This article by Bruyant and co-authors describes the data set collected by the Green Edge cruise team in Baffin Bay during late spring/early summer 2016. 

The data set itself is extensive and multipartite and a significant contribution to the available observations around phytoplankton blooms in the Arctic. Here, I am solely commenting on the article itself (and not the data). 

While the overall structure and writing of the article are adequate, I was somewhat taken aback by the substantial number of typographical and formatting errors and inconsistensies. This makes the reading a more arduous process than necessary and is particularly lamentable given the large number of co-authors who apparently did not read the text with care.

The collection, management, and archiving of such a large-scale and complex data set requires great care and attention to detail (as the authors emphasize in Section 7). To instill confidence in the data set I believe it paramount that the accompanying article is also crafted with care and attention to detail.

I note that the author contribution statement provides a detailed overview of data collection and analysis but does not discuss who actually wrote the article. 

I recommend publication after careful and thorough copy-editing and consideration of my comments below.

(Please note that this is not a comprehensive list of copy-editing issues).

General comments:

1. The main confusing part for me was the relation between this article, the ice camp article, and the data sets stored on SEANOE:

- In my understanding there was a perivous, separate ESSD article on the Green Edge ice camps: Massicotte et al 2019 (https://essd.copernicus.org/articles/12/151/2020)

- And a companion data set, entitled "The Green Edge initiative: understanding the processes controlling the under-ice Arctic phytoplankton spring bloom", also by Massicotte et al 2019 (https://www.seanoe.org/data/00487/59892/)

- Then there is the present article on the Green Edge cruise: Bruyant et al 2022 (https://essd.copernicus.org/preprints/essd-2022-41/)

- But what is the companion data set? The article on l.76 and l.483 refers to the data set above by Massicotte et al 2019 (https://www.seanoe.org/data/00487/59892/). Does that mean both articles describe different parts of the same data set?

- Or is the companion data set to this article the much earlier one by Bruyant et al 2016 "The Green Edge cruise: following the evolution of the Arctic phytoplankton spring bloom, from ice-covered to open waters" (https://www.seanoe.org/data/00752/86417/)? This would make sense, since the abstracts for the article and this early data set are basically identical. But the 2016 SEANOE entry is not mentioned anywhere in the ESSD article. 

Please clarify! It is striking that this article is about a data set and it is unclear what the data set is or where to find it. 

2. Relatedly, it would be helpful to get an overview of what papers have been published on what subsets of the data so far. Maybe add a table?

3. There are a couple of points where you reference data that were collected as part of this effort but that are stored elsewhere (not with the main SEANOE data set), e.g., l.137, l.315. It might be worth adding a table or similar to provide an overview of where all data is to be found? Or at least add those separately stort data links to the data availability statement. 

Some detailed comments:

l.72 misplaced comma

l.74 Canadian Arctic should be capitalized (capitalization issues throughout the article)

l.75 "data set" or "dataset" ? (be consistent throughout)

l.76 not just spring, maybe write "late spring/early summer"?

l.77 There is no Massicotte et al (2019a) in the list of references. Nor a Massicotte et al (2019), which is referenced numerous times later in the text. There is only Massicotte et al (2020), which is the article on the Green Edge ice camps. Do you mean to cite the Bruyant et al (2016) SEANOE data set here?

l.81 It seems a little odd to speak of a MIZ when there is no perennial ice pack (presumably marginal ice is at the "margin" of something). Maybe rephrase that it may all be seasonal ice in 20 years?

l.100 "Leg 1A" rather than "Leg1A" (space missing)

l.101 "Leg 1B" rather than "leg 1B"  (also, since there are only 2 legs, why not call them leg 1 and leg 2?)

l.126 "were" instead of "was"

l.156 "presented" rather than "represented"

l.163 "First, the raw data were screened to eliminate" revise to "First, the raw data were screened to identify and where possible eliminate" (?)

l.169 "e.g.," rather than "ex."

