite saturation changes in the Southern Ocean. horizon located at ~3200-3400 m (Moy et al., 2009; Bostock This study aims to shed light on these issues by estimating et al., 2011). coccolith morphometric parameters of subantarctic C. leptoporus Radiocarbon ages obtained on planktonic foraminifera for a populations from the southwest Pacific sector of the Southern subset of the samples were available from previous studies. The sea Ocean. The study used samples from two sediment traps, a set of floor sediments south of Tasmania were dated between 3 and 7 kyrs Holocene-aged samples retrieved from the underlying sediments BP: AMS 14C dates for GC04 at a depth of 5-8 cm is 3160 yr BP, (dated between 0.6 to 8 kyr) and a sediment core covering part of GC17 at 0-2 cm is 4440 yr BP and GC14 at 5-7 cm is 7373 yr BP the Deglaciation (from ~ 14 kyr) and the Holocene from the (Connell and Sikes, 1997; Moy et al., 2009). Sediment samples from Subantarctic Zone. Comparison of the data sets with key southeast New Zealand were dated between 0.6 and 8 kyr BP: AMS environmental factors allowed (1) the reconstruction of coccolith 14C dates for the top centimetre for F150 is 609 ± 50 yr BP, S631 is morphometric variations throughout the Holocene and the industrial 1502 ± 35 yr BP, J484 is 4120 ± 25 yr BP, J1049 is 6760 ± 35 yr BP era and (2) identification of key environmental factors controlling C. and 8044 ± 35 yr BP (Prebble et al., 2013; Cortese and Prebble, leptoporus coccolith morphogenesis. Results of this study provides the 2015). It is worth noting that that it is possible the top section of the first field-based assessment of the impact of ongoing environmental piston and gravity cores may have experienced some degree of change on the biogeochemically important species C. leptoporus. stretching or loss of material during the recovery of the cores, which is consistent with the age of thousands of years suggested by some of the core tops. Based on all the above, it can be concluded that the 2 Material and methods bulk of the coccolith assemblages from the seafloor sediment samples of both sectors of the SAZ where produced during the 2.1 Samples Holocene under pre-industrial CO2 levels of roughly 260-280 ppmv (Petit et al., 1999). 2.1.1 Sediment traps Samples were used from two sediment traps from the 2.1.3 Sediment core Y16 Subantarctic SW Pacific – Southern Ocean Time Series (SOTS) – A 3-m-long Kasten core Y16 (50° 35.43’ S, 169° 45.33’ E; South of Tasmania, and SubAntarctic Mooring (SAM) – Southeast Figure 1) was collected from the Campbell Plateau at 600 m Frontiers in Marine Science 03 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 TABLE 1 Station identification code, sector, zonal system, latitude, longitude, water depth and sampling device for all the seafloor sediment and sediment core samples analyzed in this study. Station ID Sector Zonal system Latitude Longitude Water depth (m) Sampling device GC01 Indian STZ -44.17 144.18 4262 Gravity core GC04 Indian STZ -44.10 144.25 2980 Gravity core GC05 Indian STZ -44.07 145.42 2334 Gravity core GC13 Indian STZ -46.17 144.27 4452 Gravity core GC14 Indian STZ -46.45 145.24 3360 Gravity core GC15 Indian STZ -46.65 145.42 3260 Gravity core GC16 Indian STZ -46.80 145.25 3523 Gravity core GC17 Indian SAZ -47.75 145.82 3001 Gravity core GC20 Indian SAZ -48.65 146.43 3300 Gravity core GC21 Indian SAZ -49.00 145.99 4132 Gravity core GC28 Indian STZ -46.06 147.38 3065 Gravity core GC31 Indian STZ -44.55 149.06 3402 Gravity core GC35 Indian STZ -45.73 146.53 2720 Gravity core MD972107 Indian SAZ -47.71 145.78 2950 Gravity core J1049 Pacific SAZ -44.50 178.73 1339 Gravity core S631 Pacific SAZ -45.33 172.11 1329 Dredge, pipe H562 Pacific SAZ -45.87 175.49 1777 Grab H555 Pacific SAZ -45.92 178.98 2738 Gravity core F150 Pacific SAZ -49.47 174.47 501 Trawl, Agassiz medium with sediment sampler, TAN1107/09 Pacific SAZ -46.60 178.54 2778 Muticorer J484 Pacific SAZ -50.58 168.99 575 Grab Y16 Pacific SAZ -50.59 169.76 600 Kasten core water depth and archived at the National Institute of Water and on a glass slide after Flores and Sierro (1997) for all samples. This Atmospheric Research (NIWA) (Carter et al., 2002; Neil et al., was used to estimate coccosphere fluxes and coccolith 2004). A total of 16 samples were subsampled at 5 cm intervals of morphometric analyses. Since marine ‘snow’ aggregates present in the top 76 cm of the Y16 core covering the last 47 kyrs. The age sediment trap samples can introduce biases in coccolith flux model was taken from Neil et al. (2004), with chronological estimations (Bairbakhish et al., 1999), a method for the non- constraints from oxygen isotopes and radiocarbon dates on destructive disintegration of aggregates was followed for the planktic foraminifera Globorotalia inflata. preparation of the second glass slide (see Rigual-Herná ndez et al., 2020b for more details). Qualitative and quantitative analyses on all the sediment trap 2.2 Calcareous nanoplankton sample samples were performed using a Nikon Eclipse LV100 POL light preparation and analysis microscope equipped with circular polarization at 1000x magnification. The taxonomic concepts of Young et al. (2003) Sediment trap, sediment grab and sediment core samples were and the Nannotax website (Young et al., 2022) were used for prepared for coccolithophore analyses following the methodology coccolith species identification. A minimum of 300 coccoliths per outlined by Rigual-Herná ndez et al. (2020b). A total of 21 and 25 sample was counted. These data have already been published in sediment trap samples were processed for the SOTS and SAM sites, Rigual-Herná ndez et al. (2020c) and are used here to estimate respectively. For the SOTS site, a two times 1/40th aliquots were annual-flux weighted averages of C. leptoporus. dedicated to pH determination and phytoplankton analysis. The pH Since the size ranges of C. leptoporus morphotypes can exceed of all samples ranged between 8.43 to 8.68, thereby guaranteeing an or be less than those proposed in the literature (e.g. Malinverno optimum calcite preservation. For the SAM samples, a 1/25th split et al., 2015), analysis of the fine structure of their coccoliths using was used for calcareous nannoplankton analysis. Initial analyses SEM microscopy is essential to guarantee a correct morphotype using a volume of 1000 to 5000 ml of the raw sample was mounted identification. Therefore, a Carl Zeiss EVO HD25 Scanning Frontiers in Marine Science 04 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 FIGURE 1 Bathymetric map showing the location of the Southern Ocean Time Series (SOTS) observatory (yellow triangle), Subantarctic Mooring (SAM) (yellow triangle), seafloor sediment samples (grey circles) and sediment core Y16 (red diamond). STZ, Subtropical Zone; SAZ, Subantarctic Zone; PFZ, Polar Frontal Zone; STF, Subtropical Front; SAF, Subantarctic Front. Oceanic fronts after Orsi et al. (1995). Ocean Data View software (Schlitzer, 2018) was used to generate this figure. Bathymetric data from General Bathymetric Chart of the Oceans (2022). Electron Microscope (SEM) was used on all samples to identify and the same rhabdolith was taken at the beginning of each imaging classify C. leptoporus morphotypes. Samples were mounted on session and the microscope and camera settings were kept constant round glass cover slips following the same settling technique throughout each session. Then, C. leptoporus coccoliths were explained above, then the cover slips were mounted on selected and images cropped with Photoshop software. Lastly, all aluminium stubs and coated with gold. Calcidiscus leptoporus digital images were processed with C-Calcita software (Fuertes exhibits a high intraspecific genetic variability, which is reflected et al., 2014) that estimated coccolith maximum diameter, area, in the morphology of its coccoliths (Sá ez et al., 2003; Young et al., volume and weight. Coccolith thickness was estimated by dividing 2022). Currently, three C. leptoporus morphotypes are recognized volume by area. To evaluate the variability across imaging sessions, based on the size and fine structure of their coccoliths: the small the same C. leptoporus coccolith (hereinafter referred to as the form or subspecies small (coccolith diameter <5 μm); the “calibration coccolith”) was imaged after every calibration (Rigual- Intermediate type or subspecies leptoporus (5-8 μm); and the Herná ndez et al., 2020b). The standard deviation of the calibration large form or subspecies quadriperforatus (7-9 μm) (Young coccolith indicate a calibration error of ±2% and ±3% for coccolith et al., 2022). mass and length, respectively. Since the birefringence-based method Calcidiscus leptoporus coccolith counts were used to estimate used in this study employs grey scale images to estimate coccolith coccolith fluxes by applying the following the formula: thickness, it can be only applicable to coccoliths thinner than 1.55 μm (Bolton et al., 2016). As maximum thickness of all measured C. N  na A V S leptoporus was lower than 0.74 μm, this threshold does not F= dT represent a limitation in our analysis. where “F” is the daily coccolith flux (coccoliths m-2 d-1), “N” the number of coccolith counted in a given sample, “A” the total area of a Petri dish, “n” the number of fields of view analyzed, “a” the area 2.4 Statistical analyses of a field of view, “V” the dilution volume, “S” the split of original sediment trap cup analyzed, “d” the number of days of collection 2.4.1 Seasonal environmental controls on and “T” the aperture area of the sediment trap. coccolith morphometrics at SOTS Canonical Correspondence Analysis (CCA; ter Braak and Verdonschot, 1995) was used to identify the main environmental 2.3 Coccolith mass and size measurements controls on C. leptoporus morphometrics at the SOTS site (Figure 2). For this analysis, we focussed on the two most widely Morphometric parameters of C. leptoporus coccoliths were used coccolith morphometric parameters measured in the estimated using a birefringence method that relies on the literature: mass and area (directly related with coccolith systematic relationship between the thickness of a calcite particle diameter). The inclusion of more coccolith morphometric and the interference colour produced under polarized light parameters, such as coccolith thickness, was found to largely (Beaufort, 2005; Beaufort et al., 2014; Fuertes et al., 2014). An influence the relationships between the morphometrics and the apical rhabdolith of the genus Acanthoica collected by a sediment environmental parameters and yielded unrealistic results. The CCA trap at the SOTS site was used for calibration. A calibration image of was not possible for the SAM site due to the lack of accompanying Frontiers in Marine Science 05 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 A Coccolith mass (SOTS) D Coccolith mass (SAM) 40 30 Spring Coccolith mass (pg) Coccolith mass (pg) 30 20 Summer Autumn 20 10 Winter 10 0 Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct B Coccolith area (SOTS) E 30 24 Coccolith area (SAM) 25 20 20 16 15 12 10 8 5 4 Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct C Coccolith thickness (SOTS) F Coccolith thickness (SAM) 0,6 0,55 0,55 0,5 0,5 0,45 0,45 0,4 0,4 0,35 0,3 0,35 Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct FIGURE 2 Average Calcidiscus leptoporus coccolith mass (A, D), area (B, E) and thickness (C, F) with standard deviations measured on the SOTS sediment trap deployed at 1000 m between August 2011 to July 2012 and on the SAM sediment trap deployed at 1500 m between November 2009 to November 2010. in situ environmental data sets. Details on the sensor measurements Area ~ Phosphate + Silicate + Temperature + PAR (with ~ symbol for each environmental