the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Jan 2024
Seafloor sediment characterization to improve estimate of organic carbon standing stocks in continental shelves
Abstract. Continental shelf sediments contain some of the largest stocks of organic carbon (OC) on Earth and play a vital role in influencing the global carbon cycle. Quantifying how much OC is stored in shelf sediments and determining its residence time is key to assessing how human activities can accelerate the process of OC remineralization into carbon dioxide. Spatial variations in terrestrial carbon stocks are well studied and mapped at high resolution, but our knowledge of the distribution of marine OC in different seafloor settings is still very limited, particularly in the highly dynamic and spatially variable shelf environments. The lack of knowledge reduces our ability to understand and predict how much and for how long oceans sequester CO2. In this study, we use high-resolution multibeam echosounder (MBES) data from the Eastern Shore Islands offshore Nova Scotia (Canada), combined with OC measurements from discrete samples, to assess the distribution of OC content in seafloor sediments. We derive three different spatial estimates of organic carbon: i) assuming a homogenous seafloor the carbon stock estimates were scaled to the entire study region; ii) using a high-resolution substrate map, the estimates were scaled to the areas of soft substrate only, and, finally, iii) using Empirical Bayesian Regression Kriging (EBRK) regression prediction within the area of soft substrate, carbon stock estimates in areas of soft substrate were refined to account for spatial variability in the concentration of OC. These three distinct spatial models yielded dramatically different estimates of average standing stock of OC in our study area, 1275, 259 and 203 Mt of OC respectively. Our study demonstrates that high-resolution mapping is critically important for improved estimates of OC stocks on continental shelves, and to the identification of carbon hotspots that need to be considered in seabed management and climate mitigation strategies.
- Preprint
(1405 KB) - Metadata XML
-
Supplement
(65 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-5', Anonymous Referee #1, 12 Feb 2024
Brenan and co-workers present a case study that explores the utility of detailed seafloor substrate maps when estimating sedimentary organic carbon stocks in continental shelf settings. They first develop a binary substrate model (hard vs soft substrate) based on video imagery and high-resolution (2 m pixel size) predictor data fed into a random forest algorithm. Subsequently, they predict organic carbon stocks within areas of soft substrate based on a limited number of physical seafloor samples using empirical Bayesian regression kriging. The results are compared with two scenarios where the organic carbon stock is derived by scaling up average organic carbon stocks (mean of collected samples) across either the whole study site (scenario 1) or across the soft substrate area (scenario 2). They find strong differences in the stock size depending on the method of upscaling.
There is a renewed interest in organic carbon stored in seafloor sediments as it has been suggested that human activities such as demersal fishing might lead to the release of large amounts of aqueous CO2, which could partly reach the atmosphere and increase global warming. Organic carbon stock estimates are a prerequisite for such studies. Most studies that have been conducted in recent years are at regional to global scales, while this study addresses local scale variability. Such high-resolution studies are likely more relevant should the designation of marine protected areas be considered as a measure to protect organic carbon stored in seafloor sediments. Studies like the one presented here are therefore timely. However, the presented work has several shortcomings which I will discuss below. In summary, this leads me to recommend a reconsideration of the manuscript after major revisions.
Specific comments:
The main issue is the calculation of the organic carbon stocks. The authors report a total stock of 203 million tonnes in an area covering 223 km2 of seafloor according to their most refined upscaling approach (scenario 3). Compare this with results in Diesing et al. (2021), who estimated 231 million tonnes of organic carbon in the North Sea and Skagerrak (558,000 km2) or Smeaton et al. (2021), who estimated 524 million tonnes in the United Kingdom’s Exclusive Economic Zone (744,000 km2). The stock estimate in the study of Brenan et al. is on the same order of magnitude as the other two studies despite an area three orders of magnitude smaller. It would therefore appear that the estimate is too high and that there is an error in the calculations. The error can be found in equation 4, where arcsine transformed organic carbon contents are used to calculate stocks. Stock calculations should instead be done with untransformed organic carbon contents. Transformation of the data can be advisable when spatially interpolating or predicting organic carbon content. However, the results need to be back-transformed prior to stock calculations (equation 2 in Diesing et al., 2017 or equation 8 in Smeaton et al., 2021). Recalculating organic carbon stocks based on the data provided in the supplement, I got 80,901 t for scenario 1 and 16,437 t for scenario 2. I would advise the authors to calculate stocks using the untransformed organic carbon data and then assess whether a transformation is necessary for spatial prediction. They could then run the spatial predictions with transformed data and back-transform the results to get organic carbon stocks.
I find the choice of scenarios (upscaling methods) a bit artificial. Given the progress we have seen in recent years, I think (or at least hope) nobody would simply upscale the mean stock values based on measurements to a whole site (scenario 1). I suggest linking the scenarios to the methods discussed in the introduction, i.e., kriging without regression or external drift (as a new scenario 1) and upscaling based on average stocks per sediment class (similar to scenario 2). In my opinion, this would be more informative and better link up with the introduction which, among other things, summarises the evolution of the mapping methods.
The discussion is very short, which in itself is not necessary a bad thing, but I think that it leaves out opportunities. For example, the authors could discuss the impact of coarse-grained sediments on OC stock calculations more generally. All marine studies published so far calculate stocks by multiplying organic carbon content with dry bulk density and the sediment depth interval that is considered. In this study, organic carbon stocks are only calculated for soft substrates (sand, silt, and clay), while it is assumed that hard substrates do not contain organic carbon (in scenarios 2 and 3). It might be worth comparing these two contrasting approaches with the approach taken in terrestrial soil mapping, which accounts for the content of coarse fragments (>2 mm grain diameter) when calculating stocks (Hengl et al., 2014; Poeplau et al., 2017). Is this an aspect that the marine community has so far overlooked?
The first paragraph of the introduction sets out to define Blue Carbon. However, I find this section not particularly clear. In fact, there are two definitions of Blue Carbon given. The first one is extremely wide including all inorganic and organic carbon stored in the ocean. Conversely, the second definition of the IPCC is much narrower and aligned with the ‘classical’ definition of Blue Carbon, which only considers vegetated coastal ecosystems that are actionable. I strongly suggest revising this first chapter, so that it becomes clear what is considered Blue Carbon in the context of this study. Some suggested literature: Lovelock and Duarte (2019); Howard et al. (2017, 2023)
The sampling design could be explained in more detail. In particular, it is unclear to me how a stratified sampling design, which requires some form of segmentation of the area into more or less homogeneous areas can be based on the backscatter mosaic, which is continuous. Was the backscatter data categorised and if so, how?
The methods section could benefit from a short paragraph that summarises the research strategy incl. a flow diagram. This would provide the reader with a better overview of the methodology right from the start.
