High-Resolution Drone Images Show That the Distribution of Mussels Depends on Microhabitat Features of Intertidal Rocky Shores

Round 1

Reviewer 1 Report

Dear authors,

In your manuscript over utilizing drones to map microhabitats demonstrating the microhabitat features of intertidal rocky shores sought to characterize distribution of mussels  and to evaluate the role of topographical heights in mussel distributions. After reading the manuscript, it seemed neither objectives were specifically met. I found the manuscript to be an un-cohesive writing that combined mussel biology and and drone technology. Thus, the manuscript fell short for publication. In re-writing the manuscript, please either select the technology aspect (while including biology to back it up) or the biology aspect (while including technology to back it up). As written, the manuscript tries to do both and the message is lost and diminished. The work presented is valuable for those that may work with bi-valves in the future with drones, but a specific story is needed in order for this manuscript to make sense. Given many studies with bi-valves have been conducted on multiple reefs, multiple locations, and based upon multiple factors, the authors should redesign this manuscript to demonstrate how the technology can highlight the biology. If the authors are trying to demonstrate the biology based on the technology, for two reefs in a single area, the factors of their manuscript are rendered biologically useless for a bivalve.

Abstract- was well written.

-mention where the study was done.

Line 19- occupation rate? Does this mean percent cover.

Line 28-29. The last sentence of the abstract is useless. The best take home message the authors could pen was “potential advantages…. Are discussed.”? Why not be impactful and state why drone imaging should be utilized? Or should it be advantageous for the mussels?

Introduction-

Line 34- generally written as three dimensional in bivalve literature. I have never seen it as tridimensional. But still works, just caused me to pause and not continue reading the paper.

Line 41- change shellfish to bivalves. Some parts of the world collectively utilize shellfish to encompass bivalves (oysters, mussels, clams), shrimp, and crabs (essentially, if it has a shell).  Unless you are also including crabs and shrimps- which then one would question how you are determining movements of crabs and shrimp via satellite images.

Line 43 seems to be missing citations after~0.5m. Earlier in the sentence the authors wrote “but most of the methods” with no subsequent citations.

Line 45- need commas around like drones

Line 48- citations?

Lines 48-51- authors state use has increased “in the last decade” without including citations to back it up.

Lines 43-52 do not read well and it is extremely awkward. The authors seem to have stuck in information from another place and have no citations and no reference to bivalves. Then suddenly return to the bivalves.

Line 58-61- needs citation bolstering.

Line 62- tense. The studies are in the past

Line 64-66- language is awkward

Line 68-70- are you going to study the questions?

Line 76-77. The authors have literally not address microhabitats. There is a plethora of literature for similar habitat types- seasgrasses and oyster reefs, that have the same functionality and may experience the same abiotic stressors- that have microhabitats. However, up to this point the lack of cohesive synthesis of the sensing with the biological aspects has been extremely elementary. What is the definition of a microhabitat?  What species are you surveying?

Line 77-78- finally! A species. Would it not be great for the reader to have a background in this particularly important species early on?

Methods

Figure 1B does a great job of highlighting the study site.

Line 92-94- The authors have done a poor job in setting themselves up for this. The introduction was mainly about the sensing aspect, with limited aspects of the biology. Now, the reader is presented with crowding index. This should be explained (given it is a remote sensing journal..) is this a % of mussels overlapping per m²? Or based upon bare spots of a mussel bed within a m²? How many folks reading a remote sensing article are going to be familiar with crowding indices of mussels? And what does that t

Line 106- spring or neap low tide? Or does that matter at the study site?

Line 113- Change besides to additionally

Results

Orientation is still extremely confusing. The orientation of the bed? Mussel? It seems that they are going to be oriented where the tide comes in and that south to west seems to make sense based upon the map provided of land and tides.

267- microsite vs microhabitats

Figure 3- I continually referred back to figure 1B when looking at this figure. I would suggest, to strengthen your readers engagement, include figure 1b as the first panel here. That way, we (the reader) can see exactly what these awesome figures are based on. Having to go back and forth was extremely painful. If I was not reviewing the paper, I would have given up reading the paper at this exact figure.

Figure 5- so cool!