variable can be found in Rigual-Herná ndez meaning that mass and area were modelled as a function of the four et al. (2020b). environmental variables). Please note that this model did not longer As many environmental parameters display substantial exhibit collinearity issues (2.30 for Phosphate, 1.79 for Silicate, 2.94 covariability throughout the annual cycle, we followed the next for Temperature and 1.30 for PAR). Therefore, this was the model procedure to identify the model with the minimum number of finally used to evaluate the influence of environmental variability on environmental variables that explained the maximum inertia (i.e. C. leptoporus coccolith morphometrics. the amount of variation explained by the environmental variables) being statistically significant (Hastie and Pregibon, 1992; Venables 2.4.2 Correlation analyses et al., 2002). Therefore, a full model containing all the environmental To investigate the possible relationship between C. leptoporus parameters, i.e. Photosynthetically Active Radiation (PAR), water coccolith morphometric parameters and the magnitude of coccolith temperature, salinity, phosphate, Total Oxidised Nitrogen (TNOx), fluxes (used as a proxy for coccolith growth, see section 4.3 for more silicate, total CO2 (TCO2), Wcalcite and pH, was calculated first. details) measured at the SOTS and SAM sediment traps, Spearman and Then, the correlations between all environmental parameters were Pearson correlations were calculated and plotted in Figure 3. Moreover, evaluated and confirmed that several variables exhibited high the relationship between coccolith mass, area, thickness and coccolith correlations with each other. The variance inflation factor analysis fluxes was also investigated through the calculation of two correlation (VIF) collinearity statistic was also estimated to evaluate the matrices that can be found in Supplementary Figure 1. covariability of the environmental parameters and yielded the following results: 48.49 for NOx, 20.47 for Phosphate, 90.21 for Silicate, 8.22 for Salinity, 10.77 for Temperature, 14.71 for PAR, 35.02 2.4.3 Comparison between sediment trap and for TCO2, 59.89 for Calcite and 34.98 for pH. As all these values were seafloor sediment coccolith assemblages greater than 5, which is the threshold above which the model is Coccolith assemblages from the uppermost layers of the seabed considered to have collinearity (Gareth et al., 2013), we used a sediments represent an integrated assemblage of the past hundreds stepwise variable selection based on the Akaike information to thousands of years, thereby providing a robust baseline of the criterion approach. After we obtained the model with the coccolithophore assemblages during the pre-industrial state of the minimum AIC: Mass + Area ~ Phosphate + Silicate + Temperature ocean. Similarly, annual sediment trap records can be used to + PAR + TCO2, we noticed that the variable TCO2 displayed a high provide a picture of the current state of modern coccolithophore VIF in the model (3.06 for Phosphate, 12.64 for Silicate, 4.68 for assemblages. However, to be able to compare the coccolith Temperature, 1.33 for PAR, 16.99 for TCO2). Based on the high VIF morphometric parameters measured in the surface sediment and its high correlation with Silicate concentration (r=0.95), TCO2) assemblages with those measured by the traps it is necessary to was removed it from the model. Thus, the final model was Mass + eliminate the effect of seasonality in the sediment trap records. This Frontiers in Marine Science 06 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 A B FIGURE 3 Regression plots showing Pearson and Spearman correlation coefficients between coccolith fluxes and coccolith mass (A) and area (B) for the SOTS and SAM sediment trap stations. can be accomplished by estimating a flux-weighted average of the variations throughout the Deglaciation and Holocene in the SAZ. coccolith morphometric parameters measured in the sediment The following suite of environmental variables relevant for traps. The flux-weighted values allow direct comparison with the coccolithophore physiology were used in the analysis: annual surface sediment assemblages, assuming that seasonality in insolation, alkenone-derived SSTs, iron concentration, coccolithophore fluxes have remained similar to modern atmospheric CO 2 concentration and surface-water pCO 2 . conditions over the time span represented by each Holocene-aged Alkenone-derived SSTs from core MD97-2120 were obtained seafloor sample (e.g. King and Howard, 2005). To evaluate the from Pahnke and Sachs (2006) (modern SSTs obtained from the possible morphometric variations of C. leptoporus coccoliths during summer months for the same location), annual solar insolation for the industrial era (prior ca. 1750; Pachauri et al., 2014), the the 47°S latitude from Laskar et al. (2004), iron concentration from sediment trap and Holocene-aged sediment assemblages were core SO136GC-38 after Durand et al. (2017) (iron concentration for compared using the weighted Kolmogorov-Smirnov test modern conditions was derived from the most recent sample of the (Monahan, 2011). The Kolmogorov-Smirnov tests consists of a sediment core dated in 1.7 kyr), atmospheric CO2 concentration two-sample distribution-free test designed to identify possible from the Antarctic ice core composite by Bereiter et al. (2015) deviations from the null hypothesis that the distributions of the (modern value from SOTS site after Shadwick et al. (2015)), and two compared data sets are identical. If the p value is smaller than surface-water pCO2 from core MD97-2106 (Dai et al., 2022) and 0.05 then the hypothesis is rejected, i.e., the distributions of a given Shadwick et al. (2015) (for the most recent sample). As the ages of morphometric parameter in the two populations compared are proxy measurements did not coincide with that of the Y16 samples, significantly different. the values of the environmental parameters for the analyzed samples of Y16 core were calculated through linear interpolation. 2.4.4 Environmental controls on coccolith A full model containing all the environmental parameters was morphometrics from the late deglaciation calculated: Mass + Area ~ SST + atm.CO2 + pCO2 + Fe_conc + to modern Insolation. Then, a correlation matrix between all parameters was To explore the possible response of C. leptoporus to estimated. As some of the variables exhibited high correlations, the environmental variability on millennial time scales, a CCA VIF of the environmental variables was calculated and yielded the following the same procedure explained in section 2.4.1 was following results: 2.18 for SST, 526.56 for atm.CO2, 429.64 for performed. For the coccolith morphometric data, an average pCO2, 18.72 for Fe_conc and 44.82 for insolation. As some of the between the annual-flux weighted coccolith morphometrics of the coefficients were greater than 5 (i.e. the threshold above which SAM and SOTS traps was used to represent the modern C. multicollinearity is considered high), the double stepwise method leptoporus populations in the subantarctic zone, while samples was applied to obtain models that decrease the AIC of the complete from core Y16 were used to represent coccolith morphometric initial model. The model with the minimum AIC was Mass + Area Frontiers in Marine Science 07 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 ~ atm.CO2 + pCO2 + Fe_conc + Insolation. However, as some of the two stations in Rigual-Herná ndez et al. (2020c) (6.39 ± 1.49 the variables still exhibited high VIFs (516.43 for atm.CO2, 425.56 μm). Two factors likely account for these different results. Firstly, in for pCO2, 17.65 for Fe_conc and 40.54 for insolation), the the present study, C. leptoporus coccoliths were measured for all parameter with the highest value (i.e. atm.CO2) was removed samples covering a full annual cycle for both sediment traps sites, from the model. Thus, the resulting model was Mass + Area ~ and then all morphometric data were flux weighted. In contrast, in pCO2 + Fe_conc + Insolation. As this model did not longer the previous study (Rigual-Herná ndez et al., 2020c) C. leptoporus exhibited collinearity for any of the environmental parameters coccoliths were measured from seven samples selected from two (with VIF values of 2.52 for pCO2, 2.49 for Fe_conc and 2.25 for different sediment trap depths at the SOTS site (1000 and 2000 m) insolation), it is the one we used in our interpretations. and one depth at the SAM site (1500 m), and values were not flux weighted. It is possible that the morphometrics of the C. leptoporus coccoliths collected in these selected samples were not 3 Results representative of the average annual C. leptoporus coccoliths sinking in the SAZ. Secondly, the visual coccolith selection 3.1 Calcidiscus leptoporus morphotypes procedure used in Rigual-Herná ndez et al. (2020c), after initial automated selection by the C-Calcita software, may have under- All samples were analyzed under the SEM for C. leptoporus represented small coccoliths, because these tended to have organic morphotype identification. A total of 98 and 146 coccoliths were debris attached more often than the larger ones. Given these identified to morphotype level for the SOTS and SAM sediment considerations, the morphometric data produced in the current trap records. In both records, 97% of the C. leptoporus coccoliths study should be taken as a baseline for the modern state of C. identified belong to the Intermediate form while 3% corresponded leptoporus coccolith assemblages, while the morphometric values of with the Small type. For the seafloor sediment samples a total of 193 the C. leptoporus coccoliths in Rigual-Herná ndez et al. (2020c) and 86 were analyzed for the Australian and New Zealand sectors of should be viewed as an upper estimate of the morphometric the Subantarctic Zone, respectively. The Intermediate form largely parameters of the coccoliths of C. leptoporus populations dwelling dominated the seafloor sediment assemblages, accounting for 98% in the SAZ. and 100% of the identified coccoliths, respectively. Lastly, analysis Calcidiscus leptoporus coccolith fluxes were significantly correlated of 255 coccoliths from the Y16 sediment core indicates that with coccolith mass (Spearman rho = 0.40, p value < 0.01; Figure 3A) dominance of the intermediate form also held true during the last and area (Spearman rho = 0.39, p value < 0.01; Figure 3B) but ~14 Kyrs (100% of all the coccoliths analyzed). Morphotype uncorrelated with coccolith thickness (Spearman rho = 0.17, identification was not possible for samples older than ~14 Kyrs p value > 0.01; data not plotted in Figure 3). due to the stark decrease in C. leptoporus abundance in the deepest The results of the CCA in the SOTS sediment trap record section of the sediment core. suggest that photosynthetically active radiation (PAR), water temperature in the surface layer and silicate concentration are the main predictors of C. leptoporus coccolith morphometrics 3.2 Seasonal variations in C. leptoporus (Figure 4). The CCA results also indicate a secondary morphometrics at the SOTS and SAM sites contribution of phosphate concentration on coccolith morphometrics. The first axis accounted for most of the inertia A total of 481 and 628 C. leptoporus coccoliths were analyzed (99.6%) and was positively related with phosphate and temperature morphometrically for the SOTS and SAM sediment-trap samples, and negatively with PAR and silicate. Both coccolith mass and area respectively. The seasonal morphological variations of C. leptoporus are related with high silicate concentration and PAR and with low were roughly similar at the SOTS and SAM sites, although the phosphate concentration and temperature (Figure 4). seasonal amplitude of the changes were more pronounced at the SOTS site (Figure 2). Coccolith mass, area and thickness were highest in winter and/or early spring at both SOTS and SAM sites. 