Figures 1, 2 and 8 use a rainbow style colour palette. The use of such a colour scheme is generally discouraged. Please see Crameri et al. (2020) for advice on choosing a suitable colour scheme.
Technical corrections:
Please see comments in an annotated version of the manuscript.
References
Crameri, F., Shephard, G. E., and Heron, P. J.: The misuse of colour in science communication, Nat Commun, 11, 5444, https://doi.org/10.1038/s41467-020-19160-7, 2020.
Diesing, M., Thorsnes, T., and Bjarnadóttir, L. R.: Organic carbon densities and accumulation rates in surface sediments of the North Sea and Skagerrak, Biogeosciences, 18, 2139–2160, https://doi.org/10.5194/bg-18-2139-2021, 2021.
Hengl, T., de Jesus, J. M., MacMillan, R. A., Batjes, N. H., Heuvelink, G. B. M., Ribeiro, E., Samuel-Rosa, A., Kempen, B., Leenaars, J. G. B., Walsh, M. G., and Gonzalez, M. R.: SoilGrids1km — Global Soil Information Based on Automated Mapping, PLoS One, 9, e105992, 2014.
Howard, J., Sutton-Grier, A., Herr, D., Kleypas, J., Landis, E., Mcleod, E., Pidgeon, E., and Simpson, S.: Clarifying the role of coastal and marine systems in climate mitigation, Front Ecol Environ, 15, 42–50, https://doi.org/https://doi.org/10.1002/fee.1451, 2017.
Howard, J., Sutton-Grier, A. E., Smart, L. S., Lopes, C. C., Hamilton, J., Kleypas, J., Simpson, S., McGowan, J., Pessarrodona, A., Alleway, H. K., and Landis, E.: Blue carbon pathways for climate mitigation: Known, emerging and unlikely, Mar Policy, 156, 105788, https://doi.org/https://doi.org/10.1016/j.marpol.2023.105788, 2023.
Lovelock, C. E. and Duarte, C. M.: Dimensions of Blue Carbon and emerging perspectives, Biol Lett, 15, 20180781, https://doi.org/10.1098/rsbl.2018.0781, 2019.
Poeplau, C., Vos, C., and Don, A.: Soil organic carbon stocks are systematically overestimated by misuse of the parameters bulk density and rock fragment content, SOIL, 3, 61–66, https://doi.org/10.5194/soil-3-61-2017, 2017.
Smeaton, C., Hunt, C. A., Turrell, W. R., and Austin, W. E. N.: Marine Sedimentary Carbon Stocks of the United Kingdom’s Exclusive Economic Zone, Front Earth Sci (Lausanne), 9, 50, https://doi.org/10.3389/feart.2021.593324, 2021.
-
AC1: 'Reply on RC1', Craig Brown, 04 Apr 2024
We thank reviewer 1 for their insightful, thorough, and careful review of the manuscript. Below, we provide a response to each point that was raised.
Referee #1
Referee comment: The main issue is the calculation of the organic carbon stocks. The authors report a total stock of 203 million tonnes in an area covering 223 km2 of seafloor according to their most refined upscaling approach (scenario 3). Compare this with results in Diesing et al. (2021), who estimated 231 million tonnes of organic carbon in the North Sea and Skagerrak (558,000 km2) or Smeaton et al. (2021), who estimated 524 million tonnes in the United Kingdom’s Exclusive Economic Zone (744,000 km2). The stock estimate in the study of Brenan et al. is on the same order of magnitude as the other two studies despite an area three orders of magnitude smaller. It would therefore appear that the estimate is too high and that there is an error in the calculations. The error can be found in equation 4, where arcsine transformed organic carbon contents are used to calculate stocks. Stock calculations should instead be done with untransformed organic carbon contents. Transformation of the data can be advisable when spatially interpolating or predicting organic carbon content. However, the results need to be back-transformed prior to stock calculations (equation 2 in Diesing et al., 2017 or equation 8 in Smeaton et al., 2021). Recalculating organic carbon stocks based on the data provided in the supplement, I got 80,901 t for scenario 1 and 16,437 t for scenario 2. I would advise the authors to calculate stocks using the untransformed organic carbon data and then assess whether a transformation is necessary for spatial prediction. They could then run the spatial predictions with transformed data and back-transform the results to get organic carbon stocks.
Response to Referee comment: Thank you for picking up on the error in the calculations, we will correct the estimates by removing the arcsine transformation advised by the referee.
Referee comment: I find the choice of scenarios (upscaling methods) a bit artificial. Given the progress we have seen in recent years, I think (or at least hope) nobody would simply upscale the mean stock values based on measurements to a whole site (scenario 1). I suggest linking the scenarios to the methods discussed in the introduction, i.e., kriging without regression or external drift (as a new scenario 1) and upscaling based on average stocks per sediment class (similar to scenario 2). In my opinion, this would be more informative and better link up with the introduction which, among other things, summarises the evolution of the mapping methods.
Response to Referee comment: We agree with the referee that the upscaling method for scenario 1 may be somewhat artificial – but the goal was to demonstrate the challenges of spatial estimates of seafloor carbon in the absence of high-resolution seafloor mapping data, and where broad assumptions are made. The suggestion to introduce a krigging approach over the area based on the measurements made from the sample locations is of value, and we will incorporate this approach as an additional scenario into the paper, while also retaining scenario one.
Referee comment: The discussion is very short, which in itself is not necessary a bad thing, but I think that it leaves out opportunities. For example, the authors could discuss the impact of coarse-grained sediments on OC stock calculations more generally. All marine studies published so far calculate stocks by multiplying organic carbon content with dry bulk density and the sediment depth interval that is considered. In this study, organic carbon stocks are only calculated for soft substrates (sand, silt, and clay), while it is assumed that hard substrates do not contain organic carbon (in scenarios 2 and 3). It might be worth comparing these two contrasting approaches with the approach taken in terrestrial soil mapping, which accounts for the content of coarse fragments (>2 mm grain diameter) when calculating stocks (Hengl et al., 2014; Poeplau et al., 2017). Is this an aspect that the marine community has so far overlooked?
Response to Referee comment: We agree with the referee’s comment on adding to the discussion. We will include a section on comparing between marine and terrestrial soil mapping to emphasize the complexity and challenge accounting for coarse fragments in marine sediment.
