the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Jun 2026
From surface flux to seafloor function: vertical carbon export and benthic communities in high-latitude coastal systems
Abstract. The downward flux of carbon from the surface ocean to the seafloor fuels benthic invertebrate biomass and drives carbon sequestration. In polar regions, pelago-benthic carbon coupling is shaped by complex processes influenced by climate change, including glacial retreat, water stratification, and warming. Here we examine three high-latitude coastal systems along the Northeast Greenland coast: Dove Bugt, the Brede–Ardencaple Fjord system, and Kong Oscar Fjord, each differing in bathymetry, glacier influence, and exposure to Atlantic Water at depth. Using short-term sediment-traps, hydrographic profiling, sedimentary carbon analyses, and trait-based macrobenthic community data, we assess how vertical particulate organic carbon (POC) export, sedimentary organic carbon stocks (SOC), and benthic biomass and function are linked across the three coastal systems. We found that vertical POC flux and SOC were highly variable in space and showed no consistent relationship with benthic invertebrate carbon biomass. Our trait analysis of the benthic communities revealed that only a subset of trait strategies, particularly only epifaunal suspension feeders, consistently translated carbon availability into standing biomass. Latent variable modelling, which identifies underlying ecological gradients not explained by measured environmental predictors, revealed an additional axis structuring benthic trait biomass and is most consistent with bathymetric relief and associated horizontal redistribution processes. Thus, with our results showing a temporal decoupling among POC export, SOC, and benthic carbon biomass, we propose that future research investigate lateral transport pathways associated with Atlantic Water intrusion and stratification, alongside benthic production and ecosystem functioning, to better understand carbon processing, retention, and sequestration in Arctic systems.
- Preprint
(2577 KB) - Metadata XML
-
Supplement
(2604 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-3268', Anonymous Referee #1, 09 Jul 2026
-
RC2: 'Comment on egusphere-2026-3268', Anonymous Referee #2, 23 Jul 2026
This trans-disciplinary study utilized an innovative approach combining hydrographic and ecological trait-based analyses to explore dynamics in carbon supply and stores within NE Greenland fjord systems. The study finds that benthic community structure and carbon pathways (inferred from traits) are influenced by both vertical POC flux, sedimentary carbon stores, as well as the bathymetric position which is assumed to be related to local advection and depositional dynamics. Overall the study, though complex, is very well-written and supported thoroughly by published literature. There are 4 clear hypotheses stated in the introduction, although only some aspects of the first two are evaluated directly and the manuscript really focuses on the latter two hypotheses. For instance, hypothesis two refers to "stratification" which is inferred through the water mass distributions namely the change from Surface and Polar Water masses to Atlantic Water. In addition, the penetration of Atlantic Water into the fjord systems is seen through the water mass distribution along the transects but is not the focus of the study and not discussed at length. Rather, the explanatory power of water mass structure in relation to benthic community and trait structure is the focus. Regardless, the study does address aspects of all hypotheses presented. Inferences made about the study results are well thought out and described in the discussion section albeit with some very complicated and long sentences at times. The study clearly leads to the final conclusion and additional perspectives on future work are given.
The figures are generally of very high quality and, although complex, illustrate well the findings of the study. Figures 4-6 are the most difficult to interpret but paired with the text I can follow for the main findings from them.
The manuscript would benefit from very few minor adjustments made (suggested below) to improve clarity in some sections and a few comments are given regarding interpretations of the data which can be considered by the authors. These are considered very minor and the manuscript overall warrants publication.
Consider specific edits/comments:
Line 16: fix "pelago-benthic coupling" to "pelagic-benthic coupling" for consistency with the rest of the manuscript.
Line 47: change "seafloor topology" to "seafloor topography" or "seafloor bathymetry" for consistency with the rest of the manuscript
Line 59: The sentence starting with "different benthic strategies" is a little unclear - consider changing to "Organisms employing different benthic feeding strategies" for clarity
In line 65-66 it is stated that communities may shift toward strategies adapted to lower carbon supply. What are these strategies? Consider giving 1-2 as example for contrast.
Line 113: abundance and biomass data were standardized to m^-2. This may also be clearer if box core is presented as 0.1 m2 rather than in cm2.