3.3 Sediment trap and seafloor sediment The main morphometric parameters exhibited a decrease towards comparison autumn at SOTS, while the same parameters remained relatively stable during summer and autumn at the SAM site. The coccolith mass, area and thickness for the sediment traps and The annual flux-weighted average of coccolith mass, area and Holocene sediments are summarized in Figure 5. A total of 370 C. thickness at SOTS were 16.7 ± 7.4 pg (average ± standard leptoporus coccoliths were analyzed from the seafloor sediment deviation), 14.3 ± 5.5 μm2 and 0.42 ± 0.05 μm, respectively. These samples. Since part of the seabed sediment samples retrieved in the values were similar, although slightly higher, than those subtropical zone displayed important differences with those of the documented at the SAM site with 14.8 ± 6.8 pg, 12.1 ± 4.9 μm2 subantarctic zone (Figure 5), these samples were excluded for the and 0.45 ± 0.04 μm, respectively. It is important to note that the comparison with the SOTS and SAM sediment traps (both deployed average size of the C. leptoporus coccoliths measured in the traps in in the SAZ). Moreover, correlation analysis revealed that a significant the current study (diameter of 4.41 ± 0.83 μm for the SOTS site and correlation between of morphometric parameters and depth exists 4.11 ± 0.82 μm for the SAM site; average ± standard deviation) are (Supplementary Figure 2), implying that larger and heavier C. lower than that previously reported for this species using data from leptoporus coccoliths are found in the deepest seabed samples most Frontiers in Marine Science 08 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 3.4 Environmental controls on coccolith Cup 11 (Jan) morphometrics throughout the late Mass PAR Cup 4 (Oct) Cup 8 deglaciation and the Holocene in the Cup 5 (Oct) (Dec) Cup 13 (Feb) Subantarctic Zone Cup 9 (Dec) Cup 15 Cup 21 Cup 7 Phosphate (Mar) (Jul) (Nov) Cup 17 The abundance of C. leptoporus coccoliths was very low before (Apr) Cup 19 (Jun) Temperature the Late Deglaciation (i.e. before ~14 kyr) (Supplementary 0 Cup 3 (Sep) Cup 20 (Jun) Figure 4), therefore the morphometric characterization of C. Cup 16 (Apr) Cup 14 leptoporus coccoliths was only possible for the most recent nine Cup 6 (Mar) (Nov) Cup 18 samples of Y16 sediment core. Coccolith area displayed a slight Cup 12 (May) Cup 10 decrease from the late Deglaciation towards the early Holocene and RDA2 (Jan) (Feb) Silicate Cup 1 (Aug) then, it exhibited a gradual decrease towards the present Area (Figure 7B). Coccoliths mass displayed relatively stable values until the mid-Holocene, when an increase in coccolith mass is −1 observed between 7 and 6 kyr, then the values decreased steadily SOTS site towards the present (Figure 7B). Coccolith thickness decreased Cup Morphological from the late Deglaciation until ca. 8 kyr and then increased Environmental gradually towards the present with a peak between 5.6 to 4.6 kyr. The CCA results (Figure 8) indicate that changes in insolation, Cup 2 (Sep) pCO2 variations and changes in iron concentration from the deglaciation until present (Figure 7A) accounted for 100% of the −0.5 0.0 0.5 RDA1 total inertia in the dataset. Both insolation and pCO2 were the main FIGURE 4 predictors of coccolith area and mass changes, while iron input Canonical Correspondence Analysis (CCA) after model selection showing the relation between the temporal changes in coccolith played a secondary role (Figure 8). The first axis explained most of morphometrics of C. leptoporus from different cups (i.e. each the inertia (92%) and was mostly positively related with insolation sediment trap sample) throughout the year (2011-2012) at 1000 m and negatively with pCO2. The second axis explained 8% of the depth at SOTS and environmental factors measured in situ at SOTS (blue arrows) during the same interval after variable selection (see variability and was mostly related with iron concentration Rigual-Herná ndez et al., 2020b for further details) with Akaike (Figure 8). High values of coccolith area and mass were positively information criterion (AIC). Arrows indicate the direction and relative related with high insolation and negatively with dissolved CO2, importance (arrow lengths) of each environmental parameter on each axis. PAR, photosynthetically active radiation. while iron input, although secondary, was positively related with coccolith mass and negatively with coccolith area. likely due to selective dissolution of the weakest (lightest) forms. When samples retrieved beneath the calcite saturation horizon 4 Discussion (minimum depth at 3200 and 2800 in the Australian and New Zealand sector of the SAZ, respectively; Bostock et al., 2011) are 4.1 Seasonal environmental controls on C. excluded, the correlations between morphometric parameters and leptoporus coccolith morphometrics depth become non-significant, suggesting that the coccoliths from samples above the calcite saturation horizon were unbiased by calcite The high-resolution monitoring of water column properties at dissolution (Supplementary Figure 2). Consequently, only seabed the SOTS observatory (Trull et al., 2010; Eriksen et al., 2018; samples collected above the calcite saturation horizon were Shadwick et al., 2022) allows us to assess which environmental considered for the comparison with the modern (sediment trap) parameters represent the main controls on C. leptoporus coccolith C. leptoporus assemblages. morphometrics on seasonal timescales. As the growth rate of C. The results of the Kolmogorov-Smirnov test indicate that leptoporus in culture has been estimated as ca. 0.5 divisions d-1 modern subantarctic C. leptoporus coccoliths collected by the (Renaud et al., 2002; Langer et al., 2012) and the sediment trap traps are 12% lighter and 24% smaller (in terms of coccolith mass sampling resolution varied between 16 and 8 days, the delay of C. and area, respectively) than those preserved in the Holocene-aged leptoporus response to environmental variations in the surface layer sea-floor samples (Figures 6A, B). Coccolith thickness, however, is considered negligible in our study. Our analysis indicates that shows an opposite trend with coccoliths intercepted by the traps many environmental parameters known to influence coccolith exhibiting 16% thicker coccoliths (i.e. calcification per surface area morphometrics (e.g. temperature, seawater carbonate chemistry, of the coccolith) than those preserved in the sedimentary record nutrients, salinity, among others) exhibit considerable covariation (Figure 6C). Therefore, the increase in coccolith thickness is not throughout the annual cycle, thereby hampering the identification sufficient to compensate the coccolith mass loss between the pre- of the main predictors. The use of the Akaike criterion (see industrial and modern eras. methods) allowed us to reduce the number of possible predictors Frontiers in Marine Science 09 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 A B C FIGURE 5 Box and whisker plots showing C. leptoporus coccolith mass (A), area (B) and thickness (C) of the SOTS and SAM sediment trap time series (non- flux-weighted values; red) and Holocene seafloor sediment samples from the subantarctic (dark blue) and subtropical sediments (yellow) samples in the Indian and Pacific waters south of Australia and New Zealand. The median value is displayed as a solid bar within the box, the upper and lower edges of the boxes represent the 75th and 25th percentiles while the whiskers indicate ± 1.5-times the interquartile range (IQR). determining coccolith morphometrics to three: nutrient the weak relationship of C. leptoporus coccolith morphometrics concentrations, PAR and temperature (Figure 4). with phosphate and lack of relationship with nitrate (Figure 4) The CCA results suggests that macronutrients have contrasting suggest that these two macronutrients play a negligible to minor roles in the controls of coccolith morphometrics. On the one hand, role in the controls of C. leptoporus coccolith dimensions. These A B C Sea−bed sediments 8 Sediment traps 0.06 0.075 6 Density Density Density 0.04 0.050 4 0.02 0.025 2 p value = 0.001719 p value = 0 p value = 0 0.00 0.000 0 10 20 30 40 50 10 20 30 40 0.2 0.3 0.4 0.5 0.6 Coccolith mass (pg) Coccolith area (µm²) Thickness (µm) FIGURE 6 Weighted densities of the distribution of Calcidiscus leptoporus coccolith mass (A), major axis (B) and thickness (C) between the flux-weighted data of the SOTS and SAM sediment traps together (representative of modern populations) and subantarctic seafloor sediment samples above 3200 m grouped together (representative of mid- to late Holocene assemblages). Frontiers in Marine Science 10 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 A B FIGURE 7 (A) Evolution of key environmental parameters throughout the Late Deglaciation and Holocene in the SAZ south of New Zealand and Australia (see section 2.4.1 for more details). (B) C. leptoporus coccolith mass, area and thickness from sediment core Y16. 35−36 cm Iron (8.4 kyr) 0.5 Area Sediment traps (modern) 25−26 cm (7.0 kyr) 29−30 cm (7.6 kyr) 40−41 cm (13.4 kyr) pCO2 20−21 cm (6.3 kyr) Insolation 0.0 05−06 cm (4.1 kyr) RDA2 03−04 cm (3.8 kyr) −0.5 Sample Morphological 15−16 cm Environmental (5.6 kyr) 09−10 cm (4.7 cm) Mass −0.5 0.0 0.5 RDA1 FIGURE 8 Canonical Correspondence Analysis (CCA) after model selection showing the correlation between C. leptoporus coccolith mass and area and key environmental factors for the last 14 kyrs in the SAZ. Blue arrows indicate the directions and relative importance (arrow lengths) of each environmental factor on each axis. Iron – iron concentration in the seafloor; pCO2 – surface layer CO2 partial pressure; insolation – annual insolation at 47°S latitude (see section 2.4.4 for more details). Frontiers in Marine Science 11 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 results are not surprising since laboratory evidence indicates that Water temperature is identified as an important environmental phosphate and nitrate concentrations do not influence coccolith predictor in coccolith morphometrics by our CAA analysis. This morphology in C. leptoporus (Langer et al., 2012) and also because possibility is likely since Sea Surface Temperature (SST) has been phosphate and nitrate levels never reach limiting concentrations in suggested to play an important role in the biogeographical and the subantarctic surface waters south of Tasmania (Supplementary seasonal distribution of C. leptoporus morphotypes, yet conflicting Figure 3). On the other hand, the CCA analysis reveals that seasonal results have been reported (Knappertsbusch et al., 1997; Renaud changes in silicate concentration are positively related with both et al., 2002; Sá ez et al., 2003). The overwhelming dominance of the coccolith mass and area (Figure 4). Durak et al. (2016) intermediate morphotype (i.e. C. leptoporus ssp. leptoporus) in the demonstrated that silicate is an essential nutrient for coccolith SAZ traps and sediments is consistent with previous onboard formation for a group of coccolithophores species, that includes observations in the study region (Malinverno et al., 2015; C. leptoporus, characterized by the presence of diatom-like silicon Saavedra-Pellitero and Baumann, 2015) and suggest a successful transporters (SITs). The subantarctic waters south of Tasmania are adaptation of this morphotype to the cold and nitrate and characterized by low silicate concentrations year-round (<4 to 5 phosphate-rich waters of the subantarctic Southern Ocean. mmol