Referee comment: The first paragraph of the introduction sets out to define Blue Carbon. However, I find this section not particularly clear. In fact, there are two definitions of Blue Carbon given. The first one is extremely wide including all inorganic and organic carbon stored in the ocean. Conversely, the second definition of the IPCC is much narrower and aligned with the ‘classical’ definition of Blue Carbon, which only considers vegetated coastal ecosystems that are actionable. I strongly suggest revising this first chapter, so that it becomes clear what is considered Blue Carbon in the context of this study. Some suggested literature: Lovelock and Duarte (2019); Howard et al. (2017, 2023)
Response to Referee comment: Thank you for addressing your concerns about the definition of blue carbon. We will examine the suggested literature and alter the definition to clarify what blue carbon means in our study.
Referee comment: The sampling design could be explained in more detail. In particular, it is unclear to me how a stratified sampling design, which requires some form of segmentation of the area into more or less homogeneous areas can be based on the backscatter mosaic, which is continuous. Was the backscatter data categorised and if so, how?
Response to Referee comment: The sampling design did consider broad regions of similar backscatter intensity in order to design the sampling survey. Further clarification will be provided in the methods sections will be provided on the sampling design.
Referee comment: The methods section could benefit from a short paragraph that summarises the research strategy incl. a flow diagram. This would provide the reader with a better overview of the methodology right from the start.
Response to Referee comment: We will add a flow diagram to the methods section to provide a better overview of the methodology.
Referee comment: Figures 1, 2 and 8 use a rainbow style colour palette. The use of such a colour scheme is generally discouraged. Please see Crameri et al. (2020) for advice on choosing a suitable colour scheme.
Response to Referee comment: We will review the suggested paper and change the rainbow style colour scheme for Figures 1, 2 and 8. However, colour ramps such as the ones used in the listed figures are fairly standard
Technical corrections: Please see comments in an annotated version of the manuscript.
Response to Referee comment: We could not see any technical corrections on the pdf?
Citation: https://doi.org/10.5194/egusphere-2024-5-AC1
-
AC1: 'Reply on RC1', Craig Brown, 04 Apr 2024
-
RC2: 'Comment on egusphere-2024-5', Anonymous Referee #2, 08 Mar 2024
1. General Comments
The authors have presented a study comparing different spatial models to estimate organic carbon stocks in seabed sediments from the Eastern Shore Islands offshore Nova Scotia, Canada. They produce a substrate map of two classes – hard / soft substrate using video imagery and a relatively small number of physical grab samples compared to study area. They then calculate organic carbon stocks using three scenarios, increasing the accuracy of the scaling method each time. Their study highlights that stock estimates of organic carbon can vary significantly depending on the type of model and assumptions used, concluding that, high resolution mapping of sediments is critical for improving estimated for sedimentary carbon stocks. This in itself is not a novel finding; however, the study is useful in making the case for seabed substrate mapping for informed seabed management, and has generated an improved spatial sedimentary carbon dataset for the Easter Shore Islands region.
The aims of the presented study are very relevant as there is increasing interest in how the ocean stores carbon. There is a particular need to understand the extent and distribution of organic carbon within marine sediments to inform stock estimates and to understand the potential climate regulation service provided by sediments. There is merit to understanding the pitfalls of scaling-up low-resolution data – particularly to allow policy-makers to make informed decisions.
However, I think there are several areas of improvement before this manuscript is ready for publication. I have provided detailed comments within the Specific Comments and Technical Corrections Sections. These mainly reflect areas that require further clarity, improved coherence, technical corrections, and refinement of discussion, including the currently lacking, limitations of the study.
Overall, the ambition of the study is welcome and I would urge the authors to consider developing this manuscript further; it currently lacks rigour and further work is required to make the results and conclusions more robust. I am therefore recommending this manuscript be reconsidered after major revisions.
2. Specific Comments
Title
The title of the manuscript should be changed to reflect the local-regional scale of the study rather than implying the results could be applicable at the scale of ‘continental shelves’.
Introduction
The paper would benefit from restructuring the introduction to provide better flow and linkages between the paragraphs, including a stronger link between MBES, sediment type and organic carbon.
The introduction sets the scene for the requirements of the research generally; however, it needs to provide a clearer rationale for how it is novel in comparison to other mapping studies for sedimentary organic carbon and what specific contribution it is making. The paragraphs do not link together particularly well and the authors are encouraged to consider the flow between sections. Crucially, the link between sediment type and organic carbon has not been made, which is essential supporting information to explain why seabed sediment mapping using MBES can yield carbon stock results. The section on marine carbon is chaotic, for instance, it jumps from marine carbon straight into different definitions for blue carbon to benthic carbon. References that are more appropriate are needed to support statements.
- Line 32 – Blue carbon is specifically about organic carbon – if this is not explicit, there is the potential for confusion with inorganic carbon too.
- Line 33 – The term disproportionate needs to reflect per unit area. Rather than the global ocean, it is about specific habitats that can store disproportionate amounts of carbon on an area-by-area basis.
- Line 36 – A more appropriate to reference McLeod et al., 2011, who was part of coining the term Blue Carbon.
- Line 38 – A more appropriate reference would be Lovelock et al., 2019 as this paper discusses how the BC term is evolving in the science and literature.
- Line 40 – A more appropriate reference is required to highlight times of accumulation and burial of sediment over 1000s of years (See papers by Berner, 2003, or Burdige, 2007 to get a longer-term overview). The currently referenced papers investigate surface sediments, which are not where the long-terms stores of carbon are found.
- Line 41 – What is the scale for this estimate? Is that one trawl or all trawls everywhere?
- Line 42 – Only one reference provided, although next sentence refers to more than one study.
- Line 45 – Could be more specific about which anthropogenic activities. Can they be characterised?
- Line 48 – MPAs have traditionally been designated for biodiversity – designation for carbon would be a novel approach?
- Line 56 – Sampling systems – do you mean physical samples or other?
- Line 57 – Can expand on the relevance of the different sediment names – i.e. increasing grain size. What classification is this?
- Line 66 – Are these early studies based on terrestrial or marine carbon?
- Line 73 – Was MBES data used in all the referenced studies?
- Line 80 - There has been no mention yet about the relationship between sediment type and organic carbon so it is not clear why the extent of bedrock would make a difference to calculations.
- Line 86 – The Hunt et al., 2021 study results were not output at 6 m resolution. Check references used are accurate representations of the points you are making.
- Lines 91-92 – Could be more explicit in why the two studies mentioned found differences. The studies were apparently in very different geographical settings as one possibility.
- Lines 96-99 – This sentence is too long and vague - lots of challenges with global estimates and is it a realistic scale for management?
- Line 100 – Further detail about why it is an Area of Interest?
- Line 101 – Question about the relevance of the setting - I understand that this location is a good setting to test the hypothesis that different sediment types have different carbon densities? Would that be true?