l.203 "Eicken et al, 2009" is not in the references. Instead you're citing the article by Ackley (2010) which reviews the Eicken et al (2009) book. Replace that entry with " H. Eicken, F., Gradinger, R., Salganek, M., Shirasawa, K., Perovich, D. and Lepparanta, M, eds. Field techniques for sea-ice research. University of Alaska Press, 2009."

l.209 "SBE 911plus" rather than "SBE-911Plus"

l.232 "in situ" rather than "In situ"

l.249 ")), sky" (space after "))" )

l.251 "the ocean optics protocols" Which ones? Should this read "the ocean optics protocols of Mueller et al (2003) and Mobley (1999)", or similar?

l.262 "Sea-Bird SBE 19plus"

l.268 and throughout: please write units using superscripts

l.295 Please revise this sentence.

l.309 "station G719" Can you comment on this nomenclature at some point prior in the text? I assume this is transect 700 station 19? What does the "G" mean?

l.317 "Leg 1B"

l.343 first use of "SCM" - define "subsurface chlorophyll maximum"

l.391 either use "(i), (ii), (iii)", or "i), ii), iii)" 

l.400 It's unclear here what is meant by "not possible"

l.414 just delete "Leg 1" ? The whole cruise consistent of Leg 1 (A and B), correct? And 10 June - 8 July spanned both A and B?

l.420 It is strange that this paragraph couches the discussion in terms of "articles in prepariation", something that is not done anywhere else in the article. I would suggest revising this.

l.427 define "DCM"

l.434 "St 409 and St 418" - this refers to transect 4 stations 9 and 18? Please make consistent with previous nomenclature

l.577 incomplete reference

l.578 incomplete reference

l.665 reference formatting issues

Fig.1 Resolution of font is poor. It would also be helpful here to indicate with shading which side of Baffin Bay is ice covered. I was surprised to learn that all floats survived the same number of days - could you mention why? It would help for panel B if you zoomed in on the survey region to show some more of the details. 

Fig.3 Blurry, figure labels and legends not readable

Fig.4 Also blurry, I would also recommend using a different colormap for the bathymetry to differentiate the plots. 

Fig.5 Would it make sense to have panel A span the time period of Leg 1A and panel B that of Leg 1B ?

Fig.7 I recommend making the colormap divergent (or otherwise have an abrupt change) around 0 so it is obvious which stations were located in open water and which in the ice

Fig.10 Blurry

Fig.13 Missing panel labels. Also, caption should be consistent in label formats - either (a),(b),(c), or A,B,C

This article presents impressive, really and definitely impressive project and consequently its datasets. Moreover it concerns a very important and valuable topic – retreating ice in the Arctic and its impact on the primary production and its fate, thus the consequences for the whole ecosystem. The number of measurements, parameters, devices, new technologies, as well as the resolution of sampling are remarkable. I am sure that plenty important articles may emerge from this dataset. Moreover it can support many other studies as a great reference. In general the dataset as well as the description were made with caution and mostly with sufficient details, only some very minor suggestions I can provide for better utilization by other users:

Title: actually this dataset includes much more information and parameters than just phytoplankton spring bloom. I would suggest something oscillating around the processes in the marginal ice zone. Also ‘spring’ is not that accurate for the June-July period

Abstract: I think you could add a little bit more information (e.g.; list of most important measurements, parameters performed, the number of stations etc.), while the name and the abbreviation of the ship are of lesser importance in my opinion. I think it is also important to mention that not only data from cruise-based stations are available, but also high resolution measurements of gliders, MVP, ship-track water monitoring, BGC-Argo floats etc..

L86: You can consider referring here to an article about the export of the marine snow produced by the phytoplankton bloom that was based on the GreenEdge dataset : Trudnowska E., Lacour L., Ardyna M., Rogge A., Irisson J-O., Waite A., Babin M., Stemmann L. (2021) Marine snow morphology illuminates the evolution of phytoplankton blooms and determines their subsequent vertical export. Nature Communications, 12, 2816. In general it would be great if you could gather all the existing articles that published results related to GreenEdge. However, I am aware that it is not a review, so up to you and editor to decide, if that remark is relevant.

