We thank Reviewer 1 (Marius N. Müller) for providing constructive feedback on our manuscript and suggestions for areas of improvement.

Response to General comments:

1 & 2: We will take on board feedback on the language use for the revised manuscript and will omit unnecessary adverbs from the text and improve statements that are generally vague. For instance, quantifying a magnitude of change rather than using general descriptive term, especially when these occur in the results or when talking about the results. We hope that this improves the readability and accuracy of the manuscript.

3. The approximate changes in salinity that are projected to be caused by climate change will be added to the introduction for comparison with recent observations.

4. The carbonate chemistry data collected during these experiments has previously been reported in an earlier publication (Gebühr et al. 2021) and can be found as a supplement to this publication. As this information was not highlighted in the original submission, the availability of this information will be added to the relevant section of the Methods and the data availability section. We agree that cell density is relatively high, with the potential to affect carbonate chemistry and/or nutrient availability. The discussion will be revised in appropriate sections to state the change in carbonate chemistry between the beginning and end of the experiments and to additionally discuss the results in view of these data, e.g. adding to lines 400-403 to highlight the possibility of nutrient depleted conditions towards the end of the experiment linked to our data for division rates and discussing our coccolith size data in relation to carbonate chemistry.

5. The samples run through the CASY cell counter were not acidified in advance to remove the coccosphere for the purposes of this experiment. However, CASY cell counters actually measure a size that is intermediate between coccosphere size and cell size and therefore underestimate coccosphere volume (Gerecht et al. 2015) and are not true measure of cell or coccosphere size. CASY size measurements are, however, very useful for rapidly monitoring changes in size over time. We therefore agree that it is confusing to label the CASY-derived data with either “cell size” or “coccosphere size”. We will instead amend reference to CASY-derived size measurements as “size” and add a statement to the methods to state that CASY measures size between cell and coccosphere size and is used to monitor the development of the cell division cycle.

6. We have used a morphometric-based approach to estimate coccolith and cellular PIC (Young and Ziveri, 2000), which does introduce some uncertainty in the estimated PIC values. However, we aim to minimise some of the uncertainty by using shape factors that are derived from these two specific strains under the same range of salinity conditions (Linge Johnson et al. 2019). We thank the reviewer for noting that our estimated coccolith and cell PIC values may be lower than published in other literature, as this raises some additional interesting discussion points, and we confirm that it is the case that our coccolith PIC values are lower than other published values from “morphotype A” strains in some instances. For some of the examples mentioned by the reviewer, in Valença et al. (2024) morphometric-based coccolith PIC values of 0.22-0.28 pg C are reported for E. huxleyi Type A and in Müller et al. (2012), coccolith PIC for nutrient-limited E. huxleyi cultures was 0.17-0.20 pg C and for nutrient-replete cultures was 0.49-0.94 pg C (all converted from published pg CaCO₃). These published values are higher than our mean coccolith PIC values for both strains across all salinity treatments (0.077-0.191 pg C), although our data are in reasonable agreement to those reported for nutrient-deplete cultures from Müller et al. (2012). There are several potential explanations for discrepancies in coccolith PIC and cell PIC between studies. Firstly, as we mention in the discussion (section 4.4), differences in coccolith size strongly influence coccolith PIC (and therefore cell PIC), especially as estimated using morphometrics. Therefore, any difference in the size of coccoliths between our cultures and those of other studies may influence the reported coccolith and cell PIC values. This also means that differences in coccolith size between strains (either in response to or independent of any environmental treatment applied) will also influence coccolith PIC. In the example of Valença et al. (2024), the Southern Ocean strains investigated had larger mean coccolith (distal shield) length (approx. 3.25 to 3.5 µm) compared to the mean coccolith length of the strains used in our experiments (2.54 to 3.14 µm). Secondly, the choice of shape factor influences morphometric-based estimates of coccolith and cell PIC. Here we use strain-specific and salinity-treatment-specific shapes factors, which may not be the case in other studies and could introduce some discrepancies, especially as the recommended shape factor for E. huxleyi stated by Young and Ziveri (2000) is greater (0.02 for morphotype A) than the values that we have used derived specifically for strains PLYB11 (0.014-0.015) and RCC1232 (0.017-0.019), which are also classified as morphotype A. Thirdly, our estimates of cell PIC are truly cellular in that they are calculated using the number of coccoliths on each individual coccosphere, which is variable within a culture. As assay-based quantification of cell PIC averages all the PIC in the culture across the number of cells filtered, it is more likely to overestimate cell PIC by also quantifying PIC in loose coccoliths and/or dead cells. These factors may contribute to differences between our estimated coccolith and cell PIC compared to other studies and between-publication differences in estimated PIC more generally. We will add to the revised manuscript a comparison between our data and those of previously published studies and discuss the possible sources of some of these differences, as described above.