Study Area
This section could be strengthened by including a description/ characteristics of the Study Area that might be more relevant to sedimentary carbon. There is no description of what is known about seabed sediment type for instance.
- What is the relevance of temperature and nutrients to potential carbon stocks?
- Why is it an Area of Interest to the Canadian Government?
- Figure 1 - Would appreciate the location within a more generalised map of Canada somewhere as well to get the wider geographical context.
Methods
The method chapter is currently too vague and requires further development to include specific details for clarity (it is a little unclear what has been done and in what order) and for repeatability. Some areas for improvement include:
- Line 154 - Explain more about what focal statistics is and how it was applied.
- Line 166 - How many grab samples were taken overall?
- Line 167 - A stratified random sampling technique was used based on the backscatter, however no information has been given about what classification method was used for this.
- Line 169 - How large were the subsamples? How big was the Van Veen grab?
- Predictor variables – I don’t see that sediment type has been explicitly mentioned. E.g., Line 150 mentioned additional predictor variables but it is not clear what they are additional to.
- Line 168 – Backscatter is not always a good proxy – Could you caveat this by saying backscatter can be a good proxy for sediment grain size and perhaps add some more references for studies where this is the case?
- Line 172 - Only soft substrates have been sampled – what was used to ascertain what substrates were and were not suitable prior to sampling?
- Line 178 - Were the sediments sorted for grain size before being prepared for OC analysis? How were the sediments above 2 mm separated? Was bulk density of the sediment measured?
- Line 180 - Were samples dried from frozen?
- Line 183 – How was the coarse fraction estimated and removed? Did this result in any loss of integrity of the sample?
- Line 192 - How was it determined that the sediments had relatively low organic content?
- Line 192 - Were the coarse and fine fractions measured as % of mass or volume of the total? Was the full particle size distribution used to classify sediment type at all?
- Line 206 - Was the camera stationary for 3 minutes or was it taking a video transect?
- Line 213 – Two classifications is quite rudimentary and it should be acknowledged somewhere when interpreting the results as a potential limitation. Mixed sediment can exist which contain proportions of mud, sand and gravel; e.g., a gravel veneer on a muddy substrate, which will have a different OC content than a sandy or pure gravel.
- Line 223 – It is not clear what is being modelled - is it carbon content within the sediment? The title of the section is confusing.
- Line 253 – Are there any limitations of assuming the same single grain density value across the area? For instance, could it artificially over-inflate carbon stock estimates for muddier sediments?
- Line 254 – Check that the references are correct to support the method used.
- Line 261 - Did this step of modelling the OC content at unknown locations happen before or after the OC stocks were estimated in the previous section?
- Line 273 – What does a small RMSE value indicate?
- Figure 2 – In the map legends, the values of the parameters could be rounded up.
- Table 1 – formatting – centre the table headings
Results
The main issue here it that the calculated stock estimates do not appear correct when compared against other studies that have also used OC Content and Dry Bulk Density (e.g., Diesing et al., 2017, Smeaton et al., 2021, Hunt et al., 2020) - they are at least an order of magnitude higher and the calculations should be revisited.
- Line 275 – Is this a measurement of organic carbon content or concentration? See Flemming and Delafontaine, 2000.
- Line 277 - What framework is used for silt + clay = mud? Reference needed.
- Line 284 – Clarity needed – The results from Figure 4 suggests to me that % mud is a good proxy for OC - not sediment type. Sediment type is a classification based on the total composition of grain sizes.
- Line 309 – Are these statistics for across the whole area? Or for the grab samples? What useful information does this general distribution provide?
- Line 325 – Can you expand on the significance of the accuracy of the interpolation?
- Line 338 – Over what area is the assumption of a homogenous seafloor made – hard and soft substrate? It is not clear what 'average' sediment type and carbon content was used here to scale across the ‘whole’ area?
- Line 342 - Is this the same as assuming no OC present, as in scenarios 2 and 3? Needs some clarity.
- Tables 3 & 4 – further detail needed in the caption to explain what is being shown.
- Table 6 - The table format is difficult to follow. Please consider how to improve the layout. Is the average across the grab samples?
- Column 6 – Is this supposed to be density rather than stock?
- Figures 6 & 7s – The colour key (orange and purple) in the caption is the wrong way round.
- Figure 8 – Is the spatial map showing organic carbon density? The text in the results section suggests that it is OC concentration (content?). Also are the units kg/m3 correct?
I am interested in the results map in Figure 8, which generally shows very high OC densities associated with locations further offshore and within sandy sediments. It would be interesting to discuss why this might be the case – is the spatial model biased by the sample location or are there local circulation patterns occurring which may be transporting material offshore?
Discussion
Overall, this section needs developing and the results discussed further. The discussion has not acknowledged the limitations of the study - I would expect some discussion around the implications of only two sediment classes (and no sampling of the harder substrate), no direct measurement of DBD, what physical processes might be driving the spatial distributions of sediment types and/or carbon hotspots. There should be some acknowledgement of the difference between surface and deeper sediments and how this relates to OC burial if the rationale for the study are being linked to climate. Specific comments include:
- There should be further development of Paragraph 1 – it’s not clear why improved spatial modelling should always result in decreased OC stock estimates – is that what is meant?
- Line 380 – Might be appropriate to discuss that one limitation of your study was only classifying into two sediment types. For instance, how would gravelly mud be defined – as hard or soft substrate - given the binary classification?
- Line 382 – What empirical relationship is being referred to here?
- Line 386 - Why is this reference being used? - As an example of a study saying the same thing or an example of study that has assumed a homogenous seafloor? I disagree with the latter - The Smeaton et al., 2021 study supports the importance of good substrate mapping (16 Folk classification – if the data support the use of such a high-resolution study) for OC stock estimates. It does not assume a homogenous seafloor.
- Lines 395 – 405 – Can the challenges with carbon modelling on the seafloor be further elaborated? How much surface POC reaches the seafloor? Is this the only source of POC in the ocean? What might be driving the spatial distribution of carbon in the map in Figure 8?
- Line 409 - What is the uncertainty with this estimate?
- Line 420 - This study looks at stock of carbon, which is not the same as sequestration (see bullet point above). The discussion needs to better reflect the study and not over-promise on the results.
Conclusion
There is some mixed messaging in the conclusion. The authors suggest throughout the paper that their dataset is satisfactory to determine robust results however the final paragraph in the discussion and the conclusion mentions limitations in the dataset that are not discussed anywhere else.
- Technical Corrections
The definition of blue carbon in the introduction is confusing – how is it being defined in this study, and why is it something that should be cared about?
