Calcium carbonate dissolution rate changes under future climate scenarios

Trossman, David Samuel

doi:10.5194/egusphere-2026-2679

Preprints

https://doi.org/10.5194/egusphere-2026-2679

Preprints

21 May 2026

| 21 May 2026

Status: this preprint is open for discussion and under review for Biogeosciences (BG).

Calcium carbonate dissolution rate changes under future climate scenarios

David Samuel Trossman

Abstract. Calcium carbonate is known to be dissolving throughout various portions of the water column, but little is known about how this will change in the future. Model output of a large range of climate scenarios and an established diagnostic approach are used to understand the future of calcium carbonate dissolution rates. Changes in ocean salinity and mean age are of leading order importance to future calcium carbonate dissolution rates. The calcium carbonate dissolution rates can range by an order of magnitude across climate scenarios in regions such as the subtropical Atlantic Ocean basins. Some geoengineering methods are more effective than others at decreasing calcium carbonate dissolution rates in particular regions of the ocean; however, no single technique is effective everywhere. Altered calcium carbonate dissolution rates have implications for the physical ocean state projections, such as those for sea level, which are discussed in terms of direct and indirect impacts. Direct impacts due to the presence of suspended sediments comprised of calcium carbonate can be significant in some river deltas and nearby coasts.

Received: 10 May 2026 – Discussion started: 21 May 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

David Samuel Trossman

Status: open (until 17 Jul 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-2679', Anonymous Referee #1, 22 Jun 2026 reply
This manuscript investigates how calcium carbonate (CaCO₃) dissolution rates in the ocean may change under future climate scenarios, using CMIP6 (SSP5-8.5) and GeoMIP model outputs combined with the diagnostic approach of Sulpis et al. (2021). The study identifies salinity and mean water age as the dominant factors controlling future dissolution rate changes and compares the effectiveness of various geoengineering interventions. While the topic is timely and relevant, given the importance of calcium carbonate dissolution for ocean carbon and alkalinity cycles, the manuscript suffers from several significant weaknesses that limit its scientific contribution.
My primary concerns:
1. The manuscript attempts to address too many loosely connected topics (dissolution rates, multiple geoengineering scenarios, sea level, suspended sediments in river deltas) without adequately integrating them into a unified narrative. Section 4 on "Direct vs indirect climate consequences" feels particularly disconnected from the main analysis.
2. The perturbation-based approach, while computationally efficient, is crude and lacks validation. The study does not verify that applying climate model perturbations to present-day observations produces physically meaningful results.
3. Many key claims about implications (e.g., for sea level, carbon budgets) are left unquantified and deferred to "future studies," which undermines the practical value of this work.
4. The finding that salinity dominates dissolution rate changes appears to be largely a mathematical artifact of the Alk* formula (Eq. 1), where salinity has a coefficient that is much larger than the other terms.
Specific Comments
1. Scope and focus
The manuscript covers calcium carbonate dissolution, CMIP6 projections, GeoMIP geoengineering scenarios, sea level implications, and suspended sediment concentrations in river deltas. These topics are not well integrated. Specifically:
What is the logical connection between open-ocean CaCO₃ dissolution rates and suspended sediment concentrations in river deltas?

The abstract mentions "sea level" implications, but the quantitative link between dissolution rate changes and sea level is never established.

I recommend the author either (a) focus the manuscript solely on dissolution rate sensitivity analysis, or (b) provide a much clearer and quantitative framework connecting all topics.

2. Perturbation approach
The approach of adding climate model perturbations to observational data (line 37-40) is a significant simplification. The authors acknowledge this but do not adequately address:
Why is it valid to assume that the Alk*-age relationship (and its slope) remains unchanged when perturbations are applied?

Cross-isopycnal transports and mixing are not accounted for (line 283-284). How large might this error be?

The empirical formula for inferring ages from Carbon-14 (Eq. 2) is described as "crude" (line 289) and can produce unreliable results for young waters. This is a major limitation for the GeoMIP analysis.