7. Thank you for your positive feedback on the importance of this area of research, especially for those using or aiming to use desynchronised cultures.

Response to specific comments:

1. “ca.” will be changed to “approx.”.

2. The reviewer raises interesting points about the use of “exoskeleton” to describe the coccosphere as relates to exoskeletons in other organisms and a primarily protective function. Although there is no consistently used or accepted “plain language” description for the coccosphere, we agree with the points raised by the reviewer and will reword the two instances that “exoskeleton” appears in the manuscript to remove “exoskeleton” from the description of the coccosphere.

3. We agree that the definition of calcite and particulate inorganic carbon are not interchangeable and that by implying that they are by using several terms in the manuscript, this may confuse or mislead readers. We will therefore ensure consistency through the revised manuscript and use coccolith PIC or cellular PIC exclusively.

4. This is indeed a typo and Line 322 should refer to Fig. 1 (not Fig. 2). This will be corrected in the revised manuscript.

5. This will be corrected in the revised manuscript.

6. This will be corrected in the revised manuscript.

References:

Gebühr, C., Sheward, R. M., Herrle, J. O., and Bollmann, J.: Strain-specific morphological response of the dominant calcifying phytoplankton species Emiliania huxleyi to salinity change, PLoS One, 16, 1–24, https://doi.org/10.1371/journal.pone.0246745, 2021.

Gerecht, A. C., Šupraha, L., Edvardsen, B., Langer, G., and Henderiks, J.: Phosphorus availability modifies carbon production in Coccolithus pelagicus (Haptophyta), J Exp Mar Biol Ecol, 472, 24–31, https://doi.org/10.1016/j.jembe.2015.06.019, 2015.

Linge Johnsen, S. A., Bollmann, J., Gebuehr, C., and Herrle, J. O.: Relationship between coccolith length and thickness in the coccolithophore species Emiliania huxleyi and Gephyrocapsa oceanica, PLoS One, 14, 1–23, https://doi.org/10.1371/journal.pone.0220725, 2019.

Müller, M. N., Beaufort, L., Bernard, O., Pedrotti, M. L., Talec, A., and Sciandra, A.: Influence of CO₂ and nitrogen limitation on the coccolith volume of Emiliania huxleyi (Haptophyta), Biogeosciences, 9, 4155–4167, https://doi.org/10.5194/bg-9-4155-2012, 2012.

Valença, C. R., Beaufort, L., Hallegraeff, G. M., and Müller, M. N.: Technical note : A comparison of methods for estimating coccolith mass, Biogeosciences, 21, 1601–1611, https://doi.org/10.5194/bg-21-1601-2024, 2024.

Young, J. R. and Ziveri, P.: Calculation of coccolith volume and it use in calibration of carbonate flux estimates, Deep Sea Research Part II: Topical Studies in Oceanography, 47, 1679–1700, https://doi.org/10.1016/S0967-0645(00)00003-5, 2000.