l-1; Rintoul and Trull, 2001; Fripiat et al., 2011) and typically Laboratory culture experiments by Quinn et al. (2003) reach limiting concentrations (thought to be ~1 μM (Paasche, 1973) demonstrated that cold water temperatures can induce a diameter or possibly considerably higher in Southern Ocean species (Frank reduction of up to 0.5 μm on monoclonal cultures of the et al., 2000)) after a diatom bloom takes place at the onset of the Intermediate morphotype. As both minimum and maximum productive period (Dugdale et al., 1995; Lannuzel et al., 2011; annual coccolith area values are observed in winter (Figure 2), the Rigual-Herná ndez et al., 2015; Eriksen et al., 2018). In situ relationship of water temperature with coccolith area is not clear at nutrient measurements at SOTS during our sampling interval, SOTS. However, it should be noted that the seasonal variability in indicate that silicate concentration dropped below 2 mM in early water temperature at SOTS is tiny (~3°C) and substantially smaller December (Supplementary Figure 3) and remained below this than that used in Quinn et al.’s experiment (~12°C). Therefore, it is threshold until July. It is during this interval of low silicate likely that the opposite relationship with temperature in our study concentration when the lightest and smallest coccoliths are and larger variability on coccolith diameter (~1.2 μm) is not driven observed (Figure 2). This correlation might point to an effect of by temperature change. low Si concentration on coccolith mass in C. leptoporus. However, Moreover, it is worth noting that TCO2 changes could have also experimental studies analyzing the effect of low Si concentration on influenced coccolith morphometrics on a seasonal time scale even coccolith formation are rare and focus on coccolith malformations though this parameter was excluded from the final CCA model rather than size. The few data available suggest that Si owing to its high correlation with silicate concentration (r=0.95; n = concentrations need to be as low as 0.2 μM to affect coccolith 21). Changes in seawater carbonate chemistry are widely known to morphology (Langer et al., 2021). Another possibility is that the influence coccolithophore calcification (e.g. Langer et al., 2006; silicate in our correlations is acting as a proxy for dissolved iron, Beaufort et al., 2011; Meier et al., 2014). Interestingly, and similar which is also seasonally depleted in subantarctic waters (Sedwick to previous results of E. huxleyi coccoliths in our study sites (Rigual- et al., 1999; Frank et al., 2000; Boyd et al., 2001). The results Herná ndez et al., 2020b), larger and heavier C. leptoporus coccoliths presented here clearly warrant further studies analyzing the effect of were recorded at times of peak annual TCO2 (Supplementary different Si concentrations on coccolith size or mass. Figure 3). The identified positive relation of PAR with coccolith mass and Based on all the above, it can be concluded that variations in area on seasonal scale appears logical as calcification is a light silicate concentration and light are the most likely seasonal controls dependant-process (Paasche, 2002; Zondervan, 2007, and on coccolith morphometrics (mass and area) in the modern references therein) and because light is a limiting factor for subantarctic zone although the influence of other environmental phytoplankton growth in the SAZ as a result of deep summer parameters, such as TCO2, cannot be ruled out. mixed layers (70-100 m; Rintoul and Trull, 2001) and cloudiness (Bishop and Rossow, 1991). The few available laboratory data on the effect of temperature on C. leptoporus coccoliths morphometrics 4.2 Evolution of Calcidiscus leptoporus indicates that some strains of C. leptoporus might be sensitive to low coccolith morphometrics from the late light levels reducing their diameter up to ~0.7 μm (Quinn et al., deglaciation to the modern 2003). However, it should be noted that Quinn et al. (2003) documented different responses across the strains analyzed, with Comparison of the C. leptoporus coccoliths intercepted by the some of them being unaffected by low-light stress. Moreover, traps with those from the seafloor sediments indicate that the several studies have documented that calcification in the model modern subantarctic C. leptoporus populations produce about coccolithophore species E. huxleyi decreases at low irradiances 15% lighter and 25% smaller coccoliths (in terms of coccolith (Paasche, 1999; Zondervan et al., 2002) most likely as a result of a mass and area, respectively) than those of the preindustrial lower amount of calcite deposited in each coccolith and the lower Holocene. Coccolith thickness is the single morphometric number of coccoliths produced per cell (Paasche, 1999). Based on parameter that displays higher values in the sediment trap (i.e. all the above, light availability could account for at least part of the modern) assemblages, being about 15% thicker than the seasonal coccolith area variations documented in the traps. assemblages preserved in the seafloor sediments. Three Frontiers in Marine Science 12 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 mechanisms, operating alone or in combination, could be resulting reduction in seawater carbonate ion concentration and responsible for the observed morphometric variations of C. ocean pH (Feely et al., 2004; Doney et al., 2009) are known to leptoporus coccoliths throughout the late Holocene in both sectors impact coccolithophore growth rate, photosynthesis, calcification of the SAZ: calcite dissolution, a dominance shift in the assemblage and morphology (Riebesell et al., 2000; Langer et al., 2009; Langer and/or a physiological response of the dominant morphotype to an and Bode, 2011; Lefebvre et al., 2012; Bach et al., 2013; Schlüter environmental change. Disentangling the potential effects of each of et al., 2014b; Müller et al., 2015). However, CO2 perturbation these factors is an essential prerequisite for a meaningful experiments on C. leptoporus by Langer et al. (2006) showed that interpretation of the observed morphometric variations cellular inorganic carbon content displays a nonlinear relationship throughout the Holocene until present. with CO2 concentrations: PIC cellular content increased steadily 1) Calcite dissolution in the water column and/or sea floor from ~100 μatm to ~350 μatm, and then decreased at CO2 values could induce changes in the coccolith morphology of C. leptoporus >470 μatm. Based on these laboratory observations, an increase in assemblages preserved in the sedimentary record. For example, coccolith size (i.e. area) and mass, which are directly related to PIC partial dissolution of the coccoliths could reduce the mean mass and cellular content (Langer et al., 2006), should be expected between diameter of the original C. leptoporus assemblages (Jin et al., 2019; preindustrial Holocene CO2 levels of roughly 240 ppmv to present Rigual-Herná ndez et al., 2020a), whereas selective dissolution of the ~400 ppmv instead of diminishing as shown by our results. smallest and/or lightest coccoliths could cause the opposite effect. However, the response of natural C. leptoporus populations may As mentioned above, previous research at SOTS indicate that be different than that observed in monoclonal cultures due to the coccoliths (Rigual-Herná n dez et al., 2020b) and planktonic larger genetic variability within natural populations, longer foraminifera shells (Moy et al., 2009) experience negligible timescales and the interplay between multiple stressors which changes during their transit between meso- and bathypelagic may offset or amplify the effects of single drivers (Reusch and depths, thereby ruling out the possibility of dissolution during Boyd, 2013; Collins et al., 2014; Feng et al., 2017). Lastly, it is their transit throughout the water column. Secondly, all the important to note that the CCA results are markedly influenced by seafloor sediment samples used in our analysis were retrieved the most recent sample (Figure 8), that exhibit substantially higher above the calcite saturation horizon. Lastly, SEM observations of pCO2 values (between 80-130 matm) than the pre-industrial all sediment samples shows a good preservation for most of the C. samples. Therefore, our data provides no clear evidence that leptoporus coccoliths indicating the calcite dissolution did not have ocean carbonate chemistry alone has an appreciable contributory an important impact on the seafloor assemblages. Based on these role to overall variation in coccolith size and area throughout the lines of evidence, it can be concluded that calcite dissolution played last 14 kyrs but may represent an important control on C. leptoporus a negligible role in the observed coccolith morphometric variations physiology during the industrial era. between seabed and sediment trap data sets. Moreover, the CCA results also indicate that changes in 2) A shift in the abundance of the dominant morphotype could insolation at 47°S latitude are positively related with variations of lead to a change in the average of a given coccolith morphometric coccolith area and mass (Figure 8). As light limitation is known to parameter, e.g. an increase in the relative abundance of larger represent an important control on phytoplankton productivity in the morphotypes would result in an increase in the average area of SAZ (Deppeler and Davidson, 2017), it is likely that changes in the population. However, observation of all samples under the SEM insolation alone and/or in combination with changes of mixed layer indicates that the C. leptoporus assemblages preserved in the trap depths (largely determined by insolation changes) could have and sedimentary records are nearly completely composed of the induced a physiological response on C. leptoporus. However, it sub-species leptoporus (intermediate morphotype). This could be argued that the insolation variation from the late observation is in agreement with previous research of modern Holocene (~4 Wm-2) until present is too small to be directly plankton assemblages (Malinverno et al., 2015; Saavedra-Pellitero responsible for the rapid and pronounced reduction in coccolith and Baumann, 2015) and renders the possibility of an “ecologically” mass and area observed between preindustrial and industrial driven change (Bach et al., 2012) in coccolith morphology unlikely. datasets (Figure 6). Lastly, the CCA also suggests a secondary 3) A change in one or several environmental parameters could contribution of iron concentration in the controls of coccolith induce a response of the dominant C. leptoporus strain or morphometrics (Figure 8). Iron transported into the SAZ as morphotype, thereby driving a “physiologically” driven change in mineral dust is known to play an important role in the average of a given coccolith morphometric parameter. Since coccolithophore physiology (e.g. Balch, 2018, and references calcite dissolution and a shift in the dominant C. leptoporus therein). However, its opposite effect on coccolith mass and morphotype can be disregarded as possible factors of such a limited variability explained by the second CCA axis (mostly change, next the possible influence of key environmental factors related with iron concentration) suggests that the influence of iron that could have induced a “physiologically” driven change in the could be an artifact result of the limited sample size of the study. subantarctic C. leptoporus assemblages is discussed. Therefore, our results suggest that insolation could represent a Our CCA analysis (Figure 8) identifies variations in surface- possible modulating factor —direct or indirect— of the physiology water CO2 partial pressure (pCO2) from the Late Deglaciation to the of subantarctic