the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Jun 2026
Taxon-specific coccolithophore carbonate stocks and production across a Pacific trophic gradient from oligotrophic gyre to upwelling zone
Abstract. Coccolithophores contribute substantially to marine carbonate production, yet species-resolved estimates of coccosphere-associated CaCO₃ stocks remain scarce across large-scale trophic gradients. Here we quantify multispecific coccosphere carbonate inventories and first-order production estimates along the BIOSOPE transect across the South Pacific Gyre, one of the most strongly stratified regions of the global ocean. Estimates were obtained using an upgraded SYRACO automated recognition workflow combined with bidirectional circular polarization (BCP) imaging, allowing direct reconstruction of coccosphere calcite content at the cellular scale. Coccolithophore assemblages reproduced the major regional and vertical ecological structures previously described across the transect, including the transition from nutrient-rich surface communities near the Marquesas Archipelago and the eastern upwelling-influenced region to vertically structured assemblages in the oligotrophic gyre. At the transect scale, CaCO₃ standing stocks were primarily controlled by these large-scale trophic gradients, with maxima located in surface waters of mesotrophic regions and at depth within the deep chlorophyll maximum in the stratified gyre, indicating a vertical decoupling between coccosphere abundance and carbonate inventories under oligotrophic conditions. Taxon-specific analyses show that carbonate inventories reflect both coccosphere abundance and strong interspecific contrasts in calcite content. Isochrysidales, particularly Emiliania huxleyi, dominated carbonate stocks through their numerical abundance, whereas more heavily calcified taxa contributed disproportionately relative to their cell densities. Despite structuring assemblages in the lower euphotic zone, deep-dwelling taxa represented only a minor fraction of suspended carbonate stocks in the water column, contrasting with their major contribution to sedimentary assemblages in oligotrophic regions. First-order CaCO₃ production estimates followed spatial patterns similar to standing stocks, with enhanced production in mesotrophic regions and a sustained contribution from oligotrophic communities. Although uncertainties remain related to species-specific growth rates, these estimates provide the first multispecific constraints on coccolithophore calcification dynamics in the southeastern Pacific. Overall, this study demonstrates that combining automated coccosphere recognition with BCP-based carbonate measurements provides a robust framework for resolving species-level carbonate inventories across major oceanic trophic gradients and represents an important step toward quantitative assessments of coccolithophore contributions to pelagic carbonate cycling at basin scale.
- Preprint
(2263 KB) - Metadata XML
-
Supplement
(132 KB) - BibTeX
- EndNote
Status: open (extended)
-
CC1: 'Comment on egusphere-2026-3055', Francisco Díaz-Rosas, 03 Oct 2026
reply
-
CC2: 'Reply on CC1', Francisco Díaz-Rosas, 03 Oct 2026
reply
Here are the references for the studied cited in the comment above, including the PANGAEA dataset containing the abundance and PIC data used in Díaz-Rosas et al. (2018):
Beaufort, L., Couapel, M., Buchet, N., Claustre, H., and Goyet, C.: Calcite production by coccolithophores in the south east Pacific Ocean, Biogeosciences, 5, 1101–1117, https://doi.org/10.5194/bg-5-1101-2008, 2008.
Beaufort, L., Probert, I., de Garidel-Thoron, T., Bendif, E. M., Ruiz-Pino, D., Metzl, N., Goyet, C., Buchet, N., Coupel, P., Grelaud, M., Rost, B., Rickaby, R. E. M., and de Vargas, C.: Sensitivity of coccolithophores to carbonate chemistry and ocean acidification, Nature, 476, 80–83, https://doi.org/10.1038/nature10295, 2011.
Díaz-Rosas, F. J., Vargas, C. A., and von Dassow, P.: Particulate inorganic carbon pools by coccolithophores in low-oxygen–low-pH waters off the Southeast Pacific margin, Biogeosciences, 22, 4405–4422, https://bg.copernicus.org/articles/22/4405/2025/.
Díaz-Rosas, F., Vargas, C. A., and von Dassow, P.: Particulate Inorganic Carbon (PIC) and associated coccospheres and detached coccoliths in waters off the Southeast Pacific margin in 2018, PANGAEA [data set], https://doi.pangaea.de/10.1594/PANGAEA.975784.
