The manuscript presents multi-factorial experiments to examine the potential impacts of ocean warming and Fe availability under future climate change scenarios in two different regions of the Southern Ocean. The manuscript demonstrates that warming does have an impact on phytoplankton growth and community structure, but that under many cases these impacts are less than the effect of Fe under current temperatures. The results raise discussion points for how additional factors, such as Manganese and light, should be studied in combination to provide a better understanding in future studies.

Overall the manuscript is very well put together, with my only main concern being that statistical tests between treatments focused primarily on bulk concentrations and not on the differences in rates (i.e. chlorophyll derived growth rates, nutrient drawdown rates etc).

This study was also quite unique in that it reports not only changes in dissolved trace metal concentrations in the treatments but also changes in particulate trace metal concentrations. However, I could not find any reporting on contributions of lithogenic fractions from the initial starting water which may impact some of the particulate results.

There are some more general comments about areas that require some greater clarity or further expansion in the PDF.

This study uses bioassay experiments in the Weddell Sea and in the Amundsen Sea Polynya to study the effects of increasing Fe and temperature conditions on natural phytoplankton communities. Given the current predictions about upcoming changes in the Southern Ocean, it is critical to understand what effects these changes may have on natural communities. The manuscript is very well written, and this study is greatly strengthened by the extent of trace metal results reported (including both dissolved and particulate data).

However, several points could be improved. First, the differences in initial conditions could be discussed in more details (difference between A1-A2 locations and same for W1-W2, difference in macronutrients, etc.). Further, the photophysiological results (Fv/Fm and more?) should be referred to more often to strengthen some statements. You could also add more information on the biological results that were briefly mentioned (e.g., Chl:C ratios) to confirm your hypothesis. The figures could also be improved with bigger panels and the statistically significant differences displayed to help result visualization and interpretation. Finally, some typos need fixing throughout the manuscript (e.g., use of abbreviations and then not, some references to fix, some commas near period).

I have provided additional comments in the PDF. 

­General comments:

A comprehensive examination of iron and temperature modulation of natural phytoplankton communities was conducted through two sets of bioassay experiments in two different regions of the Southern Ocean, namely the Amundsen Sea and the Weddell Sea, over the summer period. The experimental design was well-planned and executed with consideration of future climate change predictions. The resulting data and supporting information are structured and comprehensively presented in the manuscript. The future recommendations further enhance the relevance of this research and the propensity for continued investigations into related fields by highlighting several knowledge gaps.

Detailed and technical comments:

I have outlined some suggested changes to consider for each section, as well as some minor technical corrections. My main suggestion would be that some statements in the discussion could benefit from including more recent publications where relevant to confirm or explain the results. Additionally, the conclusion could further be finessed to enhance clarity and impact.

Introduction

The introduction adequately highlights the knowledge gaps and thus provides the rationale for the study on temperature-enhanced effects of iron on the natural phytoplankton community in the different Southern Ocean regions. However, the results and discussion additionally introduce other parameters which could have been briefly included as part of the literature in the introduction.

Line 60: The references listed for ‘phytoplankton growth often becomes limited by low iron (Fe) availability’ are some examples of the many existing publications. Rather indicate that these are some references (‘e.g.’) to the magnitude of studies that have indeed established that phytoplankton growth becomes limited by low Fe in different ocean basins.

Line 62: Consider adding ‘Milligan and Harrison, 2000’ for the ‘nitrate assimilation’ reference (https://onlinelibrary.wiley.com/doi/full/10.1046/j.1529-8817.2000.99013.x).                                                                                                     

Line 70: Another reference to consider for ‘Fe supply by increased wind-driven mixing’ is ‘Moreau et al. 2023’ (https://www.nature.com/articles/s41467-023-36992-1).

Line 71: Typo ‘L.’ in the citation: ‘L. Seyitmuhammedov et al., 2022’.