C. leptoporus populations from the Late Deglaciation modern to be significantly and negatively correlated with coccolith and most of the Holocene but its role during the Late Holocene to area (Figure 8). The increasing atmospheric CO2 levels and the modern is less clear. Frontiers in Marine Science 13 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 It should be acknowledged that the effect of other factors aside could imply a more efficient biological pump in the SAZ, as the from the ones identified in our analysis could have also exerted and efficient repackaging of carbon for deep transport by the important role in the observed coccolith morphometric variability. zooplankton community has been hypothesized to account for This is due to three reasons. Firstly, it is possible that important the greater relative POC content of the particles sinking in the environmental parameters, not considered in our study owing to SAZ compared to other Southern Ocean zonal systems further the lack of data availability (e.g. nutrient concentrations), may have south (Trull et al., 2001; Ebersbach et al., 2011). On the other hand, played an important role in coccolith morphometrics on millennial an increase of microzooplankton grazing rates may facilitate time scales. Secondly, it is likely that the observed variations in organic carbon remineralization in the upper water column, coccolith area and mass are the result of an additive, antagonistic or thereby diminishing the carbon export associated with synergistic effect of multiple drivers (e.g. De Bodt et al., 2010; Feng coccolithophores. Future mesocosm studies examining the effects et al., 2018; Beusen et al., 2022), not captured by our analysis. of coccolith size on ballasting efficiency of faecal pellets and algal Thirdly, the number of samples available for the analysis is low, and aggregation will be needed to determine the impact of shifts in therefore existing relationships with environmental parameters dominant coccolithophore size classes on the marine carbon cycle. might not be identified. For example, changes in SSTs show a medium correlation with coccolith area (0.4). It is possible that a larger sample size might have yielded additional information not 4.3 Possible application of coccolith drawn from the present study. Therefore, although ruled out by the morphometry as palaeo-proxy for Akaike criterion water temperature variations (i.e. SST) could have growth rate contributed to the observed coccolith morphometric changes since the late Deglaciation, as it represents a critical control on The significant correlation between fluxes of C. leptoporus coccolithophore growth rates (Buitenhuis et al., 2008) and species coccoliths and their mass and area in our time-series sediment distribution in the Southern Ocean (Findlay and Giraudeau, 2000; trap records (Spearman r = 0.40 and r = 0.39, respectively; N = 46, Charalampopoulou et al., 2016; Patil et al., 2017; Rigual-Herná ndez p value < 0.01; Figure 3) indicates that heavier and larger coccoliths et al., 2020c; among others). Moreover, long-term incubations of are produced at times of enhanced coccolith fluxes. This subantarctic polyclonal E. huxleyi populations suggest that this observation is consistent with results from the subpolar North species will decrease its cellular volume and PIC production by Atlantic station NABE-48, located in SubArctic waters (48°N; the year 2100 due to the synergistic or additive effect of warming Broerse et al., 2000; Renaud et al., 2002), where C. leptoporus and pH decrease (Armstrong and Law, 2023 (in press)). Given that coccolith fluxes exhibit a positive significant correlation with coccolith volume has been documented to be strongly correlated to coccolith diameter (Spearman correlation coefficient (r) of 0.62, cell volume in E. huxleyi (Müller et al., 2021), the decrease in C. N = 24, p value < 0.01). In the following we will compare this field- leptoporus coccolith mass (which equals to coccolith volume data correlation with experimentally derived correlations. multiplied by the density of calcite; Young and Ziveri, 2000) Experimental data suggest a positive correlation between during the Late Holocene and/or industrial era could be calcification related physiological rates, i.e. calcite production, interpreted as a reduction in cell volume. Interestingly, cell size coccolith production, and growth rate (Langer et al., 2006; Langer reduction is a common response of coccolithophores and other and Bode, 2011; Diner et al., 2015). phytoplankton to combined warming and decreased pH (Atkinson There are, moreover, studies that seem to support a relationship et al., 2003; Finkel et al., 2009; De Bodt et al., 2010; Schlüter et al., between coccolith mass, area and production/growth rate. In 2014a). As subantarctic waters are experiencing rapid warming carbonate chemistry manipulation experiments the percentage of (~0.3°C increase per decade since 1992; Auger et al., 2021) and coccoliths that are unusually small (either due to incompleteness or acidification (McNeil and Matear, 2008), our results suggest that the malformations) is negatively correlated with growth rate/calcite combined effect of changes in these parameters since the onset of production (Langer et al., 2006; Langer and Bode, 2011). Whether the industrial era could potentially account for a substantial part, if these experimental correlations are applicable to field data is not not all, of the differences between pre-industrial and industrial clear because firstly, neither coccolith size nor coccolith production data sets. was measured. Secondly, all physiological changes were induced by Extrapolation of our results suggests that if the significant sizable changes in carbonate chemistry which might not be reduction in C. leptoporus coccolith size (i.e. area) and mass representative of the field situation discussed here. Thirdly, it during the Holocene to the modern will continue in the future, it r e m a i n s u n k n o w n w h e t h e r r a t e c h a n g e s i n d u c e d by will result in a reduction of the ballasting capacity of this species and environmental parameters other than carbonate chemistry would consequently the efficiency of the organic carbon export in the SAZ. have the same effect. Despite these caveats we propose to apply the Moreover, as primary production consumed by microzooplankton experimental positive correlation between coccolith size and growth is known to be negatively correlated with cell size (Chen and Liu, rate to field data. We therefore tentatively propose coccolith size 2010), the reduction in cell size of C. leptoporus will most likely (i.e. area) as proxy for growth rate/coccolith production. Taken facilitate microzooplankton grazing of this species. An increase in together with the positive correlation between coccolith size and grazing could, however, exert opposing effects on the efficiency of flux (see above) we can furthermore infer a positive correlation the biological pump. On the one hand, enhanced micrograzing between growth rate and flux. Note that this inference is based on Frontiers in Marine Science 14 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 the correlations observed in field and experimental data only. It change. Comparison of high-resolution sensor data with full does not describe the complexities that co-determine flux, e.g. annual collection of coccolith fluxes by a sediment trap, suggests grazing, dissolution and so on. that variations in silicate concentration and light availability If the relationship between C. leptoporus area with growth rate represent the most likely controls on coccolith size on seasonal identified on the seasonal scale (Figure 3) would hold true for the time scales. Moreover, comparison of modern coccolith fossil record, the gradual decrease in coccolith area throughout the measurements obtained from sediment-trap annual records with Holocene in sediment core Y16 (Figure 7B) could be taken as those from a compilation of subantarctic seafloor assemblages, indicative of a reduction of C. leptoporus growth rates. However, it revealed that modern C. leptoporus populations are about 15% is important to note that, C. leptoporus coccolith accumulation rate lighter and 25% smaller than their pre-industrial counterparts. The in sediment core Y16 (Supplementary Figure 4) is not significantly limited number of proxies available to reconstruct the complexity of correlated (p-value > 0.05) with coccolith area or mass (Spearman r the environmental conditions from the Late Deglaciation to the correlation of -0.05 and -0.25, respectively; N = 9), implying that modern, together with the reduced number of samples with enough different factors other than growth rate may be equally or more C. leptoporus coccoliths for morphometric characterization, influential in regulating coccolith morphogenesis on millennial time prevented reaching conclusive evidence on the main controls of scales in the SAZ. However, it is also possible that the low coccolith morphometrics on millennial time scales. However, the sedimentation rates over the Campbell Plateau (1.5 to 3 cm/kyr interpretation of our results in light of the literature suggests that in interglacials and 0.5 to 2.5 cm/kyr in glacials; Carter et al., 2002) the cumulative effect of multiple parameters — most likely combined with post-depositional effects, such as downcore temperature, ocean acidification and insolation — is controlling variations in nannofossil concentrations by opal dilution C. leptoporus coccolith size variations in the subantarctic Southern (Bradtmiller et al., 2009), and/or sediment focusing or winnowing Ocean from the Late Deglaciation to the modern. Moreover, our associated with the strong currents of the ACC (Panitz et al., 2015) data suggests that the influence of each of these parameters has could have altered the C. leptoporus production signal preserved in varied through time, with warming and ocean acidification most the sedimentary record (i.e. the coccolith accumulation rate). likely playing a preponderant role during the industrial era. Lastly, Moreover, temperature is a critical factor controlling the our analysis also indicates that changes in iron supply into SAZ did latitudinal distribution of coccolithophores in the Southern Ocean not represent an important factor controlling coccolith (Findlay and Giraudeau, 2000; Charalampopoulou et al., 2016; morphometrics throughout the last 14 kyr. among others; Patil et al., 2017; Rigual-Herná ndez et al., 2020c), Extrapolation of our results suggests that a future decrease in and therefore it also possible that the lower SSTs during the coccolith size of C. leptoporus will have a negative impact on the deglaciation (Figure 7A and Pahnke and Sachs, 2006)) played a efficiency of the biological pump in the SAZ through a reduction of major role in C. leptoporus productivity during the deglaciation c a r b o n a t e ba l l a s t in g . H o w e v e r , t h e c o n s e q u e n c e s i n thereby overprinting the effect of other environmental parameters microzooplankton grazing and related changes in the biological on growth rate. Therefore, the possible application of coccolith area pump are less clear. Lastly, the significant correlation between C. as a palaeo-proxy for C. leptoporus growth rate remains at the first leptoporus fluxes and coccolith area in sediment trap records from phase of the typical “scientific maturation process” for a new proxy subpolar regions combined with an experimental, positive as proposed by Elderfield (2002), in which limited evidence suggests correlation between coccolith size and growth rate tentatively its great potential (Chase et al., 2018). Future culture experiments suggest that coccolith area could be used