Citation: https://doi.org/10.5194/egusphere-2026-3055-CC2 -
CC3: 'Reply on CC2', Francisco Díaz-Rosas, 05 Oct 2026
reply
As an additional point to my previous comment, coccosphere-only PIC calculated from the underlying Díaz-Rosas et al. (2025) data reached 0.164 mmol CaCO3 m-3 during late-spring 2015 at station T2 (2 meter depth, 20.109°S, 70.25°W) with 387235 coccosphere L-1 and 0.219 mmol CaCO3 m-3 with 516835 coccospheres L-1 during mid-summer 2018 at station T5 (30 m depth, 20.10°S, 70.90°W). These values exclude detached coccoliths and provide additional regional context rather than a direct comparison with the present study, given the different sampled locations, years, and methodologies. Closer to the latitude of STB18, Hyd8 (32.50°S, 72.70°W) reached 0.057 mmol CaCO3 m-3 in surface waters. Together, these observations illustrate the regional variability in coccosphere-associated CaCO3 stocks across the southeastern Pacific margin.
Citation: https://doi.org/10.5194/egusphere-2026-3055-CC3
-
CC3: 'Reply on CC2', Francisco Díaz-Rosas, 05 Oct 2026
reply
-
CC2: 'Reply on CC1', Francisco Díaz-Rosas, 03 Oct 2026
reply
-
RC1: 'Comment on egusphere-2026-3055', Yuye Han, 06 Oct 2026
reply
General comments
The article by Lassus et al. builds on Beaufort et al. (2008) by combining an upgraded SYRACO automated coccosphere recognition workflow with bidirectional circular polarization imaging to quantify coccolithophore communities and their contribution to carbonate standing stocks along the BIOSOPE transect. The approach enables reconstruction of coccosphere calcite content at the cellular scale and is used to quantify species abundance and composition, as well as to estimate taxon-specific standing stocks and CaCO3 production. The manuscript includes methodological validation through comparisons between SYRACO-derived counts and manual counts, providing useful support for automated coccolithophore identification and quantification. In addition, the study explores the distribution of coccolithophore assemblages and their contributions to CaCO3 standing stocks across environmental gradients along the BIOSOPE transect.
Overall, the study combines methodological development with a valuable ecological dataset and provides useful information on the contribution of coccolithophores to marine CaCO3 cycling. However, the manuscript currently covers a broad range of topics, including method validation, community structure, CaCO3 standing stocks, and production. The considerable breadth of the manuscript leaves limited room for a sufficiently detailed analysis and discussion of the data presented. I have a few comments that I hope will help strengthen the manuscript and improve its focus.
In the Introduction, more background should be provided on the previous application of SYRACO by Beaufort et al. (2008), including a brief summary of their main findings on coccosphere-associated carbonate stocks for Isochrysidales and the questions that remained unresolved, particularly the contribution of other coccolithophore groups to the total PIC inventory. The authors should then more clearly highlight how the present SYRACO framework has been upgraded relative to this earlier work, particularly in terms of calcite quantification and multispecific coccosphere recognition, and explain how these developments enable a broader taxon-specific assessment of CaCO3 standing stocks. In addition, the ecological background currently presented in Section 2.2 would be more appropriate in the Introduction and could be condensed and incorporated there.
There are some missing methodological details regarding the calculation of standing stocks. It is not clear how the integration depth was defined for each station and whether it was simply based on the maximum sampling depth available. If so, the authors should clarify whether this depth is intended to represent the euphotic zone depth.
The Results section contains a considerable amount of detailed information, but would benefit from further focusing to better emphasize the findings most directly relevant to the main objectives of the study. (1) Sections 3.1.1 and 3.1.4 contain extensive taxon-specific results, but much of this information is not supported by corresponding figures or tables. (2) The transect-wide integrations of abundance, CaCO3 standing stock, and production appear to add limited information beyond the clearer regional and station-based patterns and could be reconsidered. Table S1 is not cited in the main text. (3) Section 3.2.1 on coccosphere masses and diameters is largely descriptive. These results should either be explored in greater depth and discussed in relation to CaCO3 standing stocks or moved to the Supplementary Material. (4) Regional mean values are frequently reported, and the corresponding standard deviations should be provided.
The discussion of coccolithophore ecology and carbonate standing stocks remains relatively limited. Section 3.1.3 should be better integrated into the discussion section, where the observed patterns could be interpreted in the context of the extensive environmental data already available from the BIOSOPE cruise. Relevant studies from the same cruise, including Raimbault et al. (2008) and Perrin et al. (2016), should be more fully incorporated into the discussion.
