the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Jul 2026
Drivers of net community production in the bottom of Arctic pack ice
Abstract. The availability of nutrients to sea-ice microbial communities varies as drifting pack ice traverses distinct hydrographic regimes and experiences episodic mixing events. Vertical turbulent fluxes generated by processes such as ice movement and wind can enhance nutrient resupply at the ice–ocean interface, potentially influencing community composition and net community production. In this study, we evaluate the relationships between bottom sea-ice algal communities and oceanic conditions during a sea-ice drift study north of Svalbard, which crossed different branches of the Atlantic Water inflow to the Arctic over the Yermak Plateau and the Sofia Deep. Despite elevated nitrate concentrations in under-ice waters associated with the Svalbard Branch of Atlantic Water inflow, the largest nutrient and heat fluxes to the bottom ice occurred during a storm event. These fluxes coincided with increased biomass-specific net community productivity but also enhanced algal biomass loss through melt. At the same time, snow redistribution during the storm increased snow depths and reduced light availability at the bottom ice, affecting net community production. Independent of nutrient conditions, the transition from net heterotrophy (oxygen consumption) to net autotrophy (oxygen production) occurred when under-ice transmitted photosynthetically active radiation exceeded a daily range of 1.1 to 3.5 µmol m–2 s–1. Although nutrient supply remains important, our results suggest that light availability may exert a stronger control on sea-ice net community production than nutrient supply when nutrients are not critically depleted. We further highlight the importance of episodic storm events in modifying nutrient and heat vertical fluxes, and light availability at the ice–ocean interface, thereby contributing to regional variability in ice algal bloom dynamics.
- Preprint
(1610 KB) - Metadata XML
-
Supplement
(1593 KB) - BibTeX
- EndNote
Status: open (until 15 Oct 2026)
- AC1: 'Comment on egusphere-2026-4031 - MSS data', Rosalie McKay, 21 Jul 2026 reply
-
RC1: 'Comment on egusphere-2026-4031', Anonymous Referee #1, 17 Aug 2026
reply
The submitted manuscript with the title “Drivers of net community production in the bottom of Arctic pack ice” by McKay et al. provides detailed information on the sea-ice algal community found in the Central Arctic north of Svalbard and how its net community production rates (NCP) respond to changes in nutrient and light availability. This study was undertaken during a sea ice drift of a single floe encountering different water masses and a storm event during the main ice algal growth period in May. While this scenario of changes in habitat conditions is more frequently encountered by ice bottom communities, their impacts are still poorly understood, which makes this unique study an important knowledge contribution. The manuscript is overall well written and structured, presents results in a concise manner and discusses findings in detail. Overall, I suggest minor revisions that should address found inconsistencies in the format of plots, minor technical errors and requests to provide more details for certain method and result descriptions.
I also would like to ask the authors to further look into one larger concern I have with the discussion and drawn conclusions from the presented results that relate to the impact of light limitation on calculated NCP. The authors conclude that the storm event turned the overall NCP budget from net autotrophic to net heterotrophic due to a redistribution of average snow depths above the calculated snow thickness threshold that would allow for positive NCP rates. While this may be true short-term, the authors base this conclusion mainly on modelled, not measured results. Hence, I partly disagree with this conclusion for these reasons:
- Derived photosynthetic parameters from optode measurements (table 3) show a shade acclimation of second-year ice communities with higher alpha and lower compensation irradiance at higher snow depths (one above calculated 26 cm snow thickness threshold). This indicates a capacity of the present ice algae community to acclimate to higher snow depths after storm event to reach positive NCP.
- The possible impact of photoacclimation on NCP rates is only briefly discussed and only for the possibly high-light acclimated FYI community beneath lower snow thicknesses. I find this discussion very incomplete given that there are three more measurements.
- It’s quite striking that the highest NCP rate was found after the storm at the SYI site with snow depths near the threshold and despite a strong decrease in ice algal biomass. I would say this highlights how efficient the shade-acclimated community in the SYI is, especially if nutrient limitation doesn’t play a role. I would even argue that their previous conditioning to low light levels under high snow depth (see site 2, thick snow depth PE parameters) gives them an advantage to high-light acclimated FYI communities after the storm event. And as the authors point out, the floe mainly consisted of SYI, which should decrease the impact of a “struggling” FYI community. Furthermore, under-ice phytoplankton studies have shown that microalgal communities can acclimate to changing light conditions within a few days, which may further make the net heterotrophic state of the floe rather short-lived.
- The authors did not revisit a FYI side after the storm to confirm their conclusions of net heterotrophy. The sampled SYI seems to not support their conclusion. This is somewhat a limitation of this study, which should be acknowledged.
Overall, I think the relationship between NCP and ice algal photophysiology need to be discussed in more detail, similar to the detailed discussion of the impact of nutrient availability on NCP.
