the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 May 2026
Seen in flakes & drops: local heterogeneity of marine snow, biogeochemistry, and plankton in Svalbard's glacial bays
Abstract. Glacier retreat is one of the most spectacular effects of climate warming in the Svalbard Archipelago, and is having a substantial impact on physical, biogeochemical and ecological processes in marine ecosystems. Here we propose that marine snow acts as a direct agent linking water chemistry, mineral particles, and primary and secondary producers. Marine snow plays a key role in carbon export and also carries knowledge about the biophysical state of the pelagic ecosystems. We propose that structuring of marine snow, assessed via high-resolution imaging by the Underwater Vision Profiler (UVP6), may serve as an informative currency for inter-regional, inter-seasonal and fine-scale system comparisons. In this study, we investigated small-scale variability in nutrients, carbon, plankton and marine snow in fjord waters proximal to glaciers in Hornsund, Rechercherfjorden and Isfjorden. We observed strong seasonal differences in biogeochemical properties, shifts in plankton composition (protists and zooplankton) and marine snow morphology between late summers (2022, 2023) and spring 2023. Fine-scale observations show various interplays between marine snow populations with hydrography, turbidity, biogeochemistry, and composition of plankton communities. Alongside expected relations between marine snow composition and morphology with for example temperature and macronutrients, we also identified other potentially important correlations, including with manganese, iron, and total alkalinity. Given the high spatiotemporal variability in physical, biogeochemical conditions and associated particles and plankton, we conclude that seasons and local heterogeneity are the strongest drivers. These results provide new interdisciplinary insight into coupled physical-biogeochemical-ecological processes and improve our understanding of land-ocean interaction on marine production and carbon burial near glaciers at varying states of recession.
- Preprint
(4802 KB) - Metadata XML
-
Supplement
(1425 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-2515', Anonymous Referee #1, 24 Jun 2026
-
RC2: 'Comment on egusphere-2026-2515', Anonymous Referee #2, 28 Jun 2026
This manuscript presents an extensive and valuable dataset from glacially influenced Svalbard fjords spanning multiple seasons, years, fjords, and glacier settings. The dataset has considerable potential to provide new insights into the interactions between physical forcing, biogeochemistry, plankton communities, and marine snow dynamics. However, I found that the current version does not yet fully capitalise on the richness of the dataset and the main scientific message remains difficult to identify.
One of my main concerns is that the biogeochemical dataset is not analysed or interpreted in sufficient depth to support the discussion. The study includes numerous informative variables, including nutrients, carbon, suspended particles, and stable isotopes, yet these are not suitably quantitatively interpreted. For example, Figure 4 presents cruise-averaged values that combine different fjords, glacier types, degrees of glacial influence, and depths. This aggregation makes it difficult to distinguish whether the observed physical and biogeochemical differences reflect seasonal variability, spatial gradients, glacier influence, or depth-dependent processes.
Similarly, several potentially informative variables, particularly δ¹³C and δ¹⁵N, receive no attention in the discussion despite their potential to provide insight into sources, biological processing, and carbon and nitrogen cycling. Throughout the manuscript, the discussion focuses primarily on the direction of change (for example, increases or decreases indicated by arrows) with no clear supporting evidence for this (maybe this is covered by the many supplementary figures, but this would need to be introduced more clearly) and no discussion on what the measured values themselves imply about the underlying biogeochemical processes is included.
Related to this, several interpretations would benefit from stronger quantitative support. Statements regarding nutrient depletion, nutrient supply, productivity, or glacier-driven differences should incorporate analyses of vertical structure, nutrient stoichiometry, isotope signatures, or differences among fjords and glacier settings. Nutrient limitation is mentioned briefly, but the manuscript does not explore whether nutrient ratios or isotope data support this interpretation. Making fuller use of the available biogeochemical dataset would greatly strengthen the mechanistic interpretation and better exploit what is potentially a valuable observational dataset.
A second major issue concerns the role of glacier type. The introduction places considerable emphasis on the contrasting influences of marine-terminating and land-terminating glaciers, and Section 4.3 argues that these differences are important. However, the conclusions ultimately state that no consistent patterns emerge among glacier types. This may indeed be an important finding, but the reasoning leading to this conclusion is not entirely clear. Because many of the analyses combine fjords, seasons, depths, and varying degrees of glacial influence, it is difficult to determine whether glacier-type effects are genuinely absent or whether they cannot be isolated within the current presentation of the data. Clarifying this distinction would strengthen one of the central themes of the manuscript.