Additional minor comments
Lns 21-28: Include numerical values when describing the observed trends. This would make the differences more explicit and allow readers to assess the magnitude of the changes more directly.
Ln 43: “Planktonic gastropods” is a broad taxonomic group, so describing them collectively as producing aragonite may be overly general.
Ln 65-69: Clarify the major difficulties in quantifying taxon-specific coccosphere carbonate standing stocks and more explicitly highlight the advantages of the proposed approach compared with existing methods, such as those based on scanning electron microscopy.
Ln 76: Provide the full name of “SYRACO”.
Ln 85: It would be helpful to indicate the SPG and other major oceanographic regions on the map.
Ln 271: More detail is needed on the depth ranges used for vertical integration at each station.
Ln 359: Consider supporting this comparison with a quantitative diversity metric, such as the Shannon diversity index.
Ln 384: Consider using a color bar to represent sampling depth rather than including depth information in each sample label. In addition, PC1 and PC2 together explain only about 33.8% of the total variance, and this relatively low proportion should be considered when interpreting the ordination patterns. The analysis could also be strengthened by incorporating available nutrient data to better evaluate the environmental factors associated with the observed community structure.
Ln 405: Consider moving the regression equation and R2 value into the plotting area and adding the sample size (N) and corresponding p-value. “Human estimates” should also be replaced with “manual counts” for consistency with the main text.
Ln 462: Add the standard deviation associated with the mean value and check the number of significant digits used for values such as 515 × 10⁻3 mmol m⁻2.
Ln 473: Add a citation to Table 2.
Ln 505: Check the formatting of subscripts and superscripts in the figure and ensure that these are presented consistently throughout the manuscript.
Ln 527: Provide an estimate of the uncertainty associated with the calculated CaCO3 standing stocks. In addition, check the significant digits used when reporting the relative contributions.
Ln 540: Clarify how Isochrysidales recognition has improved relative to Beaufort et al. (2008), particularly the automated discrimination of Emiliania and Gephyrocapsa.
Ln 550: The explanation for weak correlations in rare taxa should be more cautious, as other factors may also contribute.
Ln 579: Should be revised to “Ecological structure of coccolithophore communities”.
Ln 590: Consider incorporating environmental variables to better support the inferred niche differentiation or alternatively discussing how cell size may relate to environmental conditions across the transect.
Ln 631: Clarify whether these “heavily calcified” taxa have larger coccospheres, more coccoliths per cell, or a higher calcite content per individual coccolith, and relate these differences to previous trait-based studies of coccolithophore calcification.
Ln 649: Consider adding quantitative comparisons among regions to support the statement.
Ln 685: Support this statement with quantitative comparisons between taxon-specific abundance and CaCO3 contribution.
Citation: https://doi.org/10.5194/egusphere-2026-3055-RC1
Data sets
Dataset for: Taxon-specific coccolithophore carbonate stocks and production across a Pacific trophic gradient from oligotrophic gyre to upwelling zone Guillaume Lassus, Luc Beaufort, Martine Couapel, and David Mendels https://doi.org/10.5281/zenodo.20419476
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 168 | 78 | 39 | 285 | 37 | 26 | 22 |
- HTML: 168
- PDF: 78
- XML: 39
- Total: 285
- Supplement: 37
- BibTeX: 26
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
I suggest expanding the discussion of the CaCO3 standing stocks by considering independent observations available farther east in the southeastern Pacific. This comparison may be particularly informative because STB18 (32.69° S, 84.07° W) is the easternmost station retained in the present quantitative analysis (Table S1), whereas the more easterly PCU stations were excluded because uncertainty in the effective filtered volume prevented reliable estimates of coccolithophore concentrations and associated CaCO3 standing stocks. Thus, the quantitative reconstruction presented here does not extend into the productive waters closer to the South American continental margin.
At the eastern end of the BIOSOPE section retained in the present analysis, Fig. 7 indicates coccosphere-associated CaCO3 stocks reaching approximately 0.04 mmol CaCO3 m−3 in the upper water column. However, the original BIOSOPE observations provide important additional context for interpreting this apparent eastern boundary. Previous analyses of the BIOSOPE material (Beaufort et al., 2008, 2011) showed a pronounced increase in coccosphere mass from STB18 toward the Chilean upwelling region. Therefore, although the more coastal BIOSOPE stations cannot be used here to derive reliable volumetric CaCO3 standing stocks because of the uncertainty in filtered volume, their exclusion should not be interpreted as evidence that STB18 represents the eastern limit of the biological pattern in coccolithophore calcification.