The author should provide at least a qualitative assessment of the errors introduced by these assumptions.
3. Model validation
The authors state (line 59-61): "We do not perform an assessment of the CMIP6 output with the GLODAPv2 data." While the temporal mismatch is a valid concern, the complete absence of any validation is problematic. Can the authors at least demonstrate that the CMIP6 control simulations produce reasonable dissolution rates compared to Sulpis et al. (2021) estimates, even if the time periods differ?
4. Salinity dominance could be mathematical artifact?
The main finding that salinity dominates future dissolution rate changes (e.g., line 266) needs to be critically examined. Looking at Equation 1:
Alk* = TA + 1.26[NO₃⁻] − 66.4S
The coefficient for salinity (66.4) is ~50× larger than for nitrate (1.26) and effectively determines Alk* changes. The authors even acknowledge this (line 114-116): "The impact of salinity on Alk* is apparent through Eq (1)... this gets magnified after scaling those factors by their coefficients."
Is the salinity dominance a genuine physical insight or simply a consequence of the formula's structure? I request the author discuss this issue explicitly and clarify what new physical understanding this result provides beyond what is already embedded in the Sulpis et al. (2021) methodology.
5. Incomplete GeoMIP analysis
Only 2 of the 6+ GeoMIP scenarios include Carbon-14 output to estimate mean age changes (line 76-77). Since mean age is identified as one of the two dominant factors, the comparisons of geoengineering effectiveness (e.g., Figure 7, conclusions) are incomplete and potentially misleading. The conclusions about which geoengineering techniques are "most effective" should be heavily caveated or removed.
6. Unquantified implications
The manuscript repeatedly defers quantitative analysis:
"Future studies will investigate the direct and indirect impacts of these changes more quantitatively" (line 298-299)

"we don't know the full range of possible calcium carbonate dissolution rates" (line 290-291)

"it is beyond the scope of the present study to quantify these effects" (line 249-250)

If the implications cannot be quantified, the "so what?" question remains unanswered. What is the practical significance of a factor-of-two or order-of-magnitude change in dissolution rates? How does this translate to Pg C or mm of sea level?
7. Section 4 – suspended sediments
This section appears to be a separate mini-study that does not connect meaningfully to the main dissolution rate analysis. Specific issues:
The assumption that TSM equals SSC and that all suspended matter is sediment (line 210-214) is crude.

The connection between SSC in river deltas and open-ocean CaCO₃ dissolution is unclear.

The "back-of-the-envelope" calculations (line 213) are insufficiently rigorous for a peer-reviewed manuscript.

I recommend either removing this section or substantially expanding it with proper methodology and clearer connection to the main analysis.
8. Figure quality and presentation
Many results are described as "not shown" (e.g., lines 120, 130, 145). If these results are important enough to mention, they should be included in supplementary material.

Figure 11 presents ranges across all biomes and scenarios but is difficult to interpret. Consider reorganizing or splitting into multiple panels.

The biome numbering system (#1-#10) requires readers to constantly refer back to figure captions. Consider using abbreviated names instead.

9. Writing clarity
The manuscript is difficult to follow in places:
The Methods section (Section 2) is too brief and does not clearly explain how perturbations are applied.

The transition between CMIP6 results (Section 3.1) and GeoMIP results (Section 3.2) is abrupt.

The logic connecting dissolution rates → carbon budget → sea level is not clearly laid out.

Technical Corrections
1. Line 217: Inconsistent minus sign format: "kg m˘3" should be "kg m⁻³" (appears multiple times)
2. Reference list: Check formatting consistency (e.g., some entries have issue numbers, others do not)
3. Line 20: "calcium dissolution rates" – missing "carbonate"?
4. Equation 2: The notation "ln(1)" equals zero, so this term appears unnecessary. Please clarify or simplify.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-2679-RC1
- CC1: 'Reply on RC1', David Samuel Trossman, 26 Jun 2026 reply
  
  Dear editor,
  
  I will be submitting an edited version of the manuscript, “Calcium carbonate dissolution
  
  rate changes under future climate scenarios,” to Biogeosciences. I have changed
  
  much of the text and presentation of the figures according to the reviewer’s suggestions.
  
  1. I now include a comparison of the calcium carbonate dissolution rate estimates
  
  from the present-day period of the CMIP6/GeoMIP output and those from the
  
  GLODAPv2 and other observationally-derived data. These estimates compare quite
  
  well, suggesting that CMIP6/GeoMIP output has sufficient fidelity to be used to
  
  make projections about future calcium carbonate dissolution rates.
  
  2. I also demonstrate that, while there is a shift in potential alkalinity towards lower
  
  values in the future climate period relative to the present-day period and there is
  
  a change in slope between alkalinity and salinity from the present-day to future
  
  climate periods, the slope of the relationship between potential alkalinity and mean
  
  age does not significantly change from the present-day period to the future climate
  
  period. This suggests the Alk*-mean age slope-based method I use to estimate
  
  future calcium carbonate dissolution rates should be about as valid as Sulpis et al.
  