More generally, the conclusions remain rather broad. Statements such as "local heterogeneity is prominent", "marine aggregates are valuable indicators", and "the interdisciplinary approach represents a step towards holistic understanding" are reasonable, but they do not clearly communicate the principal scientific advances arising from this study. Similarly, some parts of the discussion remain too general. For example, the opening paragraph discusses broad concepts but does not clearly establish the specific scientific question (what is the “paradigm” that you refer to?) or conceptual framework that motivates the remainder of the discussion. Tightening these sections would strengthen the overall narrative.
The figures would also benefit from further refinement. Many of them are visually complex, and the legends do not provide sufficient guidance for interpretation. Figure 6 is one example, where comparisons among glacier types, lagoon systems, and seasons are combined into a single schematic. It would be helpful if the legend more clearly explained the purpose of each panel, how the arrows were determined and what constitutes evidence for the reported trends. At present, the arrows appear quantitative, but the basis for assigning them is not sufficiently explained.
I also found that the discussion could be better integrated with the figures. In several places, broad interpretations are presented without clear reference to the supporting data, while elsewhere figure citations do not appear to fully support the accompanying statements. For example, the statement that late summer cruises generally experienced warmer, fresher, more turbid, less oxygenated, and nutrient-depleted conditions is attributed to Figure 3, although the supporting information appears to be distributed across multiple figures. More specific references to the relevant panels would improve readability and make the discussion easier to follow.
The overall structure of the discussion should be improved. I found that Section 4.2, which establishes the seasonal environmental context, would provide a more natural introduction to the discussion if presented before the more interpretive section on marine snow. At present, several themes are revisited multiple times, resulting in some repetition. Reorganising the discussion to first establish the seasonal and hydrographic setting, then examine glacier- and fjord-related differences, before finally interpreting the implications for marine snow, would produce a more coherent narrative.
There are also a few instances where the interpretation appears stronger than the evidence presented. For example, the manuscript suggests that May represented pre-bloom conditions with relatively low phytoplankton abundance, yet shortly afterwards states that copepod recruitment "must have been triggered by phytoplankton activity". This argument may ultimately be correct, but additional explanation would be required to reconcile these statements. More generally, I encourage the authors to adopt more cautious wording throughout the manuscript, replacing terms such as "shows", "clearly", or "must have" with expressions such as "suggests", "is consistent with", or "may indicate" where the evidence is indirect. In addition, the writing would benefit from a more scientific and concise style in places. There are several instances where the language becomes somewhat philosophical or metaphorical (e.g. referring to marine snow as "vehicles of matter and knowledge" or the title Seen in flakes & drops). While these expressions are engaging, they occasionally distract from the scientific narrative. A more direct scientific style would improve clarity and strengthen the overall presentation.
Finally, I believe the manuscript would benefit from careful editing for clarity and structure. The writing is generally understandable but often repetitive, and several paragraphs mix multiple ideas without a clearly defined focus. Improving the organisation of individual sections would help the reader follow the scientific argument and better appreciate the strengths of this substantial dataset.
Overall, I believe this dataset has considerable value and the study has the potential to make a useful contribution. However, in its current form, the manuscript does not yet fully exploit the richness of the observations or develop a sufficiently focused scientific narrative. I therefore recommend major revision, with particular attention to:
- Making fuller quantitative use of the biogeochemical dataset, including stable isotopes and nutrient stoichiometry.
- Clarifying the relative influence of season, fjord, depth, distance from meltwater input, and glacier type.
- Improving figure presentation and expanding figure legends to facilitate understanding,
- Strengthening the links between the discussion and the supporting figures.
- Revising the discussion and conclusions to emphasise a smaller number of clear, well-supported scientific messages throughout the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-2515-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 188 | 57 | 17 | 262 | 33 | 13 | 11 |
- HTML: 188
- PDF: 57
- XML: 17
- Total: 262
- Supplement: 33
- BibTeX: 13
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Overall, the manuscript presents an impressive and very detailed dataset of transects in various fjord branches with very different biological and biogeochemical characteristics. The focus on different glacial bays and the large variety of data provided are very valuable.
