General comments:

The following article addresses the impact of river-focused ocean alkalinity enhancement on carbon dioxide removal. It present’s findings that mCDR broadly scales with OAE as other studies have similarly shown. While I believe that it’s important to expand the number of OAE simulation studies and varying the means of alkalinity delivery is critical, the article is not particularly interesting. The authors could do more to differentiate their contribution, particularly given their use of an emissions-driven ESM. I was particularly surprised that they focus so little on changes in atmospheric temperatures, which appear counterintuitive. Moreover, there is no description at all of the land carbon sink and how it responds to OAE (one of the principal advantages of using a fully-coupled ESM). I would like to see both of these aspects developed in a revised manuscript. In my opinion, several of the current figures need cutting or revising to be useful to the reader.

Specific comments

L26 Is this true? Wouldn’t afforestation-based mCDR also absorb CO2 and reduce acidification?

L34-35 These are surface atmospheric temperature increases not SST increases I believe.

L53 I would use a more recent estimate of this consistent with the latest scenarios (e.g. (Smith et al., 2024))

L59 Excluding geological reservoirs.

L65-67 See previous point, other techniques could potentially also do this.

L68-70 This definition is a bit inaccurate. Alkalinity is perhaps better defined as the excess of H+ accepters over donors.

L70-71 This alkalinity decline may also be due to biotic feedbacks, (Barrett et al., 2025; Kwiatkowski et al., 2025).

L73 I’m not sure what excess H+ is in this context.

L74-75 Disequilibrium is not always enhanced. In areas of natural carbon outgassing, such as eastern boundary upwelling systems, it would likely be reduced. The net effect would be the same however, enhanced ocean carbon storage.

L103-104 There are a growing number of regional OAE simulation studies that go beyond this, some of which the authors go on to cite.

Figure 1. I don’t find this figure particularly useful. The link between weathering and atmospheric CO2 is unclear to me. Is this due to intensification of the hydrological cycle? And the role of sources and sinks of alkalinity in ocean sediments and marine biota is absent.

L130 This equation is unnecessary (and is unnumbered).

L141 Add equation number.

L145-149. The language used here is not clear. Prescribed CO2 can still be transiently changing. Are simulations concentration-driven or emissions-driven? If emissions-driven, with dynamic atmospheric CO2 this needs to be explicit here.

L153 “concentration” should be “emissions” as emissions not concentrations are prescribed in esm-hist.

L155 I don’t know what an SSP-based RCP is. You either ran an SSP or an RCP or is this some hybrid forcing I am not aware of.

L162-164 These simulation descriptions are confusing. What is meant by “based on… from 2050”?

L165-169 Is the ocean alkalinity inventory balanced in the control run? Or is there some drift?

Figure 3 In printed format it is impossible to see any of the detail of this figure. Fonts are too small, lines to thin and legends impossible to read.

L220 Clarify in the legend whether these are global zonal means or a specific transect.

L225-226 See earlier point. OAE does not always enhance disequilibrium. If it does, I would like to see a plot of this.

L235 I think uatm units should be used for partial pressures.

L245 How much later? Give the year.

L258 This seems like a trivial equation to provide, it’s just a depth integral.

L308 I would avoid describing a global pH level as “healthy”.

L332 The figure ordering is strange with respect to the text.

L334-335 Does this mean the reductions in atmospheric air temperatures are not proportional to OAE?  This is an important finding and requires discussion which appears to be absent. Why do the authors think this is the case? Is this because of internal variability? Are larger ensemble sizes of each experiment required?

L342 So the reductions in atmospheric CO2 are consistent with the extent of OAE but not the reductions in surface temperatures? Please discuss, perhaps the temperature values are type errors, it’s hard to see differences in figure 3.

L367-368 It’s primarily due to the transport of water masses into the subsurface prior to full- equilibration.

L383-375. Can the authors explain the role of the simulation time? Is this because of sediment feedbacks? Most ESMs lack such feedbacks anyway (see Planchat et al., 2023) so I’m not sure running the models for longer would make a difference.

L386-389 Are these differences in efficiency robust? Have similar effects been detailed in other studies and if so, can the authors explain the mechanism controlling this?

L398-400 Be clear that Zhou et al perform OAE locally in all grid cells and don’t rely on rivers for delivery.

L458 How do these rates of acidification and carbon uptake compare to those in the CTL simulation?

L487-489 Indicative that even riverine OAE results in loss of non-equilibrated water masses from the surface ocean, which are equilibrated of ocean circulation timescales of centuries.