Section 2.2: Little discussion was given to the choice of a single wet-weight to carbon conversion factor despite significant literature regarding the variability in such conversion factors and published conversion factors which could have been employed. In addition, no detail is given as to treatment of individual groups (e.g., shelled organisms, tube-dwellers) to understand what the data truly represent or for repeatability. The manuscript lacks discussion or evaluation of the potential error associated with the wet-weight to carbon conversions which form the basis of the data used throughout the study. It is suggested that the authors include an evaluation of the potential error and discuss this in context of the study results. In addition, this section lacks a description of the identifications made i.e. to what taxonomic level which is important for the trait data. A statement in the following section (2.3) mentions species-level carbon biomass values which implies that all taxa were identified to species level. However, it seems unlikely that all species had available trait information in the Arctic Traits Database. It is suggested that the authors add methodological details to these sections (2.2 + 2.3) indicating to which taxonomic level taxa were identified to and how trait retrieval was done if traits were lacking in the database. It is acknowledge that such a statement may be missing if traits for all species were available - this is simply unclear in the manuscript.
Line 131: remove "additional"
Section 2.4.1: It is assumed samples were homogenized and the results then represent grain size of the upper 10 cm of sediment at each site. Clarify in the text. The grain size data refers to clay which is showin in Figure S3 but the caption states "mud" which appears throughout the rest of the manuscript and figures. Authors should change to either clay or mud for consistency throughout.
Lines 157-160 describe the process of estimating POC flux at the seafloor from attenuation coefficients and deepest trap depths. It is unclear from this description what the deepest trap depth represents because in Table S11ii all deepest trap depths are shallower than the seafloor depth or deepest seafloor depth except at station DB1. In addition, Figure S11i shows the attenuation coefficient regressions where there are trap measurements existing below the "deepest trap depth" line. It is unclear what the "ratio between seafloor depth and deepest trap depth" means.
The calculation of attenuation coefficients seems straightforward but I suggest the authors comment on the clear non-linearity of POC flux profiles particularly at AFC2 which results in the only location with a positive coefficient. I suggest this because the coefficient (much like the choice of wwt-carbon conversion factor) could greatly impact the POC flux estimates used in all subsequent analyses - thus, a brief discussion of potential sources and magnitudes of error is warranted.
Lines 214-215 state that the bottom water conditions are similar across the stations. This is true at this point in time - no seasonal dynamics can be explored from the data presented in this study. This statement should be amended to reflect this.
Line 265 states that the patterns observed from GLLVM analyses provide context for interpreting spatial differences in carbon sources but very limited data in this study relate to carbon sources and nothing definitive (e.g., biomarkers, in-depth chemical composition or lability). This should be changed to "carbon stocks" or something more appropriate for the data collected and presented in this study.
Figure 3: add a color scale or state what the ends of the scale indicate in the caption.
Section 3.4.1 highlights a difference between stations AFC1 and KOF1 in terms of functional trait composition but little attention was focused on the taxonomic compositional differences between stations. Since this is only presented very coarsely at phylum level in the supplement, not much can be said and this is not the focus of the study. However, even at phylum level it is clear that there is a taxonomic compositional difference between these stations which does not seem reflected in the trait composition which is interesting in the context of the overall study.
Lines 370, 374, and 422: use "carbon accumulation" but this study measured POC flux and sedimentary carbon stocks not true accumulation rates. Change terminology for clarity.
Lines 380-383: The novel observation of Atlantic Water in this fjord system should be confirmed with more information regarding the water mass structure and temporal variability of it including the potential for water mass transformations. Description of processes which could lead to this were well explained in the subsequent lines of the manuscript.
Given the limited temporal coverage of this study, it seems inappropriate to conclude here and on like 535 that AW is a consistent background condition of this region without exploring seasonality in hydrography here. If this has been studied previously to inform this statement, literature should be referenced. Otherwise the statement seems to overreach the data in the present study.
Line 421: The statement about processes weakening the link between pelagic and benthic realms is a bit confusing since fecal pellet production can enhance carbon flux to the seabed whereas remineralisation in the water column decreases it. Consider changing "weaken" to "alter" or more clearly explain these linkages.
Line 447: This line mentions long-lived and immobile Arctic taxa - is this a statement about Arctic taxa generally or related to the traits of fauna observed in this study? If there is evidence that the sampled communities here contained long-lived and immobile taxa this statement is directly relevant to the study. Otherwise, it is commenting more generally on Arctic benthic communities and this should be clarified.
Line 453 mentions carbon lability and although not a detailed analysis of it, this study did measure sedimentary C:N ratios. It seems that the low C:N at AFC3, AFC4, KOF3 which are open to shelf waters (or on the open shelf) may indicate advective organic carbon inputs of marine origin also supporting the higher SOC measured at these stations.
Citation: https://doi.org/10.5194/egusphere-2026-3268-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 79 | 34 | 6 | 119 | 15 | 7 | 4 |
- HTML: 79
- PDF: 34
- XML: 6
- Total: 119
- Supplement: 15
- BibTeX: 7
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript by Armitage et al., is well written and explores an interesting question in terms of how much vertical carbon export and sediment carbon stock affects benthic communities in high latitudes. A great amount of work and expertise has clearly gone into shaping this manuscript. It is an extensive piece of work trying to answer 4 different hypothesis, which makes it sometimes challenging to follow through on all the different strands the authors are aiming to tie together. There are a few aspects which might need some more attention, before the manuscript is ready for application.