as a proxy for C. tailored to determine the physiological and environmental controls leptoporus growth rate. Future laboratory and field experiments (such as light, temperature and nutrient availability) on C. will be needed to assess this proxy. leptoporus coccolith production will be needed to validate this proxy and to identify its limitations. Data availability statement 5 Conclusions The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. The combination of multiple sampling efforts together with detailed taxonomic and morphometric analyses, permitted to reconstruct coccolith morphometry of the keystone Author contributions coccolithophore species Calcidiscus leptoporus from seasonal to millennial time scales in the Subantarctic Southern Ocean. Our TT, SN and RE planned and performed the sediment trap field results demonstrate that subantarctic C. leptoporus populations experiment. AR-H, JF, TT, SN, HB and FS conceived the project have been dominated by a single morphotype — type and facilitated acquisition of samples and the financial support for Intermediate — since, at least, the last 14 kyrs in the Subantarctic the project leading to this publication. JS-S performed numerical Southern Ocean. Therefore, the identified coccolith morphometric analyses. AR-H, AB, LM, TB, FR-M, AH-M and MZ-V performed variations at different timescales in the SAZ must have been driven image and SEM analyses. GL and AR-H performed identification of by a physiological response of C. leptoporus to environmental C. leptoporus morphotypes. The paper was written by AR-H and Frontiers in Marine Science 15 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 GL with input from all authors. All authors contributed to the Publisher’s note article and approved the submitted version. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated Funding organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or This project has received funding from the European Union’s claim that may be made by its manufacturer, is not guaranteed or Horizon 2020 research and innovation programme under the Marie endorsed by the publisher. Skłodowska-Curie grant agreement number 748690 – SONAR-CO2 (AR-H and JF) and by project BASELINE (PID2021-126495NBC33) funded by the MCIN/AEI/10.13039/501100011033 (AR-H). GL acknowledges funding from the Spanish Ministry of Universities Supplementary material through a Maria Zambrano grant. This work contributes to ICTA- UAB María de Maeztu” Programme for Units of Excellence of the The Supplementary Material for this article can be found online at: Spanish Ministry of Science and Innovation (CEX2019-000940-M). https://www.frontiersin.org/articles/10.3389/fmars.2023.1159884/ full#supplementary-material Acknowledgments SUPPLEMENTARY FIGURE 1 Pearson (A) and Spearman (B) rank correlation matrices for C. leptoporus coccolith morphometrics (mass, area and thickness) and C. leptoporus We are grateful to Peter Jansen for assistance in environmental coccolith fluxes at the SOTS and SAM sites. Both tests yield similar results. Non-significant correlations (p values > 0.05; N = 46 samples) are data provision from SOTS observatory. Authors wish to thank José crossed out. Ignacio Martin and Miguel Á ngel Fuertes for their assistance in nannoplankton sample preparation and calibration of C-Calcita SUPPLEMENTARY FIGURE 2 software, respectively. The authors acknowledge the assistance of Correlation between coccolith mass and depth including (A) and excluding Vito Clericò, Enrique Diez (USAL-NANOLAB) and Marta Ortiz samples below the calcite saturation horizon (~3200 m) (B). (USAL-NUCLEUS) in SEM analyses. We sincerely thank Maria SUPPLEMENTARY FIGURE 3 Vernet and Nina Bednarsek for the time dedicated to review the Environmental parameters measured at the SOTS site: water temperature, manuscript and for their valuable suggestions, which helped us to photosynthetically available radiation (PAR), salinity, phosphate improve the quality of the manuscript. concentration, total oxidised nitrogen concentration (TOxN: nitrate plus nitrite), and silicate concentration and carbonate system parameters (pH, Total CO2 and calcite saturation state). Conflict of interest SUPPLEMENTARY FIGURE 4 Downcore profiles of total nannofossil accumulation rate (NAR) (A), relative abundance of C. leptoporus (B), and C. leptoporus NAR (C) in sediment core The authors declare that the research was conducted in the Y16. Total NAR is considered to be an estimate of coccolithophore absence of any commercial or financial relationships that could be paleoproductivity (e.g. Baumann et al., 2004; Gonzá lez-Lanchas et al., construed as a potential conflict of interest. 2020) and was calculated after Flores and Sierro (1997). References Armstrong, E., and Law, C. S. (2023). Resilience of emiliania huxleyi to future photosynthesis and calcification in the coccolithophore emiliania huxleyi. New changes in subantarctic waters. PloS One. doi: 10.1371/journal.pone.0284415 Phytol. 199, 121–134. doi: 10.1111/nph.12225 Armstrong, R. A., Lee, C., Hedges, J. I., Honjo, S., and Wakeham, S. G. (2002). A new Bairbakhish, A. N., Bollmann, J., Sprengel, C., and Thierstein, H. R. (1999). mechanistic model for organic carbon fluxes in the ocean based on the quantitative Disintegration of aggregates and coccospheres in sediment trap samples. Mar. association of POC with ballast minerals. Deep-Sea Res. II 49, 219–236. doi: 10.1016/ Micropaleontology 37, 219–223. doi: 10.1016/S0377-8398(99)00019-5 S0967-0645(01)00101-1 Balch, W. M. (2018). The ecology, biogeochemistry, and optical properties of Armstrong, R. A., Peterson, M. L., Lee, C., and Wakeham, S. G. (2009). Settling coccolithophores. Annu. Rev. Mar. Sci. 10, 71–98. doi: 10.1146/annurev-marine- velocity spectra and the ballast ratio hypothesis. Deep-Sea Res. II 56 (18), 1470–8. doi: 121916-063319 10.1016/j.dsr2.2008.11.032 Balch, W. M., Drapeau, D. T., Bowler, B. C., Lyczskowski, E., Booth, E. S., and Alley, Atkinson, D., Ciotti, B. J., and Montagnes, D. J. (2003). Protists decrease in size D. (2011). The contribution of coccolithophores to the optical and inorganic carbon linearly with temperature: ca. 2.5% c– 1. Proc. R. Soc. London. Ser. B: Biol. Sci. 270, budgets during the southern ocean gas exchange experiment: new evidence in support 2605–2611. doi: 10.1098/rspb.2003.2538 of the “Great calcite belt” hypothesis. J. Geophysical Research: Oceans 116. doi: 10.1029/ Auger, M., Morrow, R., Kestenare, E., Sallé e, J.-B., and Cowley, R. (2021). Southern 2011JC006941 ocean in-situ temperature trends over 25 years emerge from interannual variability. Baumann, K.-H., Böckel, B., and Frenz, M. (2004). “Coccolith contribution to south Nat. Commun. 12, 514. doi: 10.1038/s41467-020-20781-1 Atlantic carbonate sedimentation,” in Coccolithophores: from molecular processes to Bach, L. T., Bauke, C., Meier, K., Riebesell, U., and Schulz, K. G. (2012). Influence of global impact. Eds. H. R. Thierstein and J. R. Young (Berlin, Heidelberg: Springer Berlin changing carbonate chemistry on morphology and weight of coccoliths formed by Heidelberg), 367–402. Emiliania huxleyi. Biogeosciences 9, 3449–3463. doi: 10.5194/bg-9-3449-2012 Beaufort, L. (2005). Weight estimates of coccoliths using the optical properties Bach, L. T., Mackinder, L. C., Schulz, K. G., Wheeler, G., Schroeder, D. C., Brownlee, (birefringence) of calcite. Micropaleontology 51, 289–297. doi: 10.2113/ C., et al. (2013). Dissecting the impact of CO2 and pH on the mechanisms of gsmicropal.51.4.289 Frontiers in Marine Science 16 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 Beaufort, L., Barbarin, N., and Gally, Y. (2014). Optical measurements to determine Diner, R. E., Benner, I., Passow, U., Komada, T., Carpenter, E. J., and Stillman, J. H. the thickness of calcite crystals and the mass of thin carbonate particles such as (2015). Negative effects of ocean acidification on calcification vary within the coccoliths. Nat. Protoc. 9, 633. doi: 10.1038/nprot.2014.028 coccolithophore genus calcidiscus. Marine Biology 162, 1287–1305. doi: 10.1007/ Beaufort, L., Probert, I., de Garidel-Thoron, T., Bendif, E. M., Ruiz-Pino, D., Metzl, s00227-015-2669-x N., et al. (2011). Sensitivity of coccolithophores to carbonate chemistry and ocean Doney, S. C., Fabry, V. J., Feely, R. A., and Kleypas, J. A. (2009). Ocean acidification: acidification. Nature 476, 80–83. doi: 10.1038/nature10295 the other CO2 problem. Annu. Rev. Mar. Sci. 1, 169–192. doi: 10.1146/ Bé jard, T. M., Rigual-Herná ndez, A. S., Flores, J. A., Tarruella, J. P., Durrieu de annurev.marine.010908.163834 Madron, X., Cacho, I., et al. (2023). Calcification response of planktic foraminifera to Dugdale, R. C., Wilkerson, F. P., and Minas, H. J. (1995). The role of a silicate pump environmental change in the western Mediterranean Sea during the industrial era. in driving new production. Deep Sea Res. Part I: Oceanographic Res. Papers 42, 697– Biogeosciences 20, 1505–1528. doi: 10.5194/bg-20-1505-2023 719. doi: 10.1016/0967-0637(95)00015-X Bereiter, B., Eggleston, S., Schmitt, J., Nehrbass-Ahles, C., Stocker, T. F., Fischer, H., Durak, G. M., Taylor, A. R., Walker, C. E., Probert, I., de Vargas, C., Audic, S., et al. et al. (2015). Revision of the EPICA dome c CO2 record from 800 to 600 kyr before (2016). A role for diatom-like silicon transporters in calcifying coccolithophores. Nat. present. Geophysical Res. Lett. 42, 542–549. doi: 10.1002/2014GL061957 Commun. 7, 10543. doi: 10.1038/ncomms10543 Beusen, A., Boyd, P. W., Breitburg, D., Comeau, S., Dupont, S., Hansen, P. J., et al. Durand, A., Chase, Z., Noble, T. L., Bostock, H., Jaccard, S. L., Kitchener, P., et al. (2022). Multiple ocean stressors: a scientific summary for policy makers. Paris SP, (2017). Export production in the new-Zealand region since the last glacial maximum. France: Intergovernmental Oceanographic Commission of UNESCO (UNESCO-IOC). Earth Planetary Sci. Lett. 469, 110–122. doi: 10.1016/j.epsl.2017.03.035 Available at: https://unesdoc.unesco.org/ark:/48223/pf0000380891. Ebersbach, F., Trull, T. W., Davies, D. M., and Bray, S. G. (2011). Controls on Bishop, J. K. B., and Rossow, W. B. (1991). Spatial and temporal variability of global mesopelagic particle fluxes in the Sub-Antarctic and polar frontal zones in the southern surface solar irradiance. J. Geophysical Research: Oceans 96, 16839–16858. doi: 10.1029/ ocean south of Australia in summer–perspectives from free-drifting sediment traps. 91JC01754 Deep Sea Res. Part II: Topical Stud. Oceanography 58, 2260–2276. doi: 10.1016/ Bolton, C. T., Hernandez-Sanchez, M. T., Fuertes, M.-A., Gonzalez-Lemos, S., j.dsr2.2011.05.025 Abrevaya, L., Mendez-Vicente, A., et al. (2016). Decrease in coccolithophore Elderfield, H. (2002). Foraminiferal Mg/Ca paleothermometry: expected advances calcification and CO2 since the middle Miocene. Nat. Commun. 7. doi: 10.1038/ and unexpected consequences. Geochimica Cosmochimica Acta, A213–A213. ncomms10284 Eriksen, R., Trull, T. W., Davies, D., Jansen, P., Davidson, A. T., Westwood, K., et al. Bostock, H. C., Hayward, B. W., Neil, H. L., Currie, K. I., and Dunbar, G. B. (2011). (2018). Seasonal succession of phytoplankton community structure from autonomous Deep-water carbonate concentrations in the southwest Pacific. Deep Sea Res. Part I: sampling at the Australian southern ocean time series (SOTS) observatory. Mar. Ecol. Oceanographic Res. Papers 58, 72–85. doi: 10.1016/j.dsr.2010.11.010 Prog. Ser. 589, 13–31. doi: 10.3354/meps12420 Boyd, P. W., Crossley, A. C., DiTullio, G. R., Griffiths, F. B., Hutchins, D. A., Fabry, V. J., McClintock, J. B., Mathis, J. T., and Grebmeier, J. M. (2009). Ocean Queguiner, B., et al. (2001). Control of phytoplankton growth by iron supply and acidification at high latitudes: the bellweather. Oceanography 22, 160. doi: 10.5670/ irradiance in the subantarctic southern ocean: experimental results from the SAZ oceanog.2009.105 project. J. Geophysical Res. 106, 31573–31583. doi: 10.1029/2000JC000348 Feely, R. A., Sabine, C. L., Lee, K., Berelson, W., Kleypas, J., Fabry, V. J., et al. (2004). Bradtmiller, L. I., Anderson, R. F., Fleisher, M. Q., and Burckle, L. H. (2009). Impact of anthropogenic CO2 on the CaCO3 system in the oceans. Science 305, 362– Comparing glacial and Holocene opal fluxes in the pacific sector of the southern ocean. 