References

Barrett, R. C., Carter, B. R., Fassbender, A. J., Tilbrook, B., Woosley, R. J., Azetsu-Scott, K., et al. (2025). Biological Responses to Ocean Acidification Are Changing the Global Ocean Carbon Cycle. Global Biogeochemical Cycles, 39(3), e2024GB008358. https://doi.org/10.1029/2024GB008358

Kwiatkowski, L., Planchat, A., Pyolle, M., Torres, O., Bouttes, N., Comte, A., & Bopp, L. (2025). Declining coral calcification to enhance twenty-first-century ocean carbon uptake by gigatonnes. Proceedings of the National Academy of Sciences, 122(23), e2501562122. https://doi.org/10.1073/pnas.2501562122

Planchat, A., Kwiatkowski, L., Bopp, L., Torres, O., Christian, J. R., Butenschön, M., et al. (2023). The representation of alkalinity and the carbonate pump from CMIP5 to CMIP6 Earth system models and implications for the carbon cycle. Biogeosciences, 20(7), 1195–1257. https://doi.org/10.5194/bg-20-1195-2023

Smith, S., Geden, O., Gidden, M., Lamb, W. F., Nemet, G. F., Minx, J., et al. (2024). The State of Carbon Dioxide Removal - 2nd Edition. https://doi.org/10.17605/OSF.IO/F85QJ

Zhu et al. (2025) discussed river-based ocean alkalinity enhancement (OAE) for carbon dioxide removal in Earth system models. The main innovation of this paper lies in its specific focus on river-based OAE, distinguishing it from previous studies that typically assumed OAE on a broader scale, such as in open ocean basins (i.e., Lenton et al., 2018) or coastal areas (He and Tyka, 2023). Additionally, in contrast to global studies that cover estuarine regions (e.g., Zhou et al. 2024), this study uniquely utilizes an emission-driven Earth System Model (ESM), which provides an opportunity to further investigate atmospheric feedback effects (Tyka, 2025). However, the manuscript appears to capture such feedback but does not yet attempt to further distinguish and discuss these atmospheric feedback effects. Refining this section would enhance the scientific significance of the paper. Furthermore, there is still room for improvement in the figures and presentation. I will provide specific suggestions for improvement in the following sections. Overall, the conceptual foundation of this research is solid, and revisions and improvements would make this paper a valuable contribution to the growing body of literature on ocean alkalinity enhancement models.

Additionally, I would like to share an idea with the authors: Given that both Zhu et al. (2025) and Zhou et al. (2024) used the CESM2 framework but with different atmospheric components, and considering that Zhou et al. (2024) provide an OAE efficiency budget for various global regions, converting Zhou et al. (2024)'s open-source results to the same OAE injection areas as in Zhu et al. (2025) would not require significant additional work. However, this approach could provide potential insights into the differences in OAE budgets due to atmospheric forcing and feedback effects. Please note that this is beyond the scope of this review, and the authors are not required to address this suggestion in the revision.

Yinghuan Xie, University of Tasmania

Major comments:

195:The subtropical gyres seem to contribute to two distinct ventilation regions in around 30°N and 30°S, which are analyzed in the Discussion section (paragraph at 406) but are not mentioned in the Results. Furthermore, Fig. 3 shows that the distribution of ALK across global ocean basins is inconsistent. For example, the ALK excesses in the North Atlantic is significantly stronger than in the North Pacific. Therefore, it would be helpful to calculate the contents in Fig. 4 separately for the Atlantic, Pacific, and Indian Oceans. The same approach is also recommended for the DIC analysis in Fig. 8.

Fig. 3: It is recommended to use the anomaly for panels c-f, especially panel f. The differences between the curves in the current version are too small, which affects readability. 

269-282:The DIC decrease in the Southern Ocean and equatorial Pacific is not mentioned? This is an apparent phenomenon, and overlooking it is weird. This may be due to the atmospheric feedback effects present in the ESM, for example, the ALK injection in the northern oceans reducing PCO2atm, which could have led to net outgassing in the Southern Ocean. However, further analysis is needed to confirm this. A more detailed analysis of the atmospheric and surface ocean PCO2 outputs for both the CTRL and experimental groups is necessary to determine whether the DIC decrease is due to atmospheric feedback or other mechanisms.

Fig8: See suggestion for 195.

320: It is recommended to provide a more detailed explanation for the increase in the North Atlantic in OWE0, particularly in Hudson Bay and the Northwestern Channel. Additionally, a noticeable pH increase is also observed in the Ross Sea. Is this related to sea ice or outgassing? Further analysis and clarification would be beneficial.

332-334: The result is reasonable, but comparing the temperature values at a single time point is not appropriate. It is recommended to use the average temperature over the last 10 years or a similar metric for the analysis.

417-438: Recommend to streamline this section, as it currently appears more like a literature review rather than a targeted discussion.

Minor comments：

Figures 3-8：The numbering of the subplots, the legend, and the labels have fonts that are too small and need to be enlarged. Figures suffer from low image resolution, which affects readability. Please ensure that the images in the final published version are clear.

203:50-70N? There is a difference with Fig. 5d.

226: It is suggested to indicate the time of comparison here or in the caption of Fig. 6. Although it is provided later, it has not been explained earlier.

245: Considering that the rate of rebound is rapid, "rapidly returns" would be more appropriate than "eventually returns".392: Fig 4, not 3.

395:The logic seems unclear. It is recommended to rephrase as: Compared to the North Atlantic, the western boundary current of the North Pacific occurs outside the island chains, and a large amount of ALK excess is enriched inside the island chains, preventing it from spreading to the wider Pacific.