Line 92: Could you expand a little on the ‘Fe from a variety of sources’ apart from the seafloor?

Materials and Methods

The materials and methods section was succinct. The section on the setup verification is much appreciated to remove any doubt of contamination issues, particularly for incubations performed while out at sea.

Figure 1: I would have appreciated seeing some information on the hydrography of the sampling sites, or even the Chla distribution in the map (separately). However, I understand that it is not so trivial, given the different sampling timelines.

Line 141: What was the average PAR under the ‘dimmed light conditions?

Line 147: ’28 December 2018 to 5 January 2019’

Line 153: Table 1:

• In the methods you refer to ‘silicic acid’. Ensure it is clear that silicic acid is indeed the reported ‘silicate’. If not, make this clear.
• It should be clearer if ‘Fe’ refers to dissolved Fe (dFe) only.
• Chl a (italicize a)

Line 204: Could you please clarify what threshold/range is ‘consistently low’?

Line 298: Section 2.8: Limited information on the instrument operation for the photophysiological data acquisition is given. Was any form of post-processing conducted on the raw Fv/Fm data? Or was this not necessary based on the data acquisition from the PAM?

Have you considered investigating the effective absorption cross-section (σPSII) from the photophysiological results? Could it further support the outcomes and contextualize the results in terms of stress on the photosystem or help estimate the primary production rates?

Line 305:  Perhaps you could reference ‘Cullen and Davis, 2003’ for the choice of 0.2 µm filtered blank corrections (Cullen JJ, Davis RF (2003) The blank can make a big difference in oceanographic measurements. Limnol Oceanogr Bull 12:29–35)?

Results:

In the introduction, you highlighted that ‘Generally, the Weddell Sea has a relatively low primary productivity’, while ‘the west Amundsen Sea and specifically the Amundsen Sea Polynya (ASP) is known as one of the most productive regions in the Southern Ocean in terms of net primary production per net area’.

Based on these statements, it would be nice to see a brief contrast of the initial conditions, as well as the results obtained in these two areas. This may also be a precursor to a concluding statement about the differences in the temperature-enhanced effects of the phytoplankton from these two regions.

In the abstract, I missed specific outcomes that are expected to be different for these two regions.

Line 483 and elsewhere: When referring to significant differences in Chla concentrations from bioassays, the Chla concentrations are a useful baseline. However, Chla growth rates are nuanced for assessing significant differences in phytoplankton responses from the bioassay experiments. Thus, it might be useful to reassess the significant differences in Chla by evaluating their growth rates instead. You already mentioned elsewhere Chla growth rates, but I did not easily find how this was calculated.

Line 420: Figure 2: Typo: ‘Weddell Sea (W1: c, e, i; W2: d, f, j)’?

The red and purple lines blend too well with each other. Consider a darker shade of purple or another colour. Apply comments to other similar figures.

Line 423: missing ‘)’ after ‘day 3’?

Line 451: Figure 3: Typo: ‘Amundsen Sea A1 (a, f, i, l, o)?

Line 470: Figure 4: ‘(a), (b), (c) and (d)’ – change to lowercase to be consistent.

Line 109-110: Since short-term local temperature increases can be expected in the Weddell Sea, can you comment on the short-term temperature increase effects from your bioassays in W1 and W2? Would using lower incubation temperatures for this region be justifiable instead?

Line 399: I missed how the nutrient drawdown was calculated. It would be nice to see a summary table or figure for the nutrient drawdown and the Chla drawdown, respectively. It is confusing to follow the results otherwise.

Line 406: ‘Silicate acid’ or just ‘silicate’? Check consistency in the use of terms throughout the manuscript.

Line 551: potential typo: ‘(Fig. 6d,h, p<0.01 for all).’ And (F and TF, Fig. 6h, p < 0.01)?

Discussion:

Line 608: ‘Dissolved Mn is known to (co-)limit Southern Ocean phytoplankton growth and community composition (Balaguer et al., 2022).’