Independent quantitative observations farther east are available from Díaz-Rosas et al. (2025), who measured particulate inorganic carbon along the southeastern Pacific margin using both analytically determined total PIC (PICTotal; their Fig. 5a) and PIC reconstructed from coccolithophore abundances (PICCocco; their Fig. 5b). Across their 2018 dataset, PICTotal ranged from 0.08 to 5.86 µg C L−1 and PICCocco from 0.01 to 3.69 µg C L−1. PICCocco included both intact coccospheres and detached coccoliths and was positively related to the independently measured PICTotal. Within the upper 100 m, coccospheres plus detached coccoliths accounted for almost half of PICTotal on average (their Fig. 5d–e).
Of particular relevance, Hyd6 (32.50° S, 72.70° W) lies at almost the same latitude as STB18 but approximately 1,000 km farther east, toward the continental margin. Considering only the PIC associated with intact coccospheres, and therefore excluding detached coccoliths to provide a more consistent with the carbonate pool quantified in the present study, coccosphere-associated PIC at Hyd6 was equivalent to 0.055 and 0.057 mmol CaCO3 m−3 at 5 and 10 m, respectively (F. Díaz-Rosas, unpublished data underlying Díaz-Rosas et al., 2025). These values decreased with depth to 0.024, 0.018, and 0.003 mmol CaCO3 m−3 at 25, 50, and 100 m, respectively ((F. Díaz-Rosas, unpublished data underlying Díaz-Rosas et al., 2025). Farther south, Hyd7 (33.96° S, 73.12° W) showed a different vertical structure, with substantially lower surface values (~0.003–0.004 mmol m−3 at 5–10 m), increasing to ~0.006 mmol m−3 at 25 m and ~0.018 mmol m−3 at 50 m, before declining again to ~0.002 mmol m−3 at 100 m (F. Díaz-Rosas, unpublished data underlying Díaz-Rosas et al., 2025).
These observations are particularly relevant in the context of the original BIOSOPE measurements, which indicated increasing coccosphere mass from STB18 toward the Chilean upwelling region. Rather than constituting a direct quantitative continuation of the BIOSOPE transect, the Díaz-Rosas et al. (2025) observations provide independent quantitative estimates of coccosphere-associated PIC in the continental-margin region where the present study does not report CaCO3 standing stocks. At Hyd6, surface coccosphere-associated stocks of ~0.055–0.057 mmol CaCO3 m−3 are comparable in magnitude to the ~0.04 mmol m−3 maximum reconstructed near the eastern end of the BIOSOPE section retained in Fig. 7, while the contrasting Hyd7 profile illustrates substantial spatial variability within the margin. In this sense, the Díaz-Rosas et al. dataset complements, rather than simply extends, the present reconstruction and provides useful context for discussing how coccolithophore-associated carbonate may vary eastward beyond STB18.
The comparison is not methodologically one-to-one and should be presented accordingly. Díaz-Rosas et al. (2025) estimated individual coccolith PIC quotas from coccolith morphometry, taxon-specific shape factors, and calcite density (2.7 g cm−3), subsequently reconstructing the coccosphere contribution using published estimates of the number of coccoliths per coccosphere. For G. huxleyi, 17 coccoliths per coccosphere were assumed, and because taxonomic resolution was limited for the 2018 samples, a G. huxleyi quota of 2.5 pg CaCO3 per coccolith was applied as the maximum threshold. In contrast, the present study estimates coccosphere calcite mass directly from birefringence using bidirectional circular polarization microscopy, converting calcite thickness into mass and applying taxon-specific corrections where crystallographic orientation can lead to underestimation. The present study then combines coccosphere abundance and individual coccosphere mass and deliberately excludes free coccoliths from the standing-stock calculation because these may include previously detached material transported through the water column.
There is also an important temporal constraint: BIOSOPE was sampled during austral spring 2004, whereas Hyd6–Hyd7 were sampled during mid-summer 2018. Consequently, these datasets should not be interpreted as a direct temporal or station-to-station comparison. Nevertheless, the close latitudinal correspondence between STB18 and Hyd6, the large longitudinal separation between them, the increase in coccosphere mass toward the coast evident in the original BIOSOPE observations, and the independent quantitative PIC estimates available from the continental margin together provide useful regional context. I therefore suggest that Díaz-Rosas et al. (2025) be considered in the Discussion, while explicitly acknowledging the differences in sampling year, seasonal timing, spatial setting, and methodology.