  (2021)’s estimates.
  
  3. Lastly, because I increased the uncertainty in the mean ages in the GeoMIP calculations
  
  by propagating the error in the mean age associated with not having ages
  
  less than 200 years (only having Carbon-14-based ages) and I realized that the units
  
  were not consistent in the tracers that should be in μmolkg−1 (instead of molm−3),
  
  the figures with GeoMIP results have changed as well. These issues were not present
  
  with the CMIP6 analyses, but because alkalinity became a significant factor in es-
  
  timating calcium carbonate dissolution rates, I devised a method to predict what
  
  the future alkalinity will be based on present-day salinity and alkalinity and future
  
  salinity. I compared a Generalized Additive Model-based prediction with a
  
  Random Forest-based prediction and found that the Random Forest estimated the
  
  future alkalinity within a 10−3% error on average and within a 2.94 μmolkg−1 rootmean-
  
  square error. Then I re-estimated the calcium carbonate dissolution rates
  
  using the CMIP6 output and these future alkalinity estimates.
  
  I address each of the reviewer’s points below.
  
  1 Reviewer #1 (Comments to Author (shown to authors):
  
  • This manuscript investigates how calcium carbonate (CaCO3) dissolution rates in
  
  the ocean may change under future climate scenarios, using CMIP6 (SSP5-8.5)
  
  and GeoMIP model outputs combined with the diagnostic approach of Sulpis et al.
  
  (2021). The study identifies salinity and mean water age as the dominant factors
  
  controlling future dissolution rate changes and compares the effectiveness of various
  
  geoengineering interventions. While the topic is timely and relevant, given the
  
  importance of calcium carbonate dissolution for ocean carbon and alkalinity cycles,
  
  the manuscript suffers from several significant weaknesses that limit its scientific
  
  contribution.
  
  • Thank you for taking the time to provide such feedback. I have substantially
  
  modified the manuscript in the ways described above and respond
  
  to each of your concerns below.
  
  • My primary concerns: 1. The manuscript attempts to address too many loosely
  
  connected topics (dissolution rates, multiple geoengineering scenarios, sea level, suspended
  
  sediments in river deltas) without adequately integrating them into a unified
  
  narrative. Section 4 on ”Direct vs indirect climate consequences” feels particularly
  
  disconnected from the main analysis.
  
  • Olivier Sulpis and I have performed back-of-the-envelope estimates of
  
  how much sea-level has changed since pre-industrial times due to calcium
  
  carbonate dissolution (about 84 mm), but Olivier also suggested removing
  
  any sea-level analysis when we spoke. Because my current role focuses
  
  on sea-level, I heavily edited what I wrote. The connection between the
  
  future calcium carbonate dissolution rates is that 40% of sediments are
  
  made of calcium carbonate, sediments in sufficiently high concentrations
  
  will alter sea-level by Archimedes’ principle, and increased dissolution of
  
  calcium carbonate-based sediments in the future will therefore contract
  
  sea-level due to the reduction in density of the sediment when the calcium
  
  carbonate dissolves. This may eventually be a significant factor to
  
  account for in regional sea-level budgets, which is the reason why I included
  
  this analysis in the manuscript. I focused on river deltas because
  
  these are the regions where there are sufficiently high concentrations of
  
  sediments to impact sea-level. But I have now removed this discussion
  
  from the manuscript. As for the geoengineering scenarios, one of the
  
  purposes of my study is to examine the possible future ranges of calcium
  
  carbonate dissolution rates. The GeoMIP simulations’ output serves as a
  
  lower bound and the CMIP6 scenario’s output serves as an upper bound.
  
  I did not include other scenarios because I did not want to broach the
  
  question of which scenario is most realistic. I was only interested in the
  
  range and understanding the estimates’ sensitivities to particular factors.
  
  • 2. The perturbation-based approach, while computationally efficient, is crude and
  
  lacks validation. The study does not verify that applying climate model perturbations
  
  to present-day observations produces physically meaningful results.
  