Mn is known to (co-)limit together with? Are these co-limitations necessarily seasonal?

Consider: Pausch, et al. 2019. (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0221959)

Latour, et al. 2023. (https://online.ucpress.edu/elementa/article/11/1/00022/197210/Seasonality-of-phytoplankton-growth-limitation-by)

Line 635: Typo: ‘.,’

Line 637: Is this increased Mn uptake consistent with the needs of the phytoplankton community found at A1?

Line 664: The average Chl a:POC ratio over all treatments for the Weddell Sea bioassays were 0.003 ± 0.003. Could you comment as to why the Chl a content is very low relative to the POC? Is this because of a significant variability in this ratio across different treatments?

Line 673: The bioassays conducted by Viljoen et al. (2018) were in the Weddell Sea, while Alderkamp et al. (2019) conducted their bioassays in the Ross Sea. The bioassays presented in this manuscript were conducted in both the Amundsen Sea and Weddell Sea. The sampling season and location of W1 coincided with bioassay ‘S54–65’ by Viljoen et al. (2018). However, no comparisons seem to have been made or conclusions drawn regarding the outcomes based on similar and variable initial conditions to the overall outcomes from the bioassays.

Instead, only the ‘low light conditions’ where the light intensities differed due to Sea regions between this manuscript and Viljoen et al. (2018) as well as in Alderkamp et al. (2019) were highlighted.

Line 680: The Fv/Fm results are minimally discussed, and do not provide much insight into the changes in the phytoplankton health together with both the temperature and iron changes and the confounding influence on the changing communities.

Conclusion:

I struggle to clearly see all the concluding points made from this study, particularly in the latter half of the conclusion. The conclusion partly reads like a literature review and does not sufficiently highlight some of the main conclusions. This undermines the value based on the breadth of the experiments and outcomes of this study. I would suggest focusing on synthesizing key findings and clearly articulating the study’s contributions and implications in the conclusion.

Line 753: The reference to Brookes and Crowe (2019) appears in the conclusion with the statement that dual treatments may affect the responses. However, this referenced statement does not seem to appear among the discussion points or even as an inference.

Lines 754-759: These lines read as introduction sentences, rather than providing a strong context for the conclusion from the study: i.e. enhanced Fe input in such regions may partly overturn the warming-induced shift, given that macronutrients will not become limited.

Consider revising as this is not a compelling conclusion of the study.

Line 764: ‘only will the flow of organic carbon through the food web be affected,’

Lines 767-773: Again, I feel that these literature points can be better contextualized to your actual results.

Data availability: It seems that one needs to have an account with the NIOZ dataverse to access the data presented in the manuscript. Will this be publicly accessible later on?

Supplementary Information:

Generally, the panel sizes of all figures in the manuscript and supplementary could benefit from being slightly larger, so as to better see trends and the differences between the treatments or days.

Supplement Figure S1: This gives a nice overview of the physical setup. However, I struggled to fully comprehend and follow the details presented in the ‘Bioassay set-up’ in the supplementary text.

Supplement Figure S2: 3 Typos: ‘...the Amundsen Sea A1 (a, e, i), A2 (b, f, j) and the Weddell Sea…’

‘The black dotted line represents the control (C) treatment, the red solid line the temperature (T) treatment, the blue solid line the iron (F) treatment, and the purple solid line the combined temperature and iron (TF) treatment.’

It is not very clear to see the ‘black dotted line’ from the panels without having to zoom in significantly. Are these supposed to be ‘black solid line’? Perhaps you could change the scale/size of the y-axis to facilitate larger panels.

The purple line seems to blend too well with the red line. Consider using another contrasting colour like green or darker purple?

‘Averages of triplicates with error bars represent the standard deviation’.

Supplement Figure S3: Again here, is the ‘black dotted line’ maybe meant to be ‘black solid line’? Same comment regarding the purple and red solid lines blending.