366. doi: 10.1126/science.1097329 Paleoceanography 24. doi: 10.1029/2008PA001693 Feng, Y., Roleda, M. Y., Armstrong, E., Boyd, P. W., and Hurd, C. L. (2017). Broerse, A. T. C., Ziveri, P., van Hinte, J. E., and Honjo, S. (2000). Coccolithophore Environmental controls on the growth, photosynthetic and calcification rates of a export production, species composition, and coccolith-CaCO3 fluxes in the NE southern hemisphere strain of the coccolithophore Emiliania huxleyi. Limnology Atlantic (34°N21°W and 48°N21°W). Deep Sea Res. Part II: Topical Stud. Oceanography 62, 519–540. doi: 10.1002/lno.10442 Oceanography 47, 1877–1905. doi: 10.1016/S0967-0645(00)00010-2 Feng, Y., Roleda, M. Y., Armstrong, E., Law, C. S., Boyd, P. W., and Hurd, C. L. Buitenhuis, E. T., Pangerc, T., Franklin, D. J., Le Qué ré , C., and Malin, G. (2008). (2018). Environmental controls on the elemental composition of a southern Growth rates of six coccolithophorid strains as a function of temperature. Limnol hemisphere strain of the coccolithophore Emiliania huxleyi. Biogeosciences 15, 581– Oceanogr. 53, 1181–1185. doi: 10.4319/lo.2008.53.3.1181 595. doi: 10.5194/bg-15-581-2018 Caldeira, K., and Wickett, M. E. (2003). Anthropogenic carbon and ocean pH. Findlay, C. S., and Giraudeau, J. (2000). Extant calcareous nannoplankton in the Nature 425, 365. doi: 10.1038/425365a Australian sector of the southern ocean (austral summers 1994 and 1995). Mar. Carter, L., Neil, H. L., and Northcote, L. (2002). Late quaternary ice-rafting events in Micropaleontology 40, 417–439. doi: 10.1016/S0377-8398(00)00046-3 the SW pacific ocean, off eastern new Zealand. Mar. Geology 191, 19–35. doi: 10.1016/ Finkel, Z. V., Beardall, J., Flynn, K. J., Quigg, A., Rees, T. A. V., and Raven, J. A. S0025-3227(02)00509-1 (2009). Phytoplankton in a changing world: cell size and elemental stoichiometry. J. Charalampopoulou, A., Poulton, A. J., Bakker, D. C., Lucas, M. I., Stinchcombe, M. Plankton Res. 32, 119–137. doi: 10.1093/plankt/fbp098 C., and Tyrrell, T. (2016). Environmental drivers of coccolithophore abundance and Fiorini, S., Middelburg, J. J., and Gattuso, J.-P. (2011). Testing the effects of elevated calcification across drake passage (Southern ocean). Biogeosciences 13, 5917–5935. doi: pCO2 on coccolithophores (Prymnesiophyceae): comparison between haploid and 10.5194/bg-13-5917-2016 diploid life stages. J Phycol. 47, 1281–1291. doi: 10.1111/j.1529-8817.2011.01080.x Chase, Z., Ellwood, M. J., and van de Flierdt, T. (2018). Discovering the ocean’s past Flores, J. A., and Sierro, F. J. (1997). A revised technique for the calculation of through geochemistry. Elements: Int. Magazine Mineralogy Geochemistry Petrology 14, calcareous nannofossil accumulation rates. Micropaleontology 43, 321–324. doi: 397–402. doi: 10.2138/gselements.14.6.397 10.2307/1485832 Chen, B., and Liu, H. (2010). Relationships between phytoplankton growth and cell Frank, V. M., Brzesinski, M. A., Coale, K. H., and Nelson, D. M. (2000). Iron and size in surface oceans: interactive effects of temperature, nutrients, and grazing. silicic acid availability regulate Si uptake in the pacific sector of the southern ocean. Limnology Oceanography 55, 965–972. doi: 10.4319/lo.2010.55.3.0965 Deep-Sea Res. II 47 (15–16), 3315–38. doi: 10.1016/S0967-0645(00)00070-9 Collins, S., Rost, B., and Rynearson, T. A. (2014). Evolutionary potential of marine Frankignoulle, M., Canon, C., and Gattuso, J.-P. (1994). Marine calcification as a phytoplankton under ocean acidification. Evolutionary Appl. 7, 140–155. doi: 10.1111/ source of carbon dioxide: positive feedback of increasing atmospheric CO2. Limnology eva.12120 Oceanography 39, 458–462. doi: 10.4319/lo.1994.39.2.0458 Connell, R. D., and Sikes, E. L. (1997). Controls on late quaternary sedimentation of the Frankignoulle, M., and Gattuso, J.-P. (1993). Air-sea CO 2 exchange in coastal south Tasman rise. Aust. J. Earth Sci. 44, 667–675. doi: 10.1080/08120099708728344 ecosystems, interactions of c, n, p and s biogeochemical cycles and global change Cortese, G., and Prebble, J. (2015). A radiolarian-based modern analogue dataset for (Springer), 233–248. palaeoenvironmental reconstructions in the southwest pacific. Mar. Micropaleontology Fripiat, F., Leblanc, K., Elskens, M., Cavagna, A. J., Armand, L., André , L., et al. 118, 34–49. doi: 10.1016/j.marmicro.2015.05.002 (2011). Efficient silicon recycling in summer in both the polar frontal and subantarctic Dai, Y., Yu, J., Ren, H., and Ji, X. (2022). Deglacial subantarctic CO2 outgassing zones of the southern ocean. Mar. Ecol. Prog. Ser. 435, 47–61. doi: 10.3354/meps09237 driven by a weakened solubility pump. Nat. Commun. 13, 5193. doi: 10.1038/s41467- Fuertes, M.-Á ., Flores, J.-A., and Sierro, F. J. (2014). The use of circularly polarized 022-32895-9 light for biometry, identification and estimation of mass of coccoliths. Mar. Daniels, C. J., Poulton, A. J., Young, J. R., Esposito, M., Humphreys, M. P., Ribas- Micropaleontology 113, 44–55. doi: 10.1016/j.marmicro.2014.08.007 Ribas, M., et al. (2016). Species-specific calcite production reveals coccolithus pelagicus Gareth, J., Daniela, W., Trevor, H., and Robert, T. (2013). An introduction to as the key calcifier in the Arctic ocean. Mar. Ecol. Prog. Ser. 555, 29–47. doi: 10.3354/ statistical learning: with applications in r (Springer New York, NY: Spinger). Available meps11820 at: https://link.springer.com/book/10.1007/978-1-4614-7138-7#about-this-book. De Bodt, C., Van Oostende, N., Harlay, J., Sabbe, K., and Chou, L. (2010). Individual Gonzá lez-Lanchas, A., Flores, J.-A., Sierro, F. J., Bá rcena, M.Á ., Rigual-Herná ndez, and interacting effects of pCO 2 and temperature on emiliania huxleyi calcification: A. S., Oliveira, D., et al. (2020). A new perspective of the alboran upwelling system study of the calcite production, the coccolith morphology and the coccosphere size. reconstruction during the marine isotope stage 11: a high-resolution coccolithophore Biogeosciences 7, 1401–1412. doi: 10.5194/bg-7-1401-2010 record. Quaternary Sci. Rev. 245, 106520. doi: 10.1016/j.quascirev.2020.106520 Deppeler, S. L., and Davidson, A. T. (2017). Southern ocean phytoplankton in a Hastie, T. J., and Pregibon, D. (1992). Statistical models in s, chapter generalized changing climate. Front. Mar. Sci. 4. doi: 10.3389/fmars.2017.00040 linear models Vol. 51 (Pacific Grove: Wadsworth & Brooks/Cole). Available at: https:// Frontiers in Marine Science 17 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 www.scirp.org/(S(351jmbntv-nsjt1aadkposzje))/reference/referencespapers.aspx? Müller, M. N., Trull, T. W., and Hallegraeff, G. M. (2015). Differing responses of referenceid=2790482. three southern ocean Emiliania huxleyi ecotypes to changing seawater carbonate Honjo, S., Manganini, S. J., Krishfield, R. A., and Francois, R. (2008). Particulate chemistry. Mar. Ecol. Prog. Ser. 531, 81–90. doi: 10.3354/meps11309 organic carbon fluxes to the ocean interior and factors controlling the biological pump: Neil, H. L., Carter, L., and Morris, M. Y. (2004). Thermal isolation of Campbell a synthesis of global sediment trap programs since 1983. Prog. Oceanography 76, 217– plateau, new Zealand, by the Antarctic circumpolar current over the past 130 kyr. 285. doi: 10.1016/j.pocean.2007.11.003 Paleoceanography 19. doi: 10.1029/2003PA000975 ImageJ. Available at: http://imagej.nih.gov/ij. Nodder, S. D., Chiswell, S. M., and Northcote, L. C. (2016). Annual cycles of deep- Jackson, R., and Gabric, A. (2022). Climate change impacts on the marine cycling of ocean biogeochemical export fluxes in subtropical and subantarctic waters, southwest biogenic sulfur: a review. Microorganisms 10, 1581. doi: 10.3390/microorganisms10081581 pacific ocean. J. Geophysical Research: Oceans 121 (4), 2405–24. doi: 10.1002/ 2015JC011243 Jin, X., Liu, C., and Zhang, H. (2019). Coccolith morphological and assemblage responses to dissolution in the recent sediments of the East China Sea. Mar. Orr, J. C., Fabry, V. J., Aumont, O., Bopp, L., Doney, S. C., Feely, R. A., et al. (2005). Micropaleontology 152, 101709. doi: 10.1016/j.marmicro.2018.09.001 Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437, 681–686. doi: 10.1038/nature04095 Jonkers, L., Hillebrand, H., and Kucera, M. (2019). Global change drives modern plankton communities away from the pre-industrial state. Nature 570, 372–375. doi: Orsi, A. H., Whitworth Iii, T., and Nowlin, W. D.Jr. (1995). On the meridional extent 10.1038/s41586-019-1230-3 and fronts of the Antarctic circumpolar current. Deep Sea Res. Part I: Oceanographic Res. Papers 42, 641–673. doi: 10.1016/0967-0637(95)00021-W King, A. L., and Howard, W. R. (2005). d18O seasonality of planktonic foraminifera from southern ocean sediment traps: latitudinal gradients and implications for Paasche, E. (1973). Silicon and the ecology of marine plankton diatoms. i. paleoclimate reconstructions. Mar. Micropaleontology 56, 1–24. doi: 10.1016/ thalassiosira pseudonana (Cyclotella nana) grown in a chemostat with silicate as j.marmicro.2005.02.008 limiting nutrient. Mar. Biol. 19, 117–126. doi: 10.1007/BF00353582 Knappertsbusch, M., Cortes, M. Y., and Thierstein, H. R. (1997). Morphologic Paasche, E. (1999). Reduced coccolith calcite production under light-limited growth: variability of the coccolithophorid Calcidiscus leptoporus in the plankton, surface a comparative study of three clones of Emiliania huxleyi (Prymnesiophyceae). sediments and from the early pleistocene. Mar. Micropaleontology 30, 293–317. doi: Phycologia 38, 508–516. doi: 10.2216/i0031-8884-38-6-508.1 10.1016/S0377-8398(96)00053-9 Paasche, E. (2002). A review of the coccolithophorid Emiliania huxleyi Langer, G., and Bode, M. (2011). CO2 mediation of adverse effects of seawater (Prymnesiophyceae), with particular reference to growth, coccolith formation, and acidification in calcidiscus leptoporus. Geochemistry Geophysics Geosystems 12. doi: calcification-photosynthesis interactions. Phycologia 40, 503–529. doi: 10.2216/i0031- 10.1029/2010GC003393 8884-40-6-503.1 Langer, G., Geisen, M., Baumann, K.-H., Kläs, J., Riebesell, U., Thoms, S., et al. Pachauri, R. K., Allen, M. R., Barros, V. R., Broome, J., Cramer, W., Christ, R., et al. (2006). Species-specific responses of calcifying algae to changing seawater carbonate (2014). Climate change 2014: synthesis report. contribution of working groups I, II and chemistry. Geochemistry Geophysics Geosystems 7. doi: 10.1029/2005GC001227 III to the fifth assessment report of the intergovernmental panel on climate change (Geneva, Switzerland: Ipcc). https://www.ipcc.ch/report/ar5/syr/. Langer, G., Nehrke, G., Probert, I., Ly, J., and Ziveri, P. (2009). Strain-specific responses of Emiliania huxleyi to changing seawater carbonate chemistry. Pahnke, K., and Sachs, J. P. (2006). Sea Surface temperatures of southern Biogeosciences 6, 2637–2646. doi: 10.5194/bg-6-2637-2009 midlatitudes 0–160 kyr B.P. Paleoceanography 21. doi: 10.1029/2005PA001191 Langer, G., Oetjen, K., and Brenneis, T. (2012). Calcification of calcidiscus Panitz, S., Cortese, G., Neil, H. L., and Diekmann, B. (2015). A radiolarian-based leptoporus under nitrogen and phosphorus limitation. J. Exp. Mar. Biol. Ecol. 413, palaeoclimate history of core Y9 (Northeast of Campbell plateau, new Zealand) for the 131–137. doi: 10.1016/j.jembe.2011.11.028 last 160 kyr. Mar. Micropaleontology 116, 1–14. doi: 10.1016/j.marmicro.2014.12.003 Langer, G., Probert, I., Nehrke, G., and Ziveri, P. (2011). The morphological response Patil, S. M., Mohan, R., Shetye, S. S., Gazi, S., Baumann, K.