- Some terms have been used incorrectly. For instance, there are incorrect uses of carbon ‘concentration’ / ‘content’ / ‘stock’ and ‘density’. E.g. Line 258 - Concentration is incorrect here - it represents a mass per volume. This study is measuring content i.e. mass per mass (weight % of organic carbon) (See the paper by Flemming and Delafontaine, 2000).
- There is inconsistent formatting of units; there should be a space between the number and unit. Use either mm or µm.
- There is inconsistent formatting with the references and ‘et al.,’ should be italicised.
- Check references used are supporting and accurate representations of the arguments being made.
- Data are plural – check grammar.
- Line 37 – Use of capitals for Blue Carbon – be consistent throughout the paper.
- Language - Lines 417/422 – ‘Anthropocentric’ is not the appropriate adjective here.
References
Berner, R.A. (2003) ‘The long-term carbon cycle, fossil fuels and atmospheric composition’, Nature, 426, pp. 323–326. Available at: https://doi.org/https://doi.org/10.1038/nature02131.
Burdige, D.J. (2007) ‘Preservation of Organic Matter in Marine Sediments: Controls, Mechanisms, and an Imbalance in Sediment Organic Carbon Budgets?’, Chemical Reviews, 107, pp. 467–485. Available at: https://doi.org/10.1021/cr050347q.
Diesing, M. et al. (2017) ‘Predicting the standing stock of organic carbon in surface sediments of the North–West European continental shelf’, Biogeochemistry, 135(1–2), pp. 183–200. Available at: https://doi.org/10.1007/s10533-017-0310-4.
Flemming, B.W. and Delafontaine, M.T. (2000) ‘Mass physical properties of muddy intertidal sediments: some applications, misapplications and non-applications’, Continental Shelf Research, 20(10–11), pp. 1179–1197. Available at: https://doi.org/https://doi.org/10.1016/S0278-4343(00)00018-2.
Hunt, C. et al. (2020) ‘Quantifying Marine Sedimentary Carbon: A New Spatial Analysis Approach Using Seafloor Acoustics, Imagery, and Ground-Truthing Data in Scotland’, Frontiers in Marine Science, 7(July). Available at: https://doi.org/10.3389/fmars.2020.00588.
Hunt, C.A. et al. (2021) ‘Sounding Out the Carbon: The Potential of Acoustic Backscatter Data to Yield Improved Spatial Predictions of Organic Carbon in Marine Sediments’, Frontiers in Marine Science, 8(November), pp. 1–20. Available at: https://doi.org/10.3389/fmars.2021.756400.
Lovelock, C.E. and Duarte, C.M. (2019) ‘Dimensions of blue carbon and emerging perspectives’, Biology Letters, 15(3), pp. 1–5. Available at: https://doi.org/10.1098/rsbl.2018.0781.
McLeod, E. et al. (2011) ‘A blueprint for blue carbon: Toward an improved understanding of the role of vegetated coastal habitats in sequestering CO2’, rontiers in Ecology and the Environment, 9(10), pp. 552–560. Available at: https://doi.org/10.1890/110004.
Smeaton, C. et al. (2021) ‘Marine Sedimentary Carbon Stocks of the United Kingdom’s Exclusive Economic Zone’, Frontiers in Earth Science, 9(March), pp. 1–21. Available at: https://doi.org/10.3389/feart.2021.593324.
Citation: https://doi.org/10.5194/egusphere-2024-5-RC2 -
AC2: 'Reply on RC2', Craig Brown, 04 Apr 2024
We thank reviewer 2 for their insightful, thorough, and careful review of the manuscript. Below, we provide a response to each point that was raised.
Referee #2
Title
Referee comment: The title of the manuscript should be changed to reflect the local-regional scale of the study rather than implying the results could be applicable at the scale of ‘continental shelves’.
Response to Referee comment: Thank you for addressing the issue with the title. We will clarify by adding the location and emphasizing that it is local-regional scale.
Introduction
Referee comment: The paper would benefit from restructuring the introduction to provide better flow and linkages between the paragraphs, including a stronger link between MBES, sediment type and organic carbon.
The introduction sets the scene for the requirements of the research generally; however, it needs to provide a clearer rationale for how it is novel in comparison to other mapping studies for sedimentary organic carbon and what specific contribution it is making. The paragraphs do not link together particularly well and the authors are encouraged to consider the flow between sections. Crucially, the link between sediment type and organic carbon has not been made, which is essential supporting information to explain why seabed sediment mapping using MBES can yield carbon stock results. The section on marine carbon is chaotic, for instance, it jumps from marine carbon straight into different definitions for blue carbon to benthic carbon. References that are more appropriate are needed to support statements.
Response to Referee comment: We thank the referee for the suggested improvements to the flow of the introduction, and we will address these shortcomings by describing the relationship between sediment type and organic carbon and reordering the sections to ensure they flow better. We will also add more appropriate references and specify on one blue carbon definition that specifically discusses organic carbon and not inorganic carbon to avoid confusion.
- Line 32 – Blue carbon is specifically about organic carbon – if this is not explicit, there is the potential for confusion with inorganic carbon too.
Response to Referee comment: We will specify organic carbon to avoid confusion.
- Line 33 – The term disproportionate needs to reflect per unit area. Rather than the global ocean, it is about specific habitats that can store disproportionate amounts of carbon on an area-by-area basis.
Response to Referee comment: Thank you for your comment, we will make sure to alter the definition to reflect per unit area and highlight that it is specific habitats that can store disproportionate amounts of carbon.
- Line 36 – A more appropriate to reference McLeod et al.,2011, who was part of coining the term Blue Carbon.
- Line 38 – A more appropriate reference would be Lovelock et al., 2019 as this paper discusses how the BC term is evolving in the science and literature.
- Line 40 – A more appropriate reference is required to highlight times of accumulation and burial of sediment over 1000s of years (See papers by Berner, 2003, or Burdige, 2007 to get a longer-term overview). The currently referenced papers investigate surface sediments, which are not where the long-terms stores of carbon are found.
Response to Referee comment: Thank you for your insight - we will add all these references to the paper.
- Line 41 – What is the scale for this estimate? Is that one trawl or all trawls everywhere?
Response to Referee comment: That is a good question - we will specify this more clearly in the paper. The study by Sala et al., 2021 estimates that 4.9 million km2 or 1.3% of the global ocean is trawled each year. This disturbance to the seafloor results in an estimated 1.47 Pg of aqueous CO2 emissions. Therefore, the study is combining all trawls everywhere which creates an area of 4.9 million km2.
- Line 42 – Only one reference provided, although next sentence refers to more than one study.
Response to Referee comment: Thank you for noticing that error we will alter the grammar.
- Line 45 – Could be more specific about which anthropogenic activities. Can they be characterised?
