the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Aug 2025
Characterisation and quantification of organic carbon burial using a multiproxy approach in saltmarshes from Aotearoa New Zealand
Abstract. Blue carbon ecosystems, such as saltmarshes, play a crucial role in sequestering atmospheric carbon dioxide by storing it as buried organic carbon, also known as blue carbon, for centuries to millennia. This has generated significant interest in restoring these ecosystems for climate change mitigation benefits. While international methodologies exist for generating blue carbon credits through coastal wetland restoration, their application in Aotearoa New Zealand is limited by a lack of data on saltmarsh carbon stocks and accumulation rates. Additionally, to improve carbon mitigation estimates, research is needed to better understand the sources, composition and preservation of organic carbon in saltmarshes and the factors influencing its long-term storage. This study quantifies these metrics at five saltmarsh sites in Aotearoa New Zealand using 45 sediment cores analysed for elemental composition, stable isotopes, X-ray fluorescence (XRF), lipid biomarkers and Ramped-Pyrolysis Oxidation-Accelerator Mass Spectrometry (RPO-MS) in combination with Pyrolysis-Gas Chromatography-Mass Spectrometry (Py-GC-MS). Results show high variability in soil organic matter properties, carbon stocks (40.7 ± 9.1 to 112 ± 100.3 Mg C ha-1), and accumulation rates (0.56 ± 0.23 to 2.5 ± 0.44 Mg C ha-1 yr-1). An initial assessment indicates increased carbon accumulation following restoration at two sites. Stable isotope and lipid biomarker results show substantial contributions from saltmarsh vegetation to the organic carbon pool. Preliminary analysis suggests long-term preservation of plant-derived organic carbon in the oldest basal soil samples. The findings highlight the importance of accounting for spatial variability within saltmarsh ecosystems in carbon assessments and underscore the need for further research to determine environmental factors influencing long-term carbon storage and preservation.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Biogeosciences.
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.- Preprint
(3117 KB) - Metadata XML
-
Supplement
(3470 KB) - BibTeX
- EndNote
Status: open (until 19 Sep 2025)
- RC1: 'Comment on egusphere-2025-2949', Anonymous Referee #1, 23 Aug 2025 reply
-
RC2: 'Comment on egusphere-2025-2949', Anonymous Referee #2, 14 Sep 2025
reply
Albot et al., provide a new dataset and exploration of saltmarshes from across New Zealand exploring the quantity and composition of carbon in these system. Saltmarsh carbon data from New Zealand is rare and has been absent from global studies, making this study especially important.
The quantity and quality of the data presented in the manuscript is highly commendable as it could be easily split into several high quality papers. The density of data in the manuscript results in it being hard to follow in places, the is a decision for the authors but splitting this in to two papers would be possible and would not reduce the impact of the research.
Overall the manuscript is written well, but reads more as a thesis chapter opposed to a research article, I would suggest taken time to remove some of the unnecessary information presented.
Introduction
The introduction needs to restructured, though all the information is present in the text, the text is overloaded with unnecessary information that the manuscript does not tackle.
Methods
Line 141 - Remove grinder
Line 141 - elaborate on how - Large roots and aboveground biomass were removed
Line 142 - The samples were not size fractioned as this research focuses on bulk soil OC - in the previous sentence it was stated that large roots were removed.
Line 141/142 - Are the sentences ordered correctly did you mill the sample then remove the roots, this seem back to front.
Line 144 - change irMS to IRMS
Line 160 - Lead isotope data - state which isotopes
Line 160 - 177 - Does slicing the cores at 2cm intervals impact the quality of the radionuclide data, would doing it at 1cm resolution be more useful.
Remove section 3.4
Lines 223 - 227 - remove these line they are not required.
Should section 3.6.2 and 3.6.3 be switched as the RPO analysis uses the Py-GCMS data (line 253)
Section 299 - you do not need to explain what a PCR is, taken a more direct approach will shorten and improve the manuscript.
Figure 2 - in the methods you state the troel smith classification scheme is used, can you outline how this aligns with the soil descriptions.
Section 4.2 - the presentation of the data in the text is not required as it present in table 1.
Figure 3 - could the plot be placed side by side.
Figure 4 - the outputs from the Rplum model are not the easiest to understand can you make clear which one was produced by gamma vs alpha spectrometry.
Section 4.5 - the use of detrending is interesting but the utility of the method to the manuscript has whole is questionable, I would move this to the sup mat.
Figure 7 - As d15N data has been produced, did improve source estimation.
Section 4.7 - this section does not provide any results, could you remove or provide detail on what was measured.
Section 4.8.1 - move the equation and ratio discussed to the methodology.
Section 4.9 - could the thermograms be displayed.
Discussion
The discussion in generally written well, but as with other section there is a significant amount of unnecessary text, if this could be cut down the manuscript would be much more readable. The above comments concerning the results should inform the discussion.