We are grateful to Reviewer 2 (Andres Rigual-Hernandez) for their positive feedback on the manuscript and their constructive suggestions for areas for improvement, which we address in the responses below:

1. Authors report the use of two different strains: PLYB11 from the coastline near Bergen (Norway) and RCC1232 from the NW Mediterranean Sea. Is there a particular reason why these strains were selected? I assume authors selected strains from different environments (Arctic and temperate regions). If so, I would suggest to clearly state it in the text.

Response: The two strains were selected from a range of strains from different salinity environments that we were maintaining in culture as they represented isolates from distinctly different salinity environments that represented two “end-member” marine salinity conditions. This reasoning will be added to the methods section in the revised manuscript.

2. I find the data produced in this work very useful for the scientific community working on the interpretation of changes of coccolith morphometrics in relationship with environmental variability. However, as often different strains exhibit differing response to a changing environmental parameter it is important that the reader is able to associate the cultured strains to specific E. huxleyi morphotypes (which are the forms they find in the field). For this reason, it would be useful for the reader that authors provide a description of the morphology of the coccoliths of each strain and also that they provide some SEM pictures (either in the main text or as supplementary materials). A general description of the main existing E. huxleyi morphotypes could be found here: (https://www.mikrotax.org/Nannotax3/index.php?taxon=Emiliania%20huxleyi&module=Coccolithophores).

Response: We agree that different strains can exhibit variable morphology (independent of any physiological or environmental influence) and that this is therefore important context for examining morphological changes under an environmental treatment. A detailed description of the differences in morphology between these two strains has already been presented in Gebühr et al. (2021), with illustrative images, but we agree it would be useful to provide a brief summary of the morphological characteristics (including differences in range of coccolith and coccosphere sizes) of each strain within the revised manuscript and will add this to the methods description of the two strains and provide it as context for discussing size changes in the discussion. We will also direct readers to Gebühr et al. (2021) for a more detailed description and for SEM images of the two strains showing these morphological features.

3. Authors mentioned in lines 344-345 that “Cellular PIC could not be estimated for RCC1232 under salinity conditions as coccospheres were too poorly preserved”. Providing some images of this poor preservation would be also helpful for those researchers working on the field. Perhaps, similar forms have been documented in the natural environment.

Response: We will add a plate of illustrative SEM images of poor preservation (e.g. collapsed coccospheres that prevented the measurement of coccosphere size) under salinity 25 conditions as a supplement to the revised manuscript.

4. Could authors explain a little more about how Ks values were estimated for the strains in the publications mentioned in lines 165-167 (just a couple of lines) and explain with value of the range provided for each strain was used in the present study? Line 350, heading of table 1. Could you please revise the symbol before sd? Do you mean + -?

Response: We will add a brief description of the method used by Linge Johnsen et al. (2019) to derive the shape factors that we used. Thank you for highlighting that it is not clear how the range of Ks values stated in the manuscript have been applied, this will be clarified in the text of the revised manuscript. We apologise for the error in the +/- symbol used in the table caption and elsewhere that resulted during pdf conversion, this will be corrected in the revised manuscript.

5. Lastly, it would be useful for the readers that authors mention somewhere in the text what the expected changes in salinity are for the global ocean and if the range of salinity values tested in this study fall within the expected values of salinity change.

Response: Current projections for changes in global ocean salinity will be added to the introduction (see also response to reviewer 1) and we agree that this is useful context for the motivation of the study. We will also add to the methods how the choice of salinity treatment conditions (25 and 45) relates to the range of present-day and projected future ocean salinity conditions for additional context.

References:

Gebühr, C., Sheward, R. M., Herrle, J. O., and Bollmann, J.: Strain-specific morphological response of the dominant calcifying phytoplankton species Emiliania huxleyi to salinity change, PLoS One, 16, 1–24, https://doi.org/10.1371/journal.pone.0246745, 2021.

Linge Johnsen, S. A., Bollmann, J., Gebuehr, C., and Herrle, J. O.: Relationship between coccolith length and thickness in the coccolithophore species Emiliania huxleyi and Gephyrocapsa oceanica, PLoS One, 14, 1–23, https://doi.org/10.1371/journal.pone.0220725, 2019.