However, the manuscript appears somewhat unfocused, overly lengthy, and too descriptive. I strongly suggest a more focused, hypothesis-driven approach, utilizing only the data and analyses necessary to test the hypotheses, and putting some of the plots in the supplement. The results and discussion sections are currently very long and dense, making the narrative difficult to follow, and some plots are challenging to interpret. The station examples appear very descriptive.
The hypothesis of extreme spatiotemporal variability on very small scales is compelling and could be crystallized further. Right now, the authors seem to attempt to include everything, including small-scale examples, which makes it difficult to maintain an overview. The Redundancy Analysis (RDA) appears to be the core of the paper and its most useful component; it might be worth centering the manuscript around these findings.
However, the presentation of the figures and methods for the RDA is not entirely clear. The statistical methods could be more detailed, specifically by clarifying which are the response and which are the explanatory variables. It also seems that the RDA is overparameterized, contradicting the methods section which mentions that only a subset of drivers was selected. Many arrows lack labels, and it is unclear how the most influential variables were selected or defined.
Furthermore, I am not convinced that it is possible to discuss "seasonality" based on only two time points in spring (I am not convinced it is a pre-bloom state) and summer. It might be more accurate to describe this as "temporal variability."
The English language would benefit from careful proofreading, as there are several awkward formulations and typos. I provide some examples below.
Additionally, the references in the introduction and discussion are often long, exhaustive lists. I suggest focusing on key, selected references rather than citing every study that has performed similar work. Finally, there are several formatting issues: taxa should be italicized, completed, and corrected (e.g., Cirripedia was spelled incorrectly). Subscripts and superscripts must be used consistently, and figure text needs to be large enough to be legible.
Specific Comments:
L32: The statement that no two fjords have the same drivers seems oversimplified; some drivers are likely universal (e.g., temperature).
L60: Do some of these autonomous vehicles take physical samples as well?
L65: There are numerous seasonal studies, especially on hydrography and phytoplankton. I disagree with the assertion that this is a gap; perhaps specify if you are referring specifically to seasonal marine snow studies?
L77: Typo: "faecel" should be "fecal."
L100f: This is a very long and awkward sentence with grammatical issues, an example of where professional proofreading would improve the manuscript.
Fig 1: Great figure. However, why were no reference stations sampled outside the fjord bays? This might be worth discussing or justifying.
L117: More details about the glass bottles would be helpful. Were they acid-washed? Did you use a crimper to ensure they were gas-free?
L119: Should "AT" be "TA" (Total Alkalinity)?
L188: Did you use a binocular or a stereo microscope?
L236ff: As noted above, I am not sure it is possible to discuss seasonality based on two time points in late winter (post-spring bloom) and summer. "Temporal variability" would be more accurate.
L245: Typo: "Cyrripedia" should Cirripedia. I suggest double-checking all taxonomic names in the manuscript for spelling and ensuring they are in italics.
L257: Awkward phrasing. It is unclear which year this "low O2 high Chl" observation refers to.
Fig 3C: Please use full, italicized taxonomic names and verify their accuracy (e.g.,Cirripedia).
Fig 3: Some labels are too small to read. Units need to be in superscript (e.g., ind m-2)
Fig 4: Typically, units should use superscripts.
Fig 5: See comments for Fig 3.
L345: The "2" inO2 needs to be in subscript.
Ch 3.3 and Fig 8: This seems to be the core of the paper. However, it remains unclear. Please detail the statistics in the methods (response vs. explanatory variables). The RDA appears overparameterized. Many arrows lack labels, and the criteria for selecting the most influential ones are unclear. Fig 8c is missing an x-axis label, making it difficult to interpret.
L435: Is "paradigm" the right term here?
L445: Please be careful discussing seasonality with only two time points; consider using "temporal variability" instead.
L454: Would the decaying phytoplankton help explain the seasonal timing? It appears you may have sampled a post-bloom scenario.
L465f: It is interesting to mention TEP, but since you did not measure it, perhaps frame this as an outlook for future research?
L490: I am not convinced by your interpretation. Low silicate and high Cirripedia abundances point to a post-bloom scenario. In a pre-bloom stage, nutrients should not be depleted and zooplankton should not be abundant. Perhaps, justify with a key reference if you disagree.
L569ff: This section is largely repetitive of the previous discussion.