I further structured my detailed comments for the authors into two sections (“Specific comments” and “Technical comments”) as suggested by the journal review guidelines.
Specific comments:
Line 14: For a plain language summary, I suggest not using the term “physical forcing” and use something more explanatory such as “changes to habitat conditions”
Line 66: Please further define “topography” – Does it refer to “bathymetry” or “sea ice topography”?
Lines 90 and 91: I suggest clarification “autotrophic and heterotrophic processes WITHIN ICE”
Figure 1: Chlorophyll a concentration in Fig. 1b) should be provided in unit “mg m-2” to match text values. Please include explanation of abbreviation “MSS”. Spelling of first-year and second-year sea ice is incorrect.
Line 150: It would be helpful if it is made clear which of the MSS probes are used to estimate vertical flux later. Otherwise it is somewhat of a surprise in section 2.8.2. And when was the MSS instrument calibrated?
Section 2.5: How optical terms are started is somewhat convoluted and confusing for anyone who isn’t an expert in marine optics. Hence I would suggest some modifications:
- Line 175: Delete the term “irradiance” after radiation (PAR) as the term is now already defined as “photosynthetically active radiation” and doesn’t need the add-on “irradiance”.
- Throughout this section and following, just use “PAR” instead of “irradiance”, “irradiance withing the PAR spectrum”, except for line 176, where you also should add the actual PAR range: “surface downwelling and upwelling irradiance within the PAR spectral range (as photon flux density, 400 – 700 nm, μmol photons m–2 s–1)”.
- State type of radiometer and keep model in brackets to be concise: e.g.
Lines 178 – 180: “Measurements were recorded with a LI–COR LI–1500 logger and the upwelling/downwelling surface PAR was consecutively recorded using an integrated planar PAR sensor (LI–190R 2π, LI–COR)….”
Lines 180 – 181: “Under-ice transmitted PAR was recorded using an integrating scalar PAR sensor (LI–193 4 π)…”
Line 182: Was the snow put back into the hole during the under-ice arm deployment to avoid artificial increase in under-ice light values?
Line 185: Does the used model to calculate surface PAR at the sampling site include the cloud cover during that time period. Please specify. If it’s not included, this could have let to an overestimation of surface PAR.
Lines 195 – 196: Why was only surface water sampled if this study investigated vertical nutrient fluxes from the deeper Atlantic water? This study does not provide any measured or literature AW nutrient values for this region to support their discussion of increased surface nutrient levels due to vertical upwelling. (this comment also applies to lines 505 – 507: try to provide some literature values).
Line 224: Was ice core temperature measurements started from he top or bottom of the core? Please specify.
Line 225: When was the conductivity meter calibrated and what is its accuracy?
Line 235: Can you provide filtration volumes for these parameters, which would enable a repetition of this sampling?
Line 236: For how long and at what temperature were POC/PON filters combusted?
Line 237: How and when was the Turner fluorometer calibrated?
Section 2.8.1: I suggest including the NCP formula that uses all the defined variables. Also see my comment about “E” instead of “I” for irradiance
Lines 360 – 362: In this explanation of how the snow thickness threshold to reach zero NCP was derived, it is not explained if PE parameters were kept the same in the NCP calculation. If so, it should be highlighted in the discussion that this does not necessarily reflect the in situ algal photophysiological response.
Line 382: How was the mixed layer depth derived?
Line 383: How thick was the “warmer AW core” at Site 2?
Line 398: If the statistical test does not show any significant differences between measurement points for silicic acid, you cannot describe it as “decreasing”. The stats result says, there is no change. (also line 501 – “insignificant decrease” – statistically speaking, those two words rule each other out.)
Line 400: Was the diapycnal diffusivity enhanced at all sites?
Line 423: The text says that table 2 is only summarizing physical characteristics, while the table is actually also providing biochemical results. Those aren’t described in the text at all. Please add a description of those results.
Section 3.1.: I suggest including a plot to visualize flow cytometry and total cell abundance results. It’s difficult to grasp these big numbers and differences from the text.
Line 475: Was this transmittance value of 2.56% measured?
Line 548: The floe did not only consist of SYI as stated here.
Line 550: As stated in the methods, ice thickness has a minimal impact on light transmission compared to snow.
Section 4.2.2.: Please see my general comment of a lack of sufficient discussion in this section.
Technical comments:
Line 29: unit is missing term “photons”
Lines 99: Providing more than four digits in GPS coordinates is unreasonable
Throughout manuscript, make sure spelling of first-year sea ice and second-year sea ice is correct (dash is often missing).
