the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Mar 2026
Contrasting effects of river and erosion-derived inputs on Arctic Ocean acidification
Abstract. Although the Arctic Ocean is relatively small in volume, its extensive coastline delivers large quantities of terrigenous material from rivers and coastal erosion. As a result, the Arctic Ocean is impacted more strongly by terrigenous material than most other parts of the global ocean. Yet the effect of this material on carbon cycling and ocean acidification remains poorly quantified. In this study, we use an ocean biogeochemical model driven by observation-based estimates of terrigenous carbon, alkalinity, and nutrients to evaluate their contribution to the mean state, depth pattern, and seasonal cycle of ocean acidification, as measured by the aragonite saturation state. Riverine alkalinity generally mitigates acidification, whereas organic carbon from coastal erosion intensifies it. Nutrients from both sources mitigate ocean acidification at the surface by stimulating primary production, but intensify it at depth through subsequent remineralisation. Together, riverine and erosion-derived inputs account for about 20–40 % of the seasonal variability in the saturation state of the surface ocean. This amplification of the natural seasonal cycle is primarily caused by an increase in the summertime maximum of the saturation state. Terrigenous inputs also reduce the Arctic Ocean's capacity to absorb atmospheric CO2 by 17–25 %. Accurately representing carbon and nutrient inputs from rivers and coastal erosion in biogeochemical models is therefore important for reliable assessments of ocean acidification, ecosystem health, and carbon budgets in the Arctic Ocean.
- Preprint
(6625 KB) - Metadata XML
-
Supplement
(37620 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-683', Anonymous Referee #1, 16 Apr 2026
-
RC2: 'Comment on egusphere-2026-683', Anonymous Referee #2, 21 Apr 2026
Summary
The authors estimate the impact of riverine and coastal erosion fluxes on the carbonate system of the Arctic Ocean, with focus on aragonite saturation state. They find that erosion has opposing effect, leading to a lower aragonite saturation state, compared to riverine input, which increases the aragonite saturation. The authors nicely discuss the uncertainties related to fluxes, and highlight the need for improved estimates to better constrain the Arctic Ocean carbon cycle.
General comments
- I think that your discussion related to the uncertainties in fluxes from erosion, and how it would impact your results, is very interesting, and it would be good to highlight this with a sentence in the abstract.
- I’m wondering about the validity of assuming that coastal erosion supplies no alkalinity? Are there some references that you could add to support this assumption?
- You are estimating the effect of the uncertainties related to your erosion fluxes by scaling the change in the carbonate system linearly to the uncertainties in the fluxes. However, the carbonate system is not linear, so I am wondering about the validity of this approach?
- You have not indicated any spinup time of the model. While the spinup time may be less important for the results in the surface ocean, it will have more impact in the deep (where the volume turnover is longer), and especially the results in section 3.2.2. Could you comment on that maybe in the description of the simulations, and potentially in the discussion?
Specific comments
L6-7: “Riverine alkalinity generally mitigates acidification, whereas organic
carbon from coastal erosion intensifies it.” Consider changing to “Riverine input generally mitigates acidification, whereas input from coastal erosion intensifies it”, because it is not only about the alkalinity and the organic carbon, it is about the combined effect of carbon and alkalinity in both fluxes.L9: surface ocean -> surface Arctic Ocean?
L8-9: “Together, riverine and erosion-derived inputs account for about 20–40% of the seasonal variability in the saturation state of the surface ocean. This amplification of
the natural seasonal cycle is primarily caused by an increase in the summertime maximum of the saturation state.” -> This is quite difficult to understand first time reading. Consider reformulating. May help with only changing seasonal variability to seasonal amplitude.Section 1: very nicely written!
L91-92: is there a reference for the low alkalinity input from erosion?
L120: I would avoid using the word “please” in a scientific paper
L138-139: How did you distribute the input from the erosion, is it at the top surface layer?
L143-145: How did you decide on this partitioning for the material coming from coastal erosion? Is some of this put into the particulate material pool?
L155: why didn’t the simulations include erosion up to 1990? (earlier you wrote that the observational estimates go back to 1950)
Section 2.3: you have not written anything about the spinup of the model. How long is the spinup time? What fields was the model initialized from?
L211-213: Has the CO2 been generated on land, or in the freshwater?
L211-213: Why is the plume becoming supersaturated when reaching seawater? Is it an effect of salinity?
Section 3.2.2: Would it be possible to see if the model simulations become more realistic when including both erosion and riverine input, by comparing to observations? I’m thinking that there should be vertical profiles of carbon system parameters from single cruises, but is possible that there are not enough observations for such a comparison.
L461-531: These references may be useful here (first one touching on the effect of lability of terrestrial DOC on the spatial distribution of remineralization, second one uses a model with full stoichiometric variability, and discusses the effect of tDOC remineralization on pCO2, and the effect of tDOC on light attenuation):
Fransner, F., J.Nycander, C.-M.Mörth, C.Humborg, H. E.Markus Meier, R.Hordoir, E.Gustafsson, and B.Deutsch (2016), Tracing terrestrial DOC in the Baltic Sea—A 3-D model study, Global Biogeochem. Cycles, 30, 134–148, doi:10.1002/2014GB005078.
Fransner, F., Fransson, A., Humborg, C., Gustafsson, E., Tedesco, L., Hordoir, R., and Nycander, J.: Remineralization rate of terrestrial DOC as inferred from CO2 supersaturated coastal waters, Biogeosciences, 16, 863–879, https://doi.org/10.5194/bg-16-863-2019, 2019.