I am not totally convinced by the oceanographic aspect of the paper and the finding that high arctic fjords with high sills are influenced by Atlantic water, which is normally characterised by temperatures > 2 degrees and high salinity. I could not find this in the supplementary material (which shows temperatures below 2 degrees) and the authors do not provide thresholds on what they defined as Atlantic water (in Figure 1 this seems to be primarily based on salinity). Based on the figures provided, I do not think bottom waters at 0-2 degrees are an indication for fjords being influenced by warm Atlantic water. This needs to better explained or possibly dropped.
When it comes to the estimation of carbon content from wet-mass, I am not convinced by the very generic conversion that has been employed here. Behrisch and Zwerschke (2026) have shown that a conversion from wet mass to dry or afdm is not always possible with specific taxa, and that all of these conversion factors are highly taxa dependent. Anemones for example may have a large amount of water stored in their wet-weight compared to bryozoans, which would result in an overestimation of their carbon content. Barnes and Sands (2017) conversion to carbon content was based on functional groups and afdw and not a blanket conversion for all taxa. This could very much bias the results when aiming to link carbon standing stock to sediment carbon and vertical carbon transport. Thus, I would urge to reconsider this aspect of the manuscript.
The other aspect is that vertical carbon transport is unlikely to be completely vertical in a system regulated by tides and currents. This means that surface carbon content may not be linked to sediment carbon stock, particularly in areas with high flux, which is not necessarily linked to benthic-pelagic coupling but dictated by the abiotic environment and it would be good to have this acknowledged or considered when calculating the vertical carbon flux.
One aspect that has been neglected in this paper is, that substratum availability, generally dictated by abiotic environments as well, is a great driver of benthic communities. To find large amounts of high biomass sessile suspension feeders, hard substratum tends to be required, which is found in the high energy environments. The authors touch upon this, by linking their LV gradient to bathymetry and steepness, but it would be nice to see a clearer acknowledgement of this.
Although the paper is very well written, there is a tendency to overcomplicate statements, which sometimes makes it hard to follow and figure out what it is the authors are trying to say. E.g a temporal decoupling of POC, SOC and benthic carbon biomass in the abstract, or greater redistribution of carbon across co-occurring feeding modes under advective conditions in the discussion. A simplification and greater precision of language here might help the reader to better understand the point that is being made. This is particularly evident in the discussion, where the authors are making bold statements, which are not necessarily underpinned by the data that is presented.
The figures are very nice and clear but I am not sure it is necessary to have the same figure three times. It makes it difficult to compare between the differences in traits across the latent variable scores. Maybe all the trait-biomass diagrams could be combined into one, based on their z score.
I would also recommend spelling out some of the abbreviations such as surface water or Atlantic water or as a minimum spell them out in a new paragraph/section again, so the reader is reminded what it means.
Some more specific comments
Line 49: modify in which way?
Line 54: Bathymetric highs?
Line 53-56: Check and consider Buydens 2026
Line 82- 86: sentence is being repeated
Line 130-132: was this analysed in different sections, e.g. surface and lower depth, if so how?
Line 154: When was it deeper
Line 165-170: Can you explain what the coeffcient means, for people not familiar with analysis? You speak of it being stable or high or moderate. Is that the intervariability between sites or stations etc etc
Line 180-183: Not clear
Line 242: I would not call this comparable
Line 251-2: Is this linked to primary production
Line 318: this biomass is greater than other stations, such as KOF3
Line 321: Why would the deep stations be more prone to disturbance from deposition and sediment resuspension
Discussion:
It is not always clear whether some of the statements are supported by a statistical analysis, as some of this factors such as particle size have ben measured but there is no indication about a formal analysis here
Line 428-430: unclear
Line 435-438: unclear
Line 455-457: Would this not be a seasonal pattern
Line 470 – 472: Would this not be the other way around
Line 472-475: this contradicts your previous statements
Line 485-486: exposure to what
Line 487-488: like what substratum type, depth, currents
Line 492-3: Well yes because there is more hard substratum, no?
Line 504: unclear what you mean with hydrographic scale
Line 512-514: This is very hard to interpret
Line 516-518: I am not sure I agree here and whether this is clearly supported in your data
Line 525: what processes
Line 525-530: But this is not what you did, no?
Line 535-537: I am not sure you can say that you have shown that
Line 539-541: But you did not measure that