  • Sulpis et al. (2021) performed the validation relative to sediment trap
  
  data for the calcium carbonate dissolution rate estimates from the Alk*-
  
  mean age slope method they used and I use here. The CMIP6 biogeochemical
  
  output has been compared with independent observational
  
  data products, at least in the Southern Ocean, by Cheng, M., N. Maher,
  
  M. J. Ellwood, 2025: Evaluating the performance of CMIP6 models
  
  in simulating Southern Ocean biogeochemistry. Biogeosciences, 22,
  
  7269–7291, 2025 https://doi.org/10.5194/bg-22-7269-2025. In addition,
  
  I now compare the calcium carbonate dissolution rate estimates using
  
  the same GLODAPv2 and other observationally-derived data that Sulpis
  
  et al. (2021) did and those using the near-present-day years of the
  
  CMIP6/GeoMIP simulation output. The correlation between these two
  
  dissolution rate estimates globally is about 0.93. A figure is shown later
  
  in this response. While potential alkalinity using these two data sets
  
  has a correlation of about 0.8, it is the slope of the potential alkalinity
  
  versus mean age relationship that matters, which is why the correlation
  
  between the calcium carbonate dissolution rate estimates can be higher
  
  than the correlation between the potential alkalinity estimates. If there
  
  are other comparisons you would like to see using available data, I am
  
  open to suggestions.
  
  • 3. Many key claims about implications (e.g., for sea level, carbon budgets) are left
  
  unquantified and deferred to ”future studies,” which undermines the practical value
  
  of this work.
  
  • The sea-level discussion has now been removed from the manuscript,
  
  which is mostly (if not entirely) what you’re referring to here. As you
  
  seemed to imply above, the sea-level discussion belongs in its own separate
  
  manuscript, where I should address the issues that I deferred to
  
  future studies.
  
  • 4. The finding that salinity dominates dissolution rate changes appears to be largely
  
  a mathematical artifact of the Alk* formula (Eq. 1), where salinity has a coefficient
  
  that is much larger than the other terms.
  
  • Yes, in the previously submitted draft of the manuscript, I explained why
  
  the calcium carbonate dissolution rates are most sensitive to salinity and
  
  mean ages mathematically. The underlying issue here is whether the
  
  Alk*-mean age slope-based method is valid in a future climate, and not
  
  just in the past/present climate in which it was empirically derived. Because
  
  this cannot directly be tested (using future observational data that
  
  doesn’t exist yet), I made an attempt to be clear about my assumption
  
  that this method is valid in a future climate. To assess this assumption,
  
  I have looked at the phase space of alkalinity, nitrate, and salinity
  
  (the variables that explicitly come into the Alk* formula). The phase
  
  spaces for the present-day and future climate periods are now shown
  
  (see later in this response). The main issue with using this method of
  
  calculating the calcium carbonate dissolution rates is the extent to which
  
  cross-isopycnal transports and mixing become more or less important in
  
  a future climate. According to the studies cited in Figure 2.2 of Melet, A.
  
  V., R. Hallberg, D. P. Marshall, 2022: Chapter 2 - The role of ocean mixing
  
  in the climate system, Editor(s): Michael Meredith, Alberto Naveira
  
  Garabato, Ocean Mixing, Elsevier, 5-34, https://doi.org/10.1016/B978-
  
  0-12-821512-8.00009-8, there can be regional differences in ocean mixing
  
  of up to 40%, but only 5% globally. Together with the strong relationship
  
  between our estimates using observationally-derived data and using
  
  present-day climate model output, we suggest that the estimates from
  
  Sulpis et al. (2021) are approximately as valid over basin scales as our
  
  calculations in a future climate context.
  
  • Specific Comments 1. Scope and focus The manuscript covers calcium carbonate
  
  dissolution, CMIP6 projections, GeoMIP geoengineering scenarios, sea level implications,
  
  and suspended sediment concentrations in river deltas. These topics are not
  
  well integrated. Specifically: What is the logical connection between open-ocean
  
  CaCO3 dissolution rates and suspended sediment concentrations in river deltas?
  
  The abstract mentions ”sea level” implications, but the quantitative link between
  
  dissolution rate changes and sea level is never established. I recommend the author
  
  either (a) focus the manuscript solely on dissolution rate sensitivity analysis, or (b)
  
  provide a much clearer and quantitative framework connecting all topics.
  