Response to Referee comment: Yes, we can be more specific and emphasize anthropocentric activities like bottom trawling and dredging.
- Line 48 – MPAs have traditionally been designated for biodiversity – designation for carbon would be a novel approach?
Response to Referee comment: Yes, we emphasize in the paper that MPA’s are often determined based on high areas of biodiversity and spatially mapping carbon could expand the definition of MPA’s to become areas that need to be protected due to high amount of CO2 sequestration.
- Line 56 – Sampling systems – do you mean physical samples or other?
Response to Referee comment: Yes, we will make sure to specify that it is physical sampling
- Line 57 – Can expand on the relevance of the different sediment names – i.e. increasing grain size. What classification is this?
Response to Referee comment: We will expand on the relevance of different sediment names and specify that in this study we utilized the Wentworth scale.
- Line 66 – Are these early studies based on terrestrial or marine carbon?
Response to Referee comment: Good question, these studies are marine carbon based studies so we will ensure that is clarified.
- Line 73 – Was MBES data used in all the referenced studies?
Response to Referee comment: When examining the referenced studies Smeaton et al. 2019 was the only study that examined MBES data, which emphasizes how novel our approach is in marine carbon mapping research. We will make sure to elaborate on that in the paper.
- Line 80 - There has been no mention yet about the relationship between sediment type and organic carbon so it is not clear why the extent of bedrock would make a difference to calculations.
Response to Referee comment: Thank you for addressing that point, we will make sure to add this information into the paper to make it clearer for the reader.
- Line 86 – The Hunt et al., 2021 study results were not output at 6 m resolution. Check references used are accurate representations of the points you are making.
Response to Referee comment: That is a great point, we will specify that the Hunt et al., 2021 performed one calculation of organic carbon stock using predictions from the linear mixed model with backscatter at 48 m resolution.
- Lines 91-92 – Could be more explicit in why the two studies mentioned found differences. The studies were apparently in very different geographical settings as one possibility.
Response to Referee comment: Yes, we agree with your point and will add to that paragraph discussing why the estimates in these studies are different emphasizing that one reason is due to the different locations and approaches.
- Lines 96-99 – This sentence is too long and vague - lots of challenges with global estimates and is it a realistic scale for management?
Response to Referee comment: Thank you for that insight with will shorten this sentence and add more detail.
- Line 100 – Further detail about why it is an Area of Interest?
Response to Referee comment: We will add more detail about the significance of the site and why it is an area of interest for future conservation/protection by Fisheries and Oceans Canada.
- Line 101 – Question about the relevance of the setting - I understand that this location is a good setting to test the hypothesis that different sediment types have different carbon densities? Would that be true?
Response to Referee comment: Yes, previous studies have explored carbon mapping in a homogenous seabed, therefore our study is novel since we are exploring carbon within a heterogenous seabed environment.
Study Site
This section could be strengthened by including a description/ characteristic of the Study Area that might be more relevant to sedimentary carbon. There is no description of what is known about seabed sediment type for instance.
Response to Referee comment: We will revise to include a description of the study site that includes what was known prior to this study modelling the substrate. There was some limited work conducted at this location conducted by the Geological Survey of Canada which we will reference.
- What is the relevance of temperature and nutrients to potential carbon stocks?
Response to Referee comment: We wanted to provide general information about the location since temperature and nutrients all have a relationship with carbon and could provide insight on why there is substantial organic carbon in certain areas on the map.
- Why is it an Area of Interest to the Canadian Government?
Response to Referee comment: The eastern Shore islands has been identified as an Ecologically and Biologically Significant Area. It has unique coastal and marine habitats and significant concentrations of kelp beds and eelgrass. It also is an area used by juvenile Atlantic cod (Endangered species in Canada), white hake, and pollock. It is a spawning area for Atlantic herring and important habitat for Atlantic salmon (Endangered species in Canada). Furthermore, it has significant foraging area for a variety of sea- and shorebirds, including Harlequin duck (Special Concern – SARA), purple sandpiper, and Roseate tern (Endangered – SARA)]. We will make sure to add this information to the paper to provide context for selection of this site for our study.
- Figure 1 - Would appreciate the location within a more generalised map of Canada somewhere as well to get the wider geographical context.
Response to Referee comment: We will add a more generalised map of Canada to the figure.
Methods
The method chapter is currently too vague and requires further development to include specific details for clarity (it is a little unclear what has been done and in what order) and for repeatability. Some areas for improvement include:
- Line 154 - Explain more about what focal statistics is and how it was applied.
Response to Referee comment: We will add more detail about how focal statistics works and how we applied it to our research.
- Line 166 - How many grab samples were taken overall?
Response to Referee comment: 17 grab samples were taken overall. We will specify that in the paper.
- Line 167 - A stratified random sampling technique was used based on the backscatter, however no information has been given about what classification method was used for this.
Response to Referee comment: The sampling design did consider broad regions of similar backscatter intensity in order to design the sampling survey. Further clarification will be provided in the methods sections will be provided on the sampling design.
- Line 169 - How large were the subsamples? How big was the Van Veen grab?
Response to Referee comment: We already state that the penetration depth of the grab sampler was 10 cm (line 250). We will add further clarification that the Van Veen was a 0.1m2 grab, along with further clarification on the subsampling volume/procedure (subsamples were the 32 oz plastic containers - we will rephrase that sentence to make sure that is clear).
- Predictor variables – I don’t see that sediment type has been explicitly mentioned. E.g., Line 150 mentioned additional predictor variables but it is not clear what they are additional to.
Response to Referee comment: Sediment type is not a predictor variable - it is the response variable being modelled utilizing the predictor variables. Also, we were trying to explain that from the primary bathymetric dataset, additional predictor variables like slope, bathymetric position index (BPI) and vector ruggedness measure (VRM) were derived. We will make sure to make that clearer – and rephrase sentences that may be unclear.
- Line 168 – Backscatter is not always a good proxy – Could you caveat this by saying backscatter can be a good proxy for sediment grain size and perhaps add some more references for studies where this is the case?
Response to Referee comment: Thank you for this insight, we will alter the wording here and add additional references to prove our case.
- Line 172 - Only soft substrates have been sampled – what was used to ascertain what substrates were and were not suitable prior to sampling?
Response to Referee comment: We did not decide this prior to sampling. Based on the samples we collected the only samples we could measure using the elemental analyzer had to have a grain size below 2 mm diameter. We will emphasize this in the paper.
- Line 178 - Were the sediments sorted for grain size before being prepared for OC analysis? How were the sediments above 2 mm separated? Was bulk density of the sediment measured?