The format of all figures should be the same throughout the manuscript, including using the same letter format for “(a)” vs “a)” vs “a”, using variable “depth” (not pressure) with positive numbers for profiles (negative in Fig. 4), and using a “real” minus in axis labels and units. Plot axes should also always start and end with a number and not have numbers only show somewhere in the middle of the axis (Figs. 2 & 4).
Line 132: “sampling site”
Line 133: Figure 1b) does not indicate water sampling at Site 3.
Line 135: Table 1 does not provide coordinates even though text is claiming it does.
Table 1: Table caption or sampling date in the table should indicate which year it was sampled.
Line 185: Specify “The resultant PAR transmittance”
Line 208: Specify “9 cm diameter”
Line 209 (and throughout manuscript): Only biochemical measurements were performed, no “geo”
Line 210: Include “..and TO make data from…”
Line 230: Specify “Nutrient samples from melted sea-ice bottom and water samples..”
Line 235: Specify “Chlorophyll a, POC and PON from melted sea-ice bottom samples…”
Line 245: Use same format “2–20 µm”
Line 247: The term “diluted melt” should be rephrased.
Line 268 (and throughout manuscript): The official symbol for irradiance is “E” not “I” (used in old literature). Hence, it should be “PE” parameters and “Es” and “Ec” for photosynthetic efficiency and photoacclimation parameter, respectively.
Line 270 (and throughout manuscript)” Use “mL” for unit
Lines 272 -273: Be concise “measured with an integrating scalar PAR probe (Walz US–SQS/L) and logger (Walz model ULM–500)” (I’m adding the the term “integrating” because none of these radiometers measures spectral light, they just provide one PAR value)
Line 284: Unit needs capital “L” for liters
Line 313: Equation 1 – variable “z” is not defined (just further below)
Line 340: The used letter for “Under-ice PAR transmittance” as “TS” is misleading as the lower “s” is used in the equation for snow. Hence, I would suggest using “TUI”. Also be careful to define “T” as transmittance, and not “transmission” (Line 241).
Line 370: Can you provide the number of hours for the storm in the text?
Line 371: Does indicated speed refer to “drift speed”?
Figure 3: See general comments. But also, salinity should not have unit. And if the x axis label already says “Days in May”, each measurement day doesn’t need to also say May
Table 2: Explain all used abbreviations in the table caption.
Table 3: Is not just showing PE parameters. It’s also showing NCP and chlorophyll results. This should be included in the table caption. Also these are “integrated” chlorophyll values
Table 4: Again, this table is not just showing PE parameters, but also integrated Chlorophyll. Please include that in the caption.
Lines 477 – 479: The grammar of this sentence is off, which makes it hard to understand. Please try to rephrase.
Line 552: Add “Here, the highest chl a was FOUND in the bottom of FYI…”.
Citation: https://doi.org/10.5194/egusphere-2026-4031-RC1 -
RC2: 'Comment on egusphere-2026-4031', Anonymous Referee #2, 01 Oct 2026
reply
Review of McKay et al. “Drivers of net community production in the bottom of Arctic pack ice”
The manuscript presents field observations of the bottom ice algal community of a drifting ice floe in spring near Svalbard. It presents repeated sea ice biogeochemical and accompanying atmospheric and oceanic measurements over the course of about 10 days. The paper discusses the interplay of light and nutrient availability on ice algae and net community production, including the effects of a storm event. They also estimate the area of the floe that would support net autotrophy vs heterotrophy.
I found the research interesting and valuable, as well as generally thorough and well-presented. The field sampling shows a significant coordinated sampling effort (including at least 12 cores per sea ice site!). My main suggestions surround the estimation of light and some of the presentation of the discussion, somewhat moderate revisions. Also, please note, I do not have technical expertise in turbulence measurements and cannot provide a deep assessment of them.
- The floe-scale NCP estimates are derived using % transmittance to determine a threshold snow depth that separates net autotrophic and net heterotrophic conditions. Given that the NCP curves already provide a compensation irradiance, I would recommend the use of an absolute (transmitted) PAR threshold rather than % transmittance. After all, transmittance is only half of the question—the other half being incident light intensity and duration (as the authors themselves demonstrate in L588-591, noting the effect of clouds). To then scale this, they could use a mean daily incident PAR (from Fig S9) as the starting value and use the downward exponential decay to calculate under-ice (or indeed, bottom-ice) PAR and compare it to the threshold.
- L157-159 and Figure S6 indicate that measurements of ice thickness were measured along the transect. If the authors adopt the approach in Point 1, then they could also consider ice thickness in their calculations, instead assuming a constant ice thickness. Although it is true that the Kd of snow is about 10x that of ice, differences in ice thickness of ~70 cm (e.g., Site 1 vs Site 2) could therefore have about the same difference as ~8 cm of snow (e.g., Site 1 vs Site 2 thin snow).