  • The connection between open-ocean calcium carbonate dissolution rates
  
  and suspended sediment concentrations is likely negligible. It’s the connection
  
  between near-coastal calcium carbonate dissolution rates and suspended
  
  sediment concentrations that may be important to take into account
  
  for regional sea-level budgets applications, but because only 10%
  
  of dissolution occurs on the continental shelves (not even the near-coastal
  
  environment), the connection is tenuous. I will remove the sea-level component
  
  of the manuscript and focus solely on future dissolution rates.
  
  • 2. Perturbation approach The approach of adding climate model perturbations
  
  to observational data (line 37-40) is a significant simplification. The authors acknowledge
  
  this but do not adequately address: Why is it valid to assume that the
  
  Alk*-age relationship (and its slope) remains unchanged when perturbations are
  
  applied? Cross-isopycnal transports and mixing are not accounted for (line 283-
  
  284). How large might this error be? The empirical formula for inferring ages from
  
  Carbon-14 (Eq. 2) is described as ”crude” (line 289) and can produce unreliable
  
  results for young waters. This is a major limitation for the GeoMIP analysis. The
  
  author should provide at least a qualitative assessment of the errors introduced by
  
  these assumptions.
  
  • The Alk*-mean age relationship is valid within the range of observed
  
  combinations of environmental properties (temperature, salinity, nitrate,
  
  phosphate,...). We now include a phase space diagram that shows that
  
  the ranges of the variables explicitly involved in the Alk* equation (alkalinity,
  
  nitrate, and salinity) do not significantly change from the present-day
  
  simulation output and the future climate output. While temperature
  
  does significantly change and the relationship between alkalinity and
  
  salinity has a slightly different slope, temperature and alkalinity do not
  
  impact Alk* by nearly as much as salinity, which does not change in its
  
  slope with mean age in our calculations from the present-day to future
  
  climate output. Only the intercept is altered going from the present-day
  
  period to the future period because there is a shift to smaller Alk*.
  
  • The error analysis is a good suggestion and I have attempted to perform
  
  this error analysis. For instance, by excluding the mean age perturbations
  
  of water less than 200 years old (but still accounting for the
  
  Carbon-14-based mean age perturbations) in the GeoMIP output, there
  
  is on average a 15.9% difference in the mean age. We now account for
  
  this in the calculations by using
  
  √0.22 + 0.1592 as the total uncertainty in
  
  mean age (i.e., assuming these two sources of uncertainty are independent).
  
  The cross-isopycnal transport and mixing was already included in
  
  the uncertainties in the calculations performed by Sulpis et al. (2021)
  
  and myself as a 6 μmolkg−1yr−1 uncertainty in the calcium carbonate dissolution
  
  rates. However, the basin-wide comparisons shown in Figure
  
  1 of Wu et al. (2025) between their estimates that account for cross-
  
  [See supplement for Figure 1, described below]
  
  isopycal transport and mixing and the estimates of Sulpis et al. (2021)
  
  suggest that there is good agreement between the two methods aside
  
  from the shallow (above the aragonite saturation depths < 500 − 1000 m)
  
  subtropical (plus one subpolar) ocean basins, but the uncertainties in
  
  the Sulpis et al. (2021) estimates were excluded from their comparison.
  
  Because ocean mixing is expected to increase by a small amount globally
  
  (5%), we suggest our future climate estimates are valid everywhere Figure
  
  1 of Wu et al. (2025) shows good agreement and place the caveat on
  
  our estimates in shallow subtropical (plus the subpolar North Atlantic)
  
  ocean basins that our estimates are likely too large when cross-isopycnal
  
  transport and mixing are accounted for.
  
  • 3. Model validation The authors state (line 59-61): ”We do not perform an assessment
  
  of the CMIP6 output with the GLODAPv2 data.” While the temporal
  
  mismatch is a valid concern, the complete absence of any validation is problematic.
  
  Can the authors at least demonstrate that the CMIP6 control simulations produce
  
  reasonable dissolution rates compared to Sulpis et al. (2021) estimates, even if the
  
  time periods differ?
  
  • This is also a good suggestion. I have estimated the calcium carbonate
  
  dissolution rates using the CMIP6/GeoMIP control simulation output to
  
  compare with those estimated with the GLODAPv2 data. They compare
  
  fairly well with each other with a correlation of 0.93.
  