-H., and Jafar, S. (2017). of emiliania huxleyi to seawater carbonate chemistry changes: an inter-strain Biogeographic distribution of extant coccolithophores in the Indian sector of the southern comparison. J. Nannoplankton Res. 32 (1), 29–34. ocean. Mar. Micropaleontology 137, 16–30. doi: 10.1016/j.marmicro.2017.08.002 Langer, G., Taylor, A. R., Walker, C. E., Meyer, E. M., Ben Joseph, O., Gal, A., et al. Petit, J. R., Jouzel, J., Raynaud, D., Barkov, N. I., Barnola, J. M., Basile, I., et al. (1999). (2021). Role of silicon in the development of complex crystal shapes in Climate and atmospheric history of the past 420,000 years from the vostok ice core, coccolithophores. New Phytol. 231, 1845–1857. doi: 10.1111/nph.17230 Antarctica. Nature 399, 429–436. doi: 10.1038/20859 Lannuzel, D., Bowie, A. R., Remenyi, T., Lam, P., Townsend, A., Ibisanmi, E., et al. Pörtner, H. O., Langenbuch, M., and Michaelidis, B. (2005). Synergistic effects of (2011). Distributions of dissolved and particulate iron in the sub-Antarctic and polar temperature extremes, hypoxia, and increases in CO2 on marine animals: from earth frontal southern ocean (Australian sector). Deep Sea Res. Part II: Topical Stud. history to global change. J. Geophysical Research: Oceans 110. doi: 10.1029/ Oceanography 58, 2094–2112. doi: 10.1016/j.dsr2.2011.05.027 2004JC002561 Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A. C. M., and Levrard, B. Prebble, J. G., Crouch, E. M., Carter, L., Cortese, G., and Nodder, S. D. (2013). (2004). A long-term numerical solution for the insolation quantities of the earth. A&A Dinoflagellate cysts from two sediment traps east of new Zealand. Mar. 428, 261–285. doi: 10.1051/0004-6361:20041335 Micropaleontology 104, 25–37. doi: 10.1016/j.marmicro.2013.08.003 Lefebvre, S. C., Benner, I., Stillman, J. H., Parker, A. E., Drake, M. K., Rossignol, P. E., Quinn, P., Thierstein, H. R., Brand, L., and Winter, A. (2003). Experimental evidence et al. (2012). Nitrogen source and pCO2 synergistically affect carbon allocation, growth for the species character of calcidiscus leptoporus morphotypes. J. Paleontology 77, and morphology of the coccolithophore emiliania huxleyi: potential implications of 825–830. doi: 10.1666/0022-3360(2003)077<0825:EEFTSC>2.0.CO;2 ocean acidification for the carbon cycle. Global Change Biol. 18, 493–503. doi: 10.1111/ Raven, J., Caldeira, K., Elderfield, H., Hoegh-Guldberg, O., Liss, P., Riebesell, U., et al. j.1365-2486.2011.02575.x (2005). Ocean acidification due to increasing atmospheric carbon dioxide (London: The Malinverno, E., Triantaphyllou, M. V., and Dimiza, M. D. (2015). Coccolithophore Royal Society). Available at: https://royalsociety.org/-/media/Royal_Society_Content/ assemblage distribution along a temperate to polar gradient in the West pacific sector of policy/publications/2005/9634.pdf. the southern ocean (January 2005). Micropaleontology 61, 489–506. doi: 10.47894/ Renaud, S., Ziveri, P., and Broerse, A. T. C. (2002). Geographical and seasonal mpal.61.6.07 differences in morphology and dynamics of the coccolithophore calcidiscus leptoporus. McNeil, B. I., and Matear, R. J. (2008). Southern ocean acidification: a tipping point Mar. Micropaleontology 46, 363–385. doi: 10.1016/S0377-8398(02)00081-6 at 450-ppm atmospheric CO2. Proc. Natl. Acad. Sci. 105, 18860–18864. doi: 10.1073/ Reusch, T. B. H., and Boyd, P. W. (2013). Experimental evolution meets marine pnas.0806318105 phytoplankton. Evolution 67, 1849–1859. doi: 10.1111/evo.12035 Meier, K. J. S., Beaufort, L., Heussner, S., and Ziveri, P. (2014). The role of ocean Riebesell, U., Zondervan, I., Rost, B., Tortell, P. D., Zeebe, R. E., and Morel, F. M. M. acidification in Emiliania huxleyi coccolith thinning in the Mediterranean Sea. (2000). Reduced calcification of marine plankton in response to increased atmospheric Biogeosciences 11, 2857–2869. doi: 10.5194/bg-11-2857-2014 CO2. Nature 407, 364–367. doi: 10.1038/35030078 Meyer, J., and Riebesell, U. (2015). Reviews and syntheses: responses of Rigual-Herná ndez, A. S., Sá nchez-Santos, J. M., Eriksen, R., Moy, A. D., Sierro, F. J., coccolithophores to ocean acidification: a meta-analysis. Biogeosciences (BG) 12, Flores, J. A., et al. (2020a). Limited variability in the phytoplankton emiliania huxleyi 1671–1682. doi: 10.5194/bg-12-1671-2015 since the pre-industrial era in the subantarctic southern ocean. Anthropocene 31, Milliman, J. D. (1993). Production and accumulation of calcium carbonate in the 100254. doi: 10.1016/j.ancene.2020.100254 ocean: budget of a nonsteady state. Global Biogeochemical Cycles 7, 927–957. doi: Rigual-Herná ndez, A. S., Trull, T. W., Bray, S. G., Cortina, A., and Armand, L. K. 10.1029/93GB02524 (2015). Latitudinal and temporal distributions of diatom populations in the pelagic Monahan, J. F. (2011). Numerical methods of statistics (Cambridge: Cambridge waters of the subantarctic and polar frontal zones of the southern ocean and their role University Press). Available at: https://www.cambridge.org/core/books/numerical- in the biological pump. Biogeosciences 12, 8615–8690. doi: 10.5194/bg-12-5309-2015 methods-of-statistics/ED2D1845F52AF845CCF560E3526B9B56#fndtn-information. Rigual-Herná ndez, A. S., Trull, T. W., Flores, J. A., Nodder, S. D., Eriksen, R., Davies, Moy, A. D., Howard, W. R., Bray, S. G., and Trull, T. W. (2009). Reduced D. M., et al. (2020b). Full annual monitoring of subantarctic emiliania huxleyi calcification in modern southern ocean planktonic foraminifera. Nat. Geosci 2, 276– populations reveals highly calcified morphotypes in high-CO2 winter conditions. Sci. 280. doi: 10.1038/ngeo460 Rep. 10, 2594. doi: 10.1038/s41598-020-59375-8 Müller, M. N., Brandini, F. P., Trull, T. W., and Hallegraeff, G. M. (2021). Coccolith Rigual-Herná ndez, A. S., Trull, T. W., Nodder, S. D., Flores, J. A., Bostock, H., volume of the southern ocean coccolithophore emiliania huxleyi as a possible indicator Abrantes, F., et al. (2020c). Coccolithophore biodiversity controls carbonate export in for palaeo-cell volume. Geobiology 19, 63–74. doi: 10.1111/gbi.12414 the southern ocean. Biogeosciences 17, 245–263. doi: 10.5194/bg-17-245-2020 Frontiers in Marine Science 18 frontiersin.org Rigual-Hernández et al. 10.3389/fmars.2023.1159884 Rintoul, S. R., and Trull, T. W. (2001). Seasonal evolution of the mixed layer in the zones of the southern ocean, south of Australia. J. Geophysical Research: Oceans 106, subantarctic zone south of Australia. J. Geophysical Research: Oceans 106, 31447– 31489–31509. doi: 10.1029/2000JC000308 31462. doi: 10.1029/2000JC000329 Trull, T. W., Passmore, A., Davies, D. M., Smit, T., Berry, K., and Tilbrook, B. (2018). Rosengard, S. Z., Lam, P. J., Balch, W. M., Auro, M. E., Pike, S., Drapeau, D., et al. Distribution of planktonic biogenic carbonate organisms in the southern ocean south (2015). Carbon export and transfer to depth across the southern ocean great calcite belt. of Australia: a baseline for ocean acidification impact assessment. Biogeosciences 15, 31– Biogeosciences 12, 3953–3971. doi: 10.5194/bg-12-3953-2015 49. doi: 10.5194/bg-15-31-2018 Saavedra-Pellitero, M., and Baumann, K.-H. (2015). Comparison of living and Trull, T. W., Schulz, E., Bray, S. G., Pender, L., McLaughlan, D., Tilbrook, B., et al. surface sediment coccolithophore assemblages in the pacific sector of the southern (2010). The Australian integrated marine observing system southern ocean time series ocean. Micropaleontology 61, 507–520. doi: 10.47894/mpal.61.6.08 facility (Sydney: OCEANS 2010 IEEE), 1–7. Sá ez, A. G., Probert, I., Geisen, M., Quinn, P., Young, J. R., and Medlin, L. K. (2003). Venables, W., Ripley, B., Venables, W., and Ripley, B. (2002). Exploratory Pseudo-cryptic speciation in coccolithophores. Proc. Natl. Acad. Sci. 100, 7163–7168. multivariate analysis. Modern Appl. Stat S, 301–330. doi: 10.1007/978-0-387-21706- doi: 10.1073/pnas.1132069100 2_11 Schlitzer, R. (2018). Ocean Data View. Available at: https://odv.awi.de/fileadmin/ Walker, C. E., Taylor, A. R., Langer, G., Durak, G. M., Heath, S., Probert, I., et al. user_upload/odv/misc/HowTo.pdf. (2018). The requirement for calcification differs between ecologically important Schlüter, L., Lohbeck, K. T., Gutowska, M. A., Gröger, J. P., Riebesell, U., and Reusch, coccolithophore species. New Phytol. 220, 147–162. doi: 10.1111/nph.15272 T. B. (2014a). Adaptation of a globally important coccolithophore to ocean warming Young, J. R., and Ziveri, P. (2000). Calculation of coccolith volume and it use in and acidification. Nat. Climate Change 4, 1024–1030. doi: 10.1038/nclimate2379 calibration of carbonate flux estimates. Deep Sea Res. Part II: Topical Stud. Schlüter, L., Lohbeck, K. T., Gutowska, M. A., Gröger, J. P., Riebesell, U., and Reusch, Oceanography 47, 1679–1700. doi: 10.1016/S0967-0645(00)00003-5 T. B. H. (2014b). Adaptation of a globally important coccolithophore to ocean warming Young, J., Geisen, M., Cross, L., Kleijne, A., Sprengel, C., Probert, I., et al (2003). A and acidification. Nat. Climate Change 4, 1024–1030. doi: 10.1038/nclimate2379 guide to extant coccolithophore taxonomy. Journal of Nanoplankton Research Special Issue 1, International Nannoplankton Association. Sedwick, P. N., DiTullio, G. R., Hutchins, D. A., Boyd, P. W., Griffiths, F. B., Crossley, A. C., et al. (1999). Limitation of algal growth by iron deficiency in the Australian Young, J. R., Bown, P. R., and Lees, J. A. (2022). Nannotax3 website. International subantarctic region. Geophysical Res. Lett. 26, 2865–2868. doi: 10.1029/1998GL002284 Nannoplankton Association. Accessed 2022. Available at: https://www.mikrotax.org/ Nannotax3. Shadwick, E. H., Rigual-Herná ndez, A. S., Eriksen, R., Jansen, P., Davies, D. M., Wynn-Edwards, C. A., et al. (2022). Changes in southern ocean biogeochemistry and Ziveri, P., de Bernardi, B., Baumann, K.-H., Stoll, H. M., and Mortyn, P. G. (2007). the potential impact on pH-sensitive planktonic organisms. Oceanography 34, 14–15. Sinking of coccolith carbonate and potential contribution to organic carbon ballasting doi: 10.5670/oceanog.2021.supplement.02-06 in the deep ocean. Deep Sea Res. Part II: Topical Stud. Oceanography 54, 659–675. doi: 10.1016/j.dsr2.2007.01.006 Shadwick, E. H., Tilbrook, B., Cassar, N., Trull, T. W., and Rintoul, S. R. (2015). Summertime physical and biological controls on O2 and CO2 in the Australian sector Zondervan, I. (2007). The effects of light, macronutrients, trace metals and CO2 on of the southern ocean. J. Mar. Syst. 147, 21–28. doi: 10.1016/j.jmarsys.2013.12.008 the production of calcium carbonate and organic carbon in coccolithophores–a review. Deep Sea Res. Part II: Topical Stud. Oceanography 54, 521–537. doi: 10.1016/ ter Braak, C. J. F., and Verdonschot, P. F. M. (1995). Canonical correspondence j.dsr2.2006.12.004 analysis and related multivariate methods in aquatic ecology. Aquat. Sci. 57, 255–289. Zondervan, I., Rost, B., and Riebesell, U. (2002). Effect of CO2 concentration on the doi: 10.1007/BF00877430 PIC/POC ratio in the coccolithophore emiliania huxleyi grown under light-limiting Trull, T. W., Bray, S. G., Manganini, S. J., Honjo, S., and François, R. (2001). Moored conditions and different daylengths. J. Exp. Mar. Biol. Ecol. 272, 55–70. doi: 10.1016/ sediment trap measurements of carbon export in the subantarctic and polar frontal S0022-0981(02)00037-0 Frontiers in Marine Science 19 frontiersin.org