- I understand the desire to order the discussion topics thematically – nutrients, light, storms, floe-scale estimates. However, I did then find it challenging at times to bounce back and forth between discussions of the different sites and, only upon a second read and with notes, did I feel I was able to pinpoint what the authors believe to be the main explanatory factors at each site. Perhaps consider adding text that clearly conveys the takeaway story for each site. E.g., at Site 1, there was Si depletion and middling NCP; at Site 2 thin snow, there was more N, but negative NCP due to higher heterotrophy; etc.
Specific comments
L82: Please consider being more specific: what kind of effects have been shown for FYI to SYI?
L185: Please specify that the POWER dataset is modelled upon its introduction and also provide a citation.
L187: It’s not clear to me how often the % transmittance is altered although the surface PAR is hourly. I guess it would be with every ice sampling date? Please specify.
L230: Missing a space in “a10%”
L259: Saying “Additional NCP experiments” before the field experiments are introduced was a bit confusing. Please consider rewording or reordering 2.7.3 and 2.7.4.
L262: “Pseudo-nitzschia” should have a dash, here and throughout the text.
Section 2.8.1: Please consider including the actual fitting equation, especially as some of the parameters appear different from Arrigo et al. 2010. Please also clarify/check if Ps is actually the maximum photosynthetic rate or is the saturated photosynthetic rate. I also agree with Reviewer 1 about using the variable E instead of I.
L300: Please change to “estimated transmitted PAR irradiance”
Section 3.1. Environmental conditions: Is it ever discussed that the FYI is far thicker than all the SYI? Do the authors have any ideas as to why this would be the case?
Note that the following section is also labelled 3.1 (Algal abundance and taxonomy in the bottom ice)
Section 3.3-3.4: Please consider presenting the environmental PAR before the lab results, so that your readers may read the lab experiments conducted at 30 umol photons/m2/s in the context of the environment.
L491: Please consider qualifying by time of year/progression of the bloom.
L501: I would just say that the Si(OH)4 is consistent/showed no significant change.
L510. The comma after (p=0.04) should probably be a semi-colon
L513-538. Is there really evidence of Si limitation? It seems like a strong claim, given no relationship between Si and NCP, and a reliance on environmental ratios, rather than nutrient limitation experiments or Ks-based limitation calculations. I would like to see the stoichiometric ratios of ice algae for additional comparison, or for the language in L514, 523 for example to be more conservative than the term “limitation.”
L513: Please consider changing phytoplankton to microalgae due to the scope of the present manuscript. Same on L536.
L554-573: These two paragraphs felt out of place to me in a section that is nominally about light. Please consider moving or separating.
L645: Please change to “snow thickness primarily controls” depending on the choices to consider ice thickness.
L673: “These findings highlight the importance of nutrient fluxes…” I think this conclusion is based on the higher NCP at Site 3 with higher N fluxes due to the storm and underlying nutrient-rich waters, despite lower N in the bottom ice? However, this point wasn’t really discussed; I think it is important and should be added, probably to Section 4.1
L658-680: I don’t think these sentences are doing anything for you and would consider cutting them.
Figures and Tables
Figs. 2, 3, and 4: Please consider labeling the site numbers. I know there are slightly different water and ice sampling dates, but just choosing a convention should be fine.
Please consider adding transmitted PAR or % transmittance to either Table 2 or 3.
Table 2: Please consider consistently using cm or m for snow and ice thickness (you could chose differently for the two metrics) across this table and the rest of the paper.
Please consider moving Figure S7 to the main text, as I think the community composition is important enough to come up multiple times in the Discussion and it would make Section 3.1 easier to digest.
Citation: https://doi.org/10.5194/egusphere-2026-4031-RC2
Data sets
Sea ice biogeochemical data for the BREATHE project: Yermak Plateau 2023 R. McKay, J. Osanen, C. Laber, and K. Campbell https://doi.org/10.21334/NPOLAR.2025.18D7E643
Sea ice core data from BREATHE/SIDRiFT field school, north of Svalbard, May 2023 P. Itkin https://doi.org/10.21334/NPOLAR.2025.44A880E0
Snow pits from BREATHE/SIDRiFT field school, north of Svalbard, May 2023 P. Itkin https://doi.org/10.21334/NPOLAR.2024.A1AB4151
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 246 | 105 | 29 | 380 | 81 | 30 | 32 |
- HTML: 246
- PDF: 105
- XML: 29
- Total: 380
- Supplement: 81
- BibTeX: 30
- EndNote: 32
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The associated MSS data can be found here:
Koenig, Z. C., Campbell, K., & Saur Heiland, S. (2026). Ocean microstructure profiles (MSS) from the BREATHE cruise, Yermak plateau North of Svalbard, May 2023 [Dataset]. Norwegian Polar Institute. https://doi.org/10.21334/NPOLAR.2026.05015061