Again I would like to state that the data in this manuscript is of high quality and the interpolation is well done. However the current structure and dense text struggle to communicate the importance of the study. There is unnecessary text and data that can be removed. I would also ask the authors to consider splitting this into two papers - 1) stocks and accumulation, 2) source
Citation: https://doi.org/10.5194/egusphere-2025-2949-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 939 | 53 | 14 | 1,006 | 22 | 27 | 23 |
- HTML: 939
- PDF: 53
- XML: 14
- Total: 1,006
- Supplement: 22
- BibTeX: 27
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript entitled “Characterisation and quantification of organic carbon burial using a multiproxy approach in saltmarshes from Aotearoa New Zealand”, by Albot et al. presents an impressive body of work, using multiple approaches to characterise and quantify carbon sources, stocks, and accumulation rates in several saltmarsh systems in New Zealand. I want to highlight and commend the authors for the amount of work that went into generating this dataset and preparing the manuscript. A few minor comments are noted below.
There is a mismatch between Figure 1 and the text / supplementary S1: In Figure 1B, there are 10 core locations, but only 9 mentioned in the text. Further, from the supplementary table S1 it seems that Puk-LM4 was omitted in favour of Puk-LM1 or 2, which are located closely together. Is there a reason why this core was excluded and why is it on the figure if not used in the analysis?
Lines 139 to 141: You explain methods for DVD in two places [here coma and in line 179] which is confusing. You can combine them in the later section for clarity.
Line 179: I assume the DVD was calculated before freeze-drying. Otherwise, this sentence is confusing.
Lines 278 to 284: You specify the tests used when normality and equal variance assumptions were not met. What did you use if they were met [or you can specify here if they were all non-normal]? And what was your model structure? Did you test differences across the different sites or any other relationships between variables? Please give more details on the statistical methods.
Just a note: It is likely that the variables won't be normally distributed as they are coming from a very diverse set of sites how. However, for an anova it is more important to test for the distribution, heteroscedasticity, and independence of the model residuals rather than the underlying variable distribution.
Lines 302 to 306: There is no need to explain here what a PCA is it overloads the already heavy method section. I suggest removing these lines from the manuscript.
Lines 298 to 320: Was PCA performed on all data sets combined or separately? It is not clear from this description what was done here and why. From the results section (lines 526 and following) it seems that PCA was performed on 2 data sets (one without and one with lipids), and on the cores that were age dated (is that correct?). It is unclear as to why these separations were chosen.
Lines 351 to 352: This relates to the supplementary Table 2: you should provide more information on the statistical analysis than just the p-values. At least the test statistic in combination with the degrees of freedom should also be included.
Lines 353 to 354: You don't need to specify the units in the caption if you have them in the table. However, you should explain the acronyms.
Figure 3: I suggest using a different plot type to visualise the carbon stocks at each site and in the different zones. Bar plots are not meant to visualise continuous data like this. An option could be to plot means and standard error together with the spread of the data [points], to use box plots [with a marker for the mean], or violin or dot plots. This comment has no impact on my recommendation for the article to be accepted but should be seen as advice.
Figure 5: see my comment on bar plops above.
Lines 440 to 480: the equation and explanations of indices should go in their respective methods section. The descriptions of what they indicate may stay in their results if preferred.
Lines 527 and 544: Headings in this section are almost the same. I would suggest changing them to make the differences between the sections easier to grasp.
Lines 534 to 536: Do you have documentation on the cluster analysis? It would be informative to show the criteria of the split and the sample groupings in a supplementary table.
Line 608: Relying too heavily on the mean value here may be misleading, as it is strongly influenced by that one core that hit a peat-bed underneath the sediment layer. It may be better to use the median or at least mention it here. That was also by far the longest core. It is probably worth discussing how this impacts you results (especially since this is the core you used for further analyses).
Lines 630 to 632: This sentence explaining why SL rise is unnecessary.
Lines 662 to 665: Above, you state that C3 plants of terrestrial origin account for a large fraction of the organic material in the marsh sediments but here you say that the marsh plants themselves are responsible for the C3 signature. Does that mean that you are not able to tell the difference between allochthonous terrestrial inputs and autochthonous salt marsh production using the stable carbon isotope tracer? If correct, this limitation should be discussed as well.
Lines 768 and following: You used an impressive and resource consuming number of methods to get to the conclusions presented here. I wonder if it would be of benefit to the wider community to comment on a more feasible approach for a scalable solution for blue carbon assessments. E.g., which methods are strictly necessary, and which give important information from a scientific perspective but are less practical to apply?
Lines 793 to 794: It doesn’t make much sense to give uncertainty estimates for a range. I get what you mean (I think), which is that you show the lowest and the highest site means with standard deviation (?) but in that case you should say that. This already threw me in the Abstract where the same case applies.