  • 4. Salinity dominance could be mathematical artifact? The main finding that salinity
  
  dominates future dissolution rate changes (e.g., line 266) needs to be critically
  
  examined. Looking at Equation 1: Alk∗ = TA + 1.26[NO]66.4S The coefficient for
  
  salinity (66.4) is 50× larger than for nitrate (1.26) and effectively determines Alk*
  
  changes. The authors even acknowledge this (line 114-116): ”The impact of salinity
  
  on Alk* is apparent through Eq (1)... this gets magnified after scaling those factors
  
  by their coefficients.” Is the salinity dominance a genuine physical insight or simply
  
  a consequence of the formula’s structure? I request the author discuss this issue
  
  explicitly and clarify what new physical understanding this result provides beyond
  
  what is already embedded in the Sulpis et al. (2021) methodology.
  
  • Yes, I tried to make this clear that the salinity dominance is a direct
  
  consequence of the formulation that Carter et al. (2014) developed. The
  
  dominance of mean age comes from the slope of the mean age-Alk* re-
  
  [See supplement for Figure 2]
  
  lationship. The influence of the other factors were not clear to me so
  
  I performed the comparisons as previously submitted. Because I made
  
  the mistake of using units of molm−3 for oxygen, alkalinity, nitrate, phosphate,
  
  and DIC instead of the units they should have had (μmolkg−3),
  
  I performed the calculations again and found a surprisingly large sensitivity
  
  of the calcium carbonate dissolution rates to alkalinity in the
  
  GeoMIP output. The dissolution rates are actually most sensitive to
  
  mean age and next-most sensitive to alkalinity, not salinity. This means
  
  that the CMIP6 output-based analysis is an approximation because it
  
  assumes zero alkalinity change, whereas the GeoMIP output-based anal-
  
  ysis is an approximation because it assumes only changes in mean age
  
  greater than 300 years. Near analysis is credible near the surface, particularly
  
  in subtropical regions and the subpolar North Atlantic Ocean,
  
  according to Wu et al. (2025) so excluding these regions (where waters
  
  tend to be younger than 200-300 years old), the GeoMIP-based analysis
  
  should be fairly accurate. An estimate of the alkalinity change for the
  
  CMIP6-based analysis is difficult to derive from the GeoMIP output,
  
  but we have attempted to do this using the relationship between salinity
  
  and alkalinity shown above. Estimates of the calcium carbonate dissolution
  
  rates using the CMIP6 output with these alkalinity changes are now
  
  shown in the manuscript as well.
  
  • 5. Incomplete GeoMIP analysis Only 2 of the 6+ GeoMIP scenarios include Carbon-
  
  14 output to estimate mean age changes (line 76-77). Since mean age is identified
  
  as one of the two dominant factors, the comparisons of geoengineering effectiveness
  
  (e.g., Figure 7, conclusions) are incomplete and potentially misleading. The conclusions
  
  about which geoengineering techniques are ”most effective” should be heavily
  
  caveated or removed.
  
  • Using Carbon-14 to get a mean age perturbation was included because
  
  you are right to suggest emphasizing the caveat that I do not have sufficient
  
  data to propertly characterize the mean age perturbation in the
  
  GeoMIP simulations. I tried to explain this caveat but will edit the
  
  text to more heavily stress this caveat. Here we show that there are
  
  differences in the CMIP6-based dissolution rate estimates when mean
  
  age changes in waters younger than 200-300 years old are included versus
  
  excluded, but these differences are insignificant. The magnitude of
  
  the differences are large in the subtropical North Pacific due to large
  
  uncertainties, but otherwise are small. The difference globally is 15.9%.
  
  Part of the purpose of the phase space figures is to show how sensitive
  
  the GeoMIP-based estimates are to the mean age. Additionally, we have
  
  attempted to account for the uncertainty in mean age in the GeoMIP
  
  calculations by propagating the estimated 15.9% bias in the final mean
  
  ages as an uncertainty.
  
  • 6. Unquantified implications The manuscript repeatedly defers quantitative analysis:
  
  ”Future studies will investigate the direct and indirect impacts of these changes
  
  more quantitatively” (line 298-299) ”we don’t know the full range of possible calcium
  
  carbonate dissolution rates” (line 290-291) ”it is beyond the scope of the present
  
  study to quantify these effects” (line 249-250) If the implications cannot be quantified,
  
  the ”so what?” question remains unanswered. What is the practical significance
  
  of a factor-of-two or order-of-magnitude change in dissolution rates? How does this
  
  translate to Pg C or mm of sea level?
  
