the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 May 2026
High-resolution reconstruction of the pH-upregulation and its seasonal drivers in the temperate coral Cladocora caespitosa
Abstract. Ocean acidification (OA) and associated changes in seawater carbonate chemistry, combined with thermal stress, hampers coral calcification. By upregulating pH and dissolved inorganic carbon, corals can optimize their calcification, giving them some resilience to OA. Little is known about the seasonal- and interannual‑scale impacts of thermal stress and OA on pH upregulation and calcification in the temperate coral Cladocora caespitosa, despite it being the only zooxanthellate reef builder in the Mediterranean Sea. δ¹¹B and B/Ca were determined seasonally in C. caespitosa skeletons from two NW Mediterranean sites to reconstruct the effect of seawater temperature and pH on the carbonate chemistry of the coral calcifying fluid (CF), at a bimonthly resolution from June 2013 to August 2017 (Columbretes Islands, Spain), and June 2016 to February 2022 (Villefranche-sur-Mer, France). Cladocora caespitosa displayed a similar pH upregulation strategy to most tropical corals, albeit with an apparently lower sensitivity to seasonal environmental change. Temperature was the main driver of seasonal variability in the CF composition and coral calcification, with seawater pH having a comparatively lower seasonal variability, and acting on longer timescales. While longer coral records and investigations into inter-population variability would still be beneficial in order to fully understand the response of C. caespitosa to environmental change, our records constitute an important first step in understanding the biomineralization strategy of this ecologically important coral species.
- Preprint
(2120 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (extended)
- RC1: 'Comment on egusphere-2026-2720', David Evans, 15 Aug 2026 reply
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 193 | 89 | 27 | 309 | 24 | 16 |
- HTML: 193
- PDF: 89
- XML: 27
- Total: 309
- BibTeX: 24
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript "High-resolution reconstruction of pH upregulation and its seasonal drivers in the temperate coral Cladocora caespitosa" by Vergotti et al. presents boron isotope and trace element measurements of specimens from two locations in the Mediterranean, which the authors use to reconstruct the carbonate chemistry of the calcification site of this organism. This is of interest as C. caespitosa is the only species of reef-building coral in the Mediterranean, such that understanding the sensitivity of the calcification process to environmental change and environmental variability is a priority. The authors do so using both bulk (solution) and an in situ (femtosecond laser ablation) approach to boron isotope measurement, combining the two to demonstrate the accuracy of the laser ablation data and leveraging the spatially-resolved nature of the laser results to determine seasonal changes in calcification site pH, DIC, and Omega, thus enabling these to be linked to seasonal changes in pH and temperature via trace element-derived individual corralite age models. Ultimately, the data show that this coral appears to maintain an approximate constant calcification site Omega despite large seasonal changes in ambient pH and temperature, which the authors discuss within the context of possible mechanisms of carbon transport to the calcification site.
This thoughtfully constructed manuscript will undoubtedly be of interest to the community and presents a novel and interesting dataset, adding substantially to the small amount of published geochemical information available for this species. I have a few suggestions for the presentation and discussion of the results, especially the impact of temperature versus carbonate chemistry on the calcification site-seawater pH offset and the interpretation of the full range of variance in the laser data, but these are minor in nature and should not be difficult to address - my congratulations to the authors on a really nice piece of work!
- Separation of temperature and upregulation. There is no seasonal d11B signal in (e.g.) the Columbretes data between 2015-2017 or in the Villefranche data in 2020-2021, yet there is nonetheless a clear seasonal pHcf signal reconstructed from these data. This must mean that the seasonal pHcf reconstructions are largely driven by something other than active pH regulation (presumably temperature) which is a really interesting result that could be more clearly stated. E.g., the text on lines 334-336 notes that the seasonality in measured and reconstructed pHcf 'mostly agreed' but this is not surprising if it is in part a temperature effect, and if this is correct, it could be introduced and rephrased in these terms. The impact of temperature on pH is mentioned towards the end of the discussion so no major change is required here, other than to consider rephasing and making the link clearer to the reader earlier in the results and discussion (e.g. by describing the impact of temperature on pKB and pH).