Response to Referee comment: Yes, the samples were sorted into three grain size categories using mesh sieves (pebble/cobble (>4000 µm), gravel (>2000 µm) and fine sediment (<2000 µm)). The sample that contained fine sediment (<2000 µm) was able to be measured in the elemental analyzer. Only two samples were comprised of 30% gravel (>2000 um) and that fraction was removed and the %OC was recalculated to accommodate for that change in weight. We will make sure this information is clearer in the paper. Bulk density of the sediment was not measured and can be added as a limitation to the study.
- Line 180 - Were samples dried from frozen?
Response to Referee comment: Samples were stored in a fridge to reduce biological activity that could impact OC and then it was dried. We will ensure that this is emphasized in the methods section.
- Line 183 – How was the coarse fraction estimated and removed? Did this result in any loss of integrity of the sample?
Response to Referee comment: The coarse fraction was estimated using the mesh sieves, therefore we knew the % weight of the coarse fraction within the sample. When examining figure 7, only two samples had a coarse fraction (gravel) and we state that this in line 177-179: Two samples (ES-31, ES-35) had notable amounts of course-grained sediments (around 30% of sample). The coarse fraction was removed from these samples, and the % OC adjusted accordingly. We will modify this sentence by describing further how we altered the calculations to accommodate for the loss in sample.
- Line 192 - How was it determined that the sediments had relatively low organic content?
Response to Referee comment: Before the grain size analysis was performed on the fine sediment organic carbon content was measured and it was relatively low for all the samples, indicating that the samples did not need to be treated with acid or hydrogen peroxide. We will make sure to expand on this in the paper.
- Line 192 - Were the coarse and fine fractions measured as % of mass or volume of the total? Was the full particle size distribution used to classify sediment type at all?
Response to Referee comment: The coarse and fine fractions were measured as % mass of the total. We will make sure that it is clear in the paper. The sediment classification was determined using seafloor video imagery; therefore, the particle size distribution was not used to classify the sediment type.
- Line 206 - Was the camera stationary for 3 minutes or was it taking a video transect?
Response to Referee comment: The camera was not stationary and instead moved for 3 minutes. We will add this information to the methods section to make it clear for the reader.
- Line 213 – Two classifications is quite rudimentary, and it should be acknowledged somewhere when interpreting the results as a potential limitation. Mixed sediment can exist which contain proportions of mud, sand and gravel, e.g., a gravel veneer on a muddy substrate, which will have a different OC content than a sandy or pure gravel.
Response to Referee comment: Thank you for noting this limitation. We originally tried to add more sediment classification types, but the random forest model had high error suggesting that the sediment type was too complex and challenging for the model. Our ability to accurately resolve subtle differences in grain size from the video data was also a limitation – and a large number of physical grain size measurements would have been required to expand the sediment classification to a larger number of classes (which was beyond the scope of this project). We will expand the discussion on this point in the paper to emphasize the challenges in modelling/mapping complex, heterogenous areas of the seafloor environment. Nonetheless, we feel that even with a simple 2-substrate map of the seafloor it provides valuable insights into spatial complexities and improvements in estimating stocks of OC at the seabed.
- Line 223 – It is not clear what is being modelled - is it carbon content within the sediment? The title of the section is confusing.
Response to Referee comment: Section 3.5 is only modelling sediment type and no examination of carbon content. The sediment model provides insight of where the soft substrate is located to provide the area for interpolation of the carbon content. We will alter the title to avoid confusion for the reader.
- Line 253 – Are there any limitations of assuming the same single grain density value across the area? For instance, could it artificially over-inflate carbon stock estimates for muddier sediments?
Response to Referee comment: Yes – there are potential limitations of assuming the same single grain density value across the area. We will incorporate this as a discussion point.
- Line 254 – Check that the references are correct to support the method used.
Response to Referee comment: We will double check the reference in this sentence. Thank you for noticing this error.
- Line 261 - Did this step of modelling the OC content at unknown locations happen before or after the OC stocks were estimated in the previous section?
Response to Referee comment: The modelling of OC content at unknown locations happened after the OC stocks were estimated. But, before or after is not relevant since the OC stock utilizing the EBRK model can be calculated using the zonal statistics tool in ArcGIS Pro. We make sure to elaborate in the paper that all three scenarios can be calculated individually.
- Line 273 – What does a small RMSE value indicate?
Response to Referee comment: A small root mean square error (RMSE) indicates that the model has performed well and can predict the data accurately.
- Figure 2 – In the map legends, the values of the parameters could be rounded up.
- Table 1 – formatting – centre the table headings
Response to Referee comment: Thank you for both these insights, we will round up the values and centre the table headings.
Results
The main issue here it that the calculated stock estimates do not appear correct when compared against other studies that have also used OC Content and Dry Bulk Density (e.g., Diesing et al., 2017, Smeaton et al., 2021, Hunt et al., 2020) - they are at least an order of magnitude higher and the calculations should be revisited.
Response to Referee comment: Thank you for picking up on the error in the calculations, we will correct the estimates by removing the arcsine transformation.
- Line 275 – Is this a measurement of organic carbon content or concentration? See Flemming and Delafontaine, 2000.
Response to Referee comment: We measured organic carbon content and its relationship to the grain size composition of mud. We will make sure that is more clear in the paper.
- Line 277 - What framework is used for silt + clay = mud? Reference needed.
Response to Referee comment: We will add a reference here to explain why silt+ clay= mud. Thank you for that input.
- Line 284 – Clarity needed – The results from Figure 4 suggests to me that % mud is a good proxy for OC - not sediment type. Sediment type is a classification based on the total composition of grain sizes.
Response to Referee comment: Thank you for picking up on that distinction. We will alter our language to state that it is % mud that is a good proxy not necessarily sediment classification.
- Line 309 – Are these statistics for across the whole area? Or for the grab samples? What useful information does this general distribution provide?
Response to Referee comment: These statistics are just for the grab samples which were measured for OC content and sediment grain size. This figure helps explain the distribution of grainsize within the grab samples and demonstrate that the areas with the highest OC content have high quantities of mud (silt and clay). We will ensure that this is clear by adjusting the sentence structure.
- Line 325 – Can you expand on the significance of the accuracy of the interpolation?
Response to Referee comment: Yes, we can expand on the accuracy of the interpolation to provide further understanding for the reader.
- Line 338 – Over what area is the assumption of a homogenous seafloor made – hard and soft substrate? It is not clear what 'average' sediment type and carbon content was used here to scale across the ‘whole’ area?
Response to Referee comment: The first scenario is assuming no detailed knowledge of the sediment complexity in the area - and that the entire study area is homogenous and similar to the sediments sampled in the grabs. The ‘average’ carbon content is supposed to represent the calculated organic carbon stock (kg/m3) from each grab sample averaged and multiplied by the area for each scenario. We will make sure this is clarified in the paper.