  • I tried to quantify these numbers but could only arrive at very approximate
  
  numbers based on previous studies’ findings and theories. For
  
  example, the back-of-the-envelope estimate of 84 mm of sea-level contraction
  
  over 150 years would become about 84 mm of additional sealevel
  
  contraction by 2100 (double the dissolution rate but half the time).
  
  But the amount of sea-level change will also depend on the spatial dis-
  
  [See supplement for Figure 3]
  
  tribution of heat and freshwater in the ocean over the present-to-2100
  
  time period, which in turn will also influence the carbon budget, and the
  
  carbon budget is difficult to predict because we don’t know what the surface
  
  fluxes of carbon dioxide will be (let alone the surface fluxes of heat
  
  and salt) from our estimates. I would need to run a coupled sedimentocean
  
  biogeochemically active future climate simulation in order to do a
  
  more proper job. One purpose of this manuscript is to stress that these
  
  simulations need to be done because of how much the dissolution rate
  
  uncertainties propagate into sea-level/carbon budget uncertainties.
  
  • 7. Section 4 – suspended sediments This section appears to be a separate mini-study
  
  that does not connect meaningfully to the main dissolution rate analysis. Specific
  
  issues: The assumption that TSM equals SSC and that all suspended matter is
  
  sediment (line 210-214) is crude. The connection between SSC in river deltas and
  
  open-ocean CaCO3 dissolution is unclear. The ”back-of-the-envelope” calculations
  
  (line 213) are insufficiently rigorous for a peer-reviewed manuscript. I recommend
  
  either removing this section or substantially expanding it with proper methodology
  
  and clearer connection to the main analysis.
  
  • This section has been removed from the manuscript.
  
  • 8. Figure quality and presentation Many results are described as ”not shown”
  
  (e.g., lines 120, 130, 145). If these results are important enough to mention, they
  
  should be included in supplementary material. Figure 11 presents ranges across all
  
  biomes and scenarios but is difficult to interpret. Consider reorganizing or splitting
  
  into multiple panels. The biome numbering system (#1-#10) requires readers to
  
  constantly refer back to figure captions. Consider using abbreviated names instead.
  
  • The biome numbers have been replaced with basin abbreviations or removed
  
  altogether on the figure. I also now include figures in supplementary
  
  material instead of saying they are not shown.
  
  • 9. Writing clarity The manuscript is difficult to follow in places: The Methods
  
  section (Section 2) is too brief and does not clearly explain how perturbations are
  
  applied. The transition between CMIP6 results (Section 3.1) and GeoMIP results
  
  (Section 3.2) is abrupt. The logic connecting dissolution rates → carbon budget →
  
  sea level is not clearly laid out.
  
  • Thank you for these suggestions. I have edited the text accordingly.
  
  • Technical Corrections 1. Line 217: Inconsistent minus sign format: ”kg m˘3” should
  
  be ”kg m³” (appears multiple times) 2. Reference list: Check formatting consistency
  
  (e.g., some entries have issue numbers, others do not) 3. Line 20: ”calcium
  
  dissolution rates” – missing ”carbonate”? 4. Equation 2: The notation ”ln(1)”
  
  equals zero, so this term appears unnecessary. Please clarify or simplify.
  
  • I have incorporated edits regarding the other specific suggestions, thanks.
  
  Reply
  
  Citation: https://doi.org/10.5194/egusphere-2026-2679-CC1

David Samuel Trossman

Viewed

Total article views: 297 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
211	71	15	297	12	12

HTML: 211
PDF: 71
XML: 15
Total: 297
BibTeX: 12
EndNote: 12

Views and downloads (calculated since 21 May 2026)

Month	HTML	PDF	XML	Total
May 2026	157	45	10	212
Jun 2026	50	21	4	75
Jul 2026	4	5	1	10

Cumulative views and downloads (calculated since 21 May 2026)

Month	HTML	PDF	XML	Total
May 2026	157	45	10	212
Jun 2026	50	21	4	75
Jul 2026	4	5	1	10

Viewed (geographical distribution)

Total article views: 282 (including HTML, PDF, and XML) Thereof 282 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 13 Jul 2026

Short summary

Many marine species' exteriors and even sediment in the ocean is made up of calcium carbonate. This is the first study that examines the most important factors that determine the dissolution rates of calcium carbonate and the range of changes across future climate scenarios. We also examine how this could potentially impact sea level in addition to the known impact on the carbon cycle.


Total:	0
HTML:	0
PDF:	0
XML:	0