- Sampling strategy and biases. The discussion starts with a comparison of solution versus fsLA, which is great, but I was left wondering about the degree to which analytical noise in the laser data might bias the comparison (please see a further comment below). Is it possible that some of what you interpret as pHcf variability is analytically derived? And could this be tested by using the variance in the standard measurements to determine the probability of obtaining a dataset with a greater degree of apparent heterogeneity compared to the solution measurements?
On a similar note, the disadvantages of bulk sampling are stated (again, great) but similar issues could be raised with spatially resolved data depending on the sampling approach. E.g., tracking the laser over a centre of calcification or other structural feature could result in a d11B excursion that has little to do with environmental variability. This could be noted/addressed at the relevant point in the discussion, including whether such structural features are present in the transects and if so, whether they could be/were avoided.
- I wonder if the first discussion section, focused on the reconstruction seawater pH from coral d11B, misses a really interesting implication of the dataset. While the previous calibration of Trotter et al. (2011) aligns closely with the data presented here on average, that pHsw reconstructed using in situ d11B results in a greater degree of seasonal variability than observed presumably tells us either: i) that the seawater-calcification fluid DpH slope differs depending on whether the coral experience different pH seasonally or on average throughout their life, and/or ii) that seawater pH variability between sites or across seasons is driven mainly by additional factors such as temperature rather than active regulation (please see my earlier comment). This is alluded to in Sec. 4.3, but it might be worth discussing the observation explicitly in these terms, and perhaps the discussion could begin with this observation and/or combined with my second comment above.
Minor comments
- line 51. Change 'levels of...' to something like 'relatively high alkalinity concentration'. More broadly, it isn't clear how the second part of the sentence follows from the first. Given all else being equal, TAlk would have almost no effect on the relationship between pH and pCO2, so there must be another cause of the higher than average rate of acidification (likely the higher than average T increase).
- lines 89-90. It might be worth noting that this is not unique to C. caespitosa and also the case for many other species.
- lines 100-101. And presumably, no in situ studies that avoid complications such as acclimatisation to aquarium conditions etc.
- lines 103-107. There is no problem with this paragraph but it could be rephrased for stylistic purposes. E.g., I'm not sure the last sentence is necessary at the end of the introduction (it would fit better in the abstract or conclusions) and the first two sentences could be combined. Either way, line 106 needs to be edited to 'corals' ' or ideally something like 'the growth of temperate corals'
- line 113. State the approximate NaOCl concentration before it was diluted with DI.
- lines 118-119. Rephrase to avoid using X-ray as a verb.
- Section 2.2. Please add or comment on the following details if possible:
- What is the magnitude of diurnal pH variation at Columbretes Islands (to give a sense of whether the choice of time is important - I imagine not)
- How was the pH data calibrated/state some estimate of data quality and how it was determined
- line 142. The unit of resistivity is MOhm cm.
- lines 151-152. I didn't understand how this laser path geometry aligns with what is written about the beam diameter a few sentences below. If the laser spot size was 20 um, why were the paths 50 um wide? Please clarify.
- Line 156. Please also state representative sensitivity and gas blank voltages.
- Line 167. Please clarify whether the in-house pressed pellets were 'nanopellets' (Garbe-Schoenberg & Muller, 2014) or pressed powders with a larger grain size.
- line 174 onwards. If I understand, this implies that all standards were used to generate the Dd11B-11/10.066 ratio with the exception of the JCp-NP, is that correct? If so, I would suggest taking one of the standards out of the calibration line in order to give an estimate of data quality, or otherwise please state data quality (accuracy and precision) and how it was determined.