L119: for understanding how such complicated dataset can be used to track successive crossing of the MIZ you just refer to Randelhoff et al., 2019, but I think one sentence explaining the idea of arranging stations according to OWD (open water days) would be very helpful for potential data users. Now I see that you did so in 4.1 section, good!

Table 1 – I think it is incomplete (e.g. no mention of the UVP, LISST, sediment traps etc.). It would be great if you could list all the equipment used. Moreover, ‘Hydrobios’ is just a company producing many equipment, please be specific what you mean by this. Potentially you could also consider adding an extra row about the high resolution measurements so as to include all the devices applied in one place. While I think that Table 3 is great (impressive) with the full list of parameters, Table 1 disappoints with scattered information. In my opinion it should not be that much devoted to the ‘operations’ but rather to equipment used, because only the cruise participants may know what stays behind a specific ‘operation’, while the second-hand users prefer to know what was done with which device.

L133-135 I think those sentences about Argo are not suitable for this article, especially to this part.

L152: This header is not strictly true for the content. You say nothing about data processing, but a lot about the unification of formats. Therefore I suggest to change it accordingly, e.g. Time-for-space formatting & data quality control

L172: In my opinion you mostly provide methodology than data description. Maybe you should put a header as e.g., Description of data collection.

L176 & others: Maybe you could consider giving numbers to the parameters included in Table 3, and then to refer to those numbers also when listing them ?

L389: what are those further analyses?

In general Figures are of good quality, just a few suggestions:

Fig. 6. What is the black line representing? Is that a ship route ?

Fig. 10. Please correct a figure caption to be more specific/informative, as “Relative proportions of the different taxa within the different groups of samples,” by this we do not now that those are mostly bacteria presented etc.

Fig. 15. It can supplement Fig. 2 as e.g., b) panel

Datasets: I was not able to check all of the files provided, but within the ones I investigated at https://www.seanoe.org/data/00487/59892/ I found some problems:

• Not only data from the Cruise are provided (as indicated in an article), but also data from the ice camps (2015, 206), e.g. CDOM absorption dataset, Bacterial biomarkers dataset, Camlum (Ed) dataset, Carbon stock (DOC/TDN) any many many other files
• Erogenous latitude (e.g. 200 Lat in CDOM absorption file)
• NAs – rows with NA content in such crucial cells as ‘Station’, ‘Depth’ should be deleted as are useless (e.g., Particulate and phytoplankton absorption dataset, Bacterial biomarkers dataset)
• Different data formatting within one file – at some point the number of columns and their content starts not to agree with the other rows - e.g., Bacterial biomarkers dataset; Days of open water (DOW); Pigments, nutrients, Chlorophyll a and Phaeopigments (concentration); Sediment traps (chla, faecal pellets flux, Particulate mass, etc.)
• Headers – it is very important to name them in an identical manner, e.g. in Table 4 you claim that each .csv includes the same metadata columns. Even if that is true, but is not always, then the naming should be exactly the same, e.g. you indicate in a Table “operation code”, while it occurs by different variations of names (e.g., “code_operation” in Bacterial biomarkers dataset, Nutrients etc.), which is the same for human, but makes a huge difference for coding and automatic merging various datasets.
• Zooplankton – while there are no taxonomical results provided based on net sampling, the UVP dataset (Underwater Vision Profiler (UVP) – zooplankton) includes also other than zooplankton categories (e.g., detritus, artefact)

I think you could supplement Table 3 with an additional column saying in which file this parameter can be found.

As far as I understand the files provided at https://www.seanoe.org/data/00487/59892/

are ‘ready to use’, but many parameters described in the paper are not provided there

Whereas the raw data is provided at http://www.obs-vlfr.fr/proof/php/GREENEDGE/x_datalist_1.php?xxop=greenedge&xxcamp=amundsen

But the access to this data is mostly restricted to the owners of the specific account?

My replies to the crucial questions of the review:

Is the article itself appropriate to support the publication of a data set? – Yes

Is the data set significant – unique, useful, and complete? – Unique, Useful but with some mess and problems with access to some types of data

Is the data set itself of high quality? – Yes

Is the data set publication, as submitted, of high quality?  – Yes