Citation: https://doi.org/10.5194/egusphere-2026-683-RC2
Data sets
Gridded carbon and nitrogen land-ocean fluxes north of 60°N from rivers and coastal erosion J. Terhaar, R. Lauerwald, P. Regnier, N. Gruber, and L. Bopp https://doi.org/10.17882/76983
Aragonite saturation state (Ω) from ORCA025–PISCES sensitivity experiments on Arctic terrigenous inputs (2005–2010) H. A. L. Hollitzer, L. Bopp, and J. Terhaar https://doi.org/10.17882/111924
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 234 | 101 | 17 | 352 | 395 | 23 | 25 |
- HTML: 234
- PDF: 101
- XML: 17
- Total: 352
- Supplement: 395
- BibTeX: 23
- EndNote: 25
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Hollitzer et al. have produced a very well written and meticulously presented article. The authors demonstrate a perfect mastery of their subject. They also back up their claims with a wealth of analyses, including dozens of figures presented in a detailed supplementary section, which is invaluable for a deeper understanding. This article is so clearly explained and accessible that I am certain it could be used for educational purposes or, at the very least, effectively reach a non-specialist audience. Which I think is great. It is very difficult to find anything to criticise in terms of its presentation.
They present an analysis that enables an in-depth study (in all respects) of the carbonate system (and subsequent acidification) in relation to terrestrial inputs (particularly from rivers and coastal erosion) into the Arctic Ocean, building on previous work (Terhaar et al. 2021). They shed new light on how these inputs influence acidification, particularly at depth. Whilst acidification appears to be mitigated at the surface (increase in PPN, alkalinity input from rivers), the remineralisation of organic matter (mainly from coastal erosion) exacerbates acidification at depth. They aim to add value to the previous work of Terhaar et al. 2021 and Zhang et al. 2024. I believe they have succeeded in doing so.
I noted the study’s limitations as I read through it, but I realised at the end that a significant portion of the paper was devoted to them. I found this very interesting, almost as much as the results. I would keep it, but it would be helpful for the reader to outline the structure of the article at the end of the introduction. Otherwise, I find this article a little too detailed and long; it could have been shortened in places, but I don’t mind keeping this format.
To sum up, my only major concern relates to the protocol, which reuses an earlier simulation by Terhaars et al. (2021). However, this simulation has already been accepted for publication. I have no problem with the use of a previous simulation in principle, but the authors could have taken this opportunity to correct a previously recognised weakness in the simulation of the previous model. Since the authors are willing to acknowledge a long list of limitations, I would unhesitatingly add (1) the very short duration of the start-up protocol, and possibly (2) the absence of permanent burial. They also use a linear relationship to compensate for sensitivity tests (i.e. instantaneous remineralisation). They claim that carrying out a new simulation is out of the question without providing any justification. Do the authors not work in modelling centres?
Although I consider this simulation to have been state-of-the-art five years ago, it would have benefited from a few updates. This is all the more true given that the authors do not present any control simulations to document the model’s drift. This makes the quantitative figures regarding the contribution of terrestrial inputs rather unreliable. I understand the costs involved in running the models, and I do not wish to prevent the publication of this excellent article, particularly as it has been written by a young researcher at the start of their career. However, I believe this to be a major shortcoming of this study. The authors should, at least, acknowledge this, or, ideally, carry out a longer simulation and/or provide a model drift analysis (perhaps for the next study).
Please find below my other comments point by point (minor concerns) which should be easy to address. Kind regards.
Line 92 : This article lies on the important assumption that coastal erosion does not deliver Alkalinity. I think this is a decent assumption that coastal erosion mostly delivers OM. However, could they justify this with either a short demonstration and/or backing up with literature? Permafrost also encapsulate minerals that can undergo weathering.
Line 144: When a terrigenous DOC with specific lability is not available in a model, I understand that one should come with some assumption. And I totally agree with the claim that riverine matter contains an important share of recalcitrant DOC. However, I do not understand why the other share is considered labile (sent to DIC directly) and not to (semi-labile) DOC. I could eventually understand the instant remineralization of fresh OM from coastal erosion permafrost, but not quite this one. NOTE: I see that the authors discuss this issue later in the discussion and are constrained by the fixed stoichiometry of the model. At least, a sensitivity test would have been appreciated. Maybe next time.
Line 181 : “aligns well” seems exaggerated. Since most of the values are exceeding 2 in glodap (Fig. 1 & S2), it is hard to know what are the bias between the model and the observations, Although I acknowledge the general pattern is here. I would eventually start by acknowledging the strong biases in the inflowing Atlantic and Pacific waters.
Figure 2 concept is excellent and extensive but very hard to read and understand rapidly and is also under-exploited to synthetize the outputs of the article. Design should be improved for better readability. Subjective suggestions: One inlet map for summer sea-ice may be enough. Representation of the continental shelf must be more explicit. Increase compacity & remove blank/empty areas. Is the gray circle arrow needed ? Stating the exact months necessary (move to caption) ? Try not to repeat what is not necessary (e.g. seasons in the title for lower panels, etc). Increase fontsize.
Line 241: It is nice to see that different models come to the same conclusion (e.g. Polimene et al., Oziel et al.). This article has the nice added value and steps one step further into the complexity of the carbonate system.
Line 370: Shocking, but transparency really appreciated.
Line 382: I doubt the omega response is linear… is it ? could you justify your approach ?
Line 392: why aren’t you able to do new runs ? Don’t you have HPC resources in your institutions (e.g. IPSL) ? I find it a bit easy
Line 497: Soil DOC is labile yes, but not cessarily riverine DOC which experienced a lot of degradation before reaching the ocean and is expected to be less labile than oceanic DOC.