- Section 2.3.2. I appreciate the data given in Table B5, although some of the measured values are substantially offset from the reported values for this material, which is a bit surprising given that it is a nanopellet standard. It would be good to report the percentage accuracies in the methods section if possible and to briefly comment on whether the relatively large B/Ca inaccuracies are a source of concern when interpreting the sample data. You could also compare to the NP data reported in Jochum et al. (2019) & https://www.nature.com/articles/s41597-026-07812-0 - a bit circular in the first case given these were measured using fsLA, but it would give a sense of whether the data quality is driven by the standard versus analytical approach.
- Line 209. I think the correct term is 'intermediate precision', but this may depend on whether the data were measured over multiple analytical sessions (which intermediate precision would imply).
- Line 214. It would be great to show the age models in the appendix or a supplement, including the details of how these were pieced together using different elemental ratios.
- Line 283. 'reconstructed pHsw' should not appear here if I understand as you're (I think) comparing reconstructed calcification site pH to seawater measurements, rather than reconstructing seawater pH.
- Figure 3. Given that you've gone to the effort of reconstructing individual corralite age models and making spatially resolved measurements, it seems a shame not to show the complete dataset on this figure. I know that further details are given in subsequent sections/figures, but this could be a two panel figure showing both the corralite averages as well as the seasonal/monthly average pHsw and d11B data, which would give an immediate sense of whether seasonal pH change impact pHcf (perhaps tying into my main comment above).
- Figure 3. Please also state which conditions the reference curve is shown for (T,S).
- Lines 308-309. This is interesting, although to interpret this in terms of a larger range in pHcf identified in Fig. 4 requires the analytical noise in the laser measurements to be accounted for. Was an attempt made to take this into account?
- Figure 4. It may simply be that the in situ data align in a slightly unintuitive way, but I struggled to understand why the curves in the upper and lower panels of Fig. 4 do not look more similar. For example, solution d11B is approximately constant between 2015 and 2016 while there is a relatively large pHcf decrease in the middle of the year (presumably driven by the effect of T on pKB?). However, the lowest laser d11B data point has been corrected upwards in terms of pHcf, which isn't visible in the solution data. Is this correct?
- Figure 5. Please correct the y axis labels - B/Ca in panel (c) should not contain a cf subscript and Omega (e) is unitless. You could also give the pH scale. Note the incorrect (b) on line 320 too
- Figure 5. I would suggest changing 'reconstructed records...' to simply 'reconstructed pHcf).
- Lines 412-414. I appreciate that references are provided but wonder if this confuses cause and effect - perhaps they grow more slowly because they are less able to maintain the cf carbonate chemistry that they need to do so.
- Lines 417-420. I think this interpretation might be a bit simplistic. What you show is that this species maintains an approximately constant calcification site Omega through the year, so the DIC data could be interpreted to suggest that a lower degree of CO2 diffusion/transport is required when the pH is higher. These processes of course do not take place in isolation however, a greater pHcf and therefore steeper pH gradient to the cytosol will impact CO2 diffusion, which will result in a higher Omega if pH can be maintained and a lower Omega if not. Overall it is perhaps more likely that the organism prioritises maintaining Omega in the face of seasonally varying light, temperature (impacting pH and metabolism), and carbonate chemistry, in which case DIC increase responds to, rather than drives, the trends you see.
- Lines 432-433. Please see my earlier comment - it is perhaps more likely that they grow more slowly because they are not able to maintain a higher cf Omega.
- Line 438. 'products required for calcification'? I'm not sure I understand though. Aragonite precipiation is ultimately driven by cf carbonate chemistry (essentially the same between the two populations) and the only way to decouple this from growth rate on longer timescales is to cycle more seawater. So, the population may be more efficient in that they can do more with the same resources, if indeed the resources are the same (food, nutrients etc. - are they?), but not that they can use the products of or required for calcification more efficiently.
Typos/phrasing
- line 162. 'the signal'
- line 170. 'Section 2.3.3'
- line 315. Delete 'composition'
- line 344. The reference to panel b is incorrect and perhaps replace 'soft' with 'gradual'
- line 347. Multiplication symbols needed
- line 375. Change 'in' to 'using'.
- line 377. 'indicating that the measurements were accurate'