- Line 342 - Is this the same as assuming no OC present, as in scenarios 2 and 3? Needs some clarity.
Response to Referee comment: Scenario 1 is assuming we have no understanding of the seabed and cannot differentiate between hard and soft substrate therefore the averaged OC stock from the grab samples is multiplied to the entire study site. We can delve deeper into each scenario to make sure all 3 are clear to the reader.
- Tables 3 & 4 – further detail needed in the caption to explain what is being shown.
Response to Referee comment: We will add more detail in the caption for both these figures -thank you for that suggestion.
- Table 6 - The table format is difficult to follow. Please consider how to improve the layout. Is the average across the grab samples?
Response to Referee comment: We will alter the table format to make it easier for the reader to follow. And yes, the average is across the grab samples.
- Column 6 – Is this supposed to be density rather than stock?
Response to Referee comment: Yes – this is technically density. This will be changed.
- Figures 6 & 7s – The colour key (orange and purple) in the caption is the wrong way round.
Response to Referee comment: We will alter that, thank you for noticing the error.
- Figure 8 – Is the spatial map showing organic carbon density? The text in the results section suggests that it is OC concentration (content?). Also are the units kg/m3 correct?
I am interested in the results map in Figure 8, which generally shows very high OC densities associated with locations further offshore and within sandy sediments. It would be interesting to discuss why this might be the case – is the spatial model biased by the sample location or are there local circulation patterns occurring which may be transporting material offshore?
Response to Referee comment: This is certainly an interesting observation – and one we also noticed. It is possible that the morphology of the area, with the softer sediments occurring in low-standing depressions between bedrock reefs, combined with local currents, are accumulating higher levels of organic material such as kelp (seasonally), or other organic content from terrestrial sources. We will add material into the discussion to discuss potential reasons for these observed spatial patterns.
Discussion
Overall, this section needs developing and the results discussed further. The discussion has not acknowledged the limitations of the study - I would expect some discussion around the implications of only two sediment classes (and no sampling of the harder substrate), no direct measurement of DBD, what physical processes might be driving the spatial distributions of sediment types and/or carbon hotspots. There should be some acknowledgement of the difference between surface and deeper sediments and how this relates to OC burial if the rationale for the study are being linked to climate. Specific comments include:
Response to Referee comment: We agree with the referee that more limitations should be addressed in the discussion, and we will add insight on what physical processes are causing the spatial distribution. We will also address the different between surface and deeper sediment in relation to OC.
- There should be further development of Paragraph 1 – it’s not clear why improved spatial modelling should always result in decreased OC stock estimates – is that what is meant?
Response to Referee comment: Thank you for your comment we will expand more in paragraph one. The goal of mapping is to provide more accurate representations on OC stocks and variability in OC within a small study area. In this case it decreased the OC stocks as we increased the detail, which makes sense since without examining the spatial variability within an area it will be assumed that the estimate is even throughout.
- Line 380 – Might be appropriate to discuss that one limitation of your study was only classifying into two sediment types. For instance, how would gravelly mud be defined – as hard or soft substrate - given the binary classification?
Response to Referee comment: Yes, we will add the two-sediment classification as a limitation and express the challenge with modelling sediment type. We will add further discussion on these points into the paper.
- Line 382 – What empirical relationship is being referred to here?
Response to Referee comment: The empirical relationship is that adding more spatial detail can improve organic carbon estimates.
- Line 386 - Why is this reference being used? - As an example of a study saying the same thing or an example of study that has assumed a homogenous seafloor? I disagree with the latter - The Smeaton et al., 2021 study supports the importance of good substrate mapping (16 Folk classification – if the data support the use of such a high-resolution study) for OC stock estimates. It does not assume a homogenous seafloor.
Response to Referee comment: The reference is being used since this study said the same thing. We will alter that sentence to make that clear.
- Lines 395 – 405 – Can the challenges with carbon modelling on the seafloor be further elaborated? How much surface POC reaches the seafloor? Is this the only source of POC in the ocean? What might be driving the spatial distribution of carbon in the map in Figure 8?
Response to Referee comment: We will add more details on the challenges with carbon modelling and the lack of papers that have tried to analyze this problem. All of these are great questions that remain unanswered, and we will address these in more detail in the discussion.
- Line 409 - What is the uncertainty with this estimate?
Response to Referee comment: We are unclear as to what uncertainty the referee refers to with respect to this point/line number?
- Line 420 - This study looks at stock of carbon, which is not the same as sequestration (see bullet point above). The discussion needs to better reflect the study and not over-promise on the results.
Response to Referee comment: We will make sure to only discuss stock of carbon to ensure that we are not providing misleading statements.
Conclusion
There is some mixed messaging in the conclusion. The authors suggest throughout the paper that their dataset is satisfactory to determine robust results however the final paragraph in the discussion and the conclusion mentions limitations in the dataset that are not discussed anywhere else.
Referee comments: We will make sure to restructure our conclusion and add limitations to previous parts of the paper to ensure that no new information is addressed in the conclusion.
- Technical Corrections
The definition of blue carbon in the introduction is confusing – how is it being defined in this study, and why is it something that should be cared about?
Referee comments: Thank you for that insight we will alter our definition to make it more clear for the reader.
- Some terms have been used incorrectly. For instance, there are incorrect uses of carbon ‘concentration’ / ‘content’ / ‘stock’ and ‘density’. E.g. Line 258 - Concentration is incorrect here - it represents a mass per volume. This study is measuring content i.e. mass per mass (weight % of organic carbon) (See the paper by Flemming and Delafontaine, 2000).
Referee comments: We will review our terminology to ensure we are not using incorrect terms – and ensure consistency throughout the manuscript.
- There is inconsistent formatting of units; there should be a space between the number and unit. Use either mm or µm.
- There is inconsistent formatting with the references and ‘et al.,’ should be italicised.
- Check references used are supporting and accurate representations of the arguments being made.
- Data are plural – check grammar.
- Line 37 – Use of capitals for Blue Carbon – be consistent throughout the paper.
- Language - Lines 417/422 – ‘Anthropocentric’ is not the appropriate adjective here.
Response to referee comments: Thank you for finding these technical errors, we will review the document to fix these gramma and formatting mistakes
Citation: https://doi.org/10.5194/egusphere-2024-5-AC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 274 | 81 | 20 | 375 | 16 | 6 | 9 |
- HTML: 274
- PDF: 81
- XML: 20
- Total: 375
- Supplement: 16
- BibTeX: 6
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1