Discussion

Why genus and species here and not throughout the manuscript?

Line 386- missing comma after in addition

Line 401- don’t refer me to another paper. Tell me the factors here or I am leaving your paper and reading another one

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Brief summary:  The authors present a drone image analysis method for quantifying mussel beds on the intertidal zone. They use color and morphological features for discriminating mussel aggregations from surrounding materials. This is an interesting, drone-based application for fine-scale ecological mapping highlighting the effects of micro-habitat structures. The authors suggest a new image metric, the Mussel Visualization Index, for extracting pixels with mussel cover. Their approach provides an affordable and quick way to quantify mussel beds and correlate their coverage with environmental parameters such as the immersion time. Their analysis contributes towards the exploitation of new technologies and algorithms in coastal mapping and supports new concepts regarding conservation of intertidal species.

Broad comments: The Introduction presents sufficient background information and clearly states the aim of the study.  Methods are adequately explained and supported by quantitative information. The results are clearly presented and the discussion is well structured. One major comment is related to the decision to present only one study area. Although the available data seem to be enough for supporting the aim of the study, it should be justified by the authors why they did not consider additional study areas for assessing the influence of environmental variability in their study. The study in its current form provides very localized knowledge and it would be useful to see if/how this concept works in other areas. So there are two options for the authors: a) in case there are more data available from different geographic areas then they should be incorporated in the study, b) in case there are not more data available, the authors should outline in the Discussion the broader validity and robustness of the method in other areas and present a draft plan for future data collections.     

Specific comments: Please consider a clearer map for Fig1A showing a zoomed part of the study area. 

The manuscript can be further considered for publication once the above comment is sufficiently addressed by the authors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper explores the use of UAVs for the detection and quantification of mussels in the intertidal, as well as the potential of this techniques to be implemented as monitoring tools. The author's also explore the potential of combining mussel coverage information with geomorphological features extracted also from the images, to better understand underlying processes controlling mussel distribution.

I've found the paper very interesting, well written and easy to read. With regard to the sampling design, I'm only missing some ground control points of mussel coverage, but I imagine that image resolution was good enough to discriminate in the images between mussels and other similar cohabitants (lichens or some algae). I think the analysis of the data is appropriated and the results are clearly explained and discussed according to updated bibliography. I would recommend the publication of this paper. I only have minnor comments, detailed bellow:

Lines 92-94: I don't understand how they can be monolayered as you mention in the introduction and now in the material and methods section, crowded with and index of 1.5...maybe i dont understand the meaning of the crowding index....please explain

Lines 182-184: This enumeration gets confussing for me because I can't understand the steps of the workflow until you explain them later. I would delete the enumeration of the steps here, that's the role of figure 2.

Lines 220-223: Don't you have any sampling point on the ground to validate de presence/absence of mussels in the orthomosaic? Do you have enough resolution in your images to ensure mussel and non-mussel id?? Maybe you should mention that resolution was high enough for unequivocally identification of mussels and non mussels

Line 236: "from raster to points" shouldn't be to polygons? And then calculate the average of slope and orientation for that polygon??

Lines 238-240: I don't understand why do you change scale, first from 1x1 to 20x20 for slope and orientation calculations from DEMs and back to 1x1 to match those geomorphological variables with mussel distribution....I understand the computational problems...but how do you go back to 1x1? Please explain this better because it gets confusing 

Lines 362-364: please add some references

Line 368: In addition to predation, intra- and inter-specific competition might also play a role

Lines 371-374: Emmersion time also increases anaerobic respiration rate, ammonium excretion and as you mention decrease feeding available time, limiting the Scope for Growth (SFG), or Energy Budget of mussels in the upper intertidal. Maybe you should explain better that the upper intertidal is in general a more estressful habitat (dissecation, temperature, etc) and cite some references

Lines 409-410: Higher density of mussels instead of abundance or total covered area

Lines 407-410: Nonetheless, you are not considering the negative effect of high densities, intra-specific competition for food or space, maybe you can check some of the works of Cubillo A. on that topic and the self-thinning phenomena and include something about that,

Lines 461-463: That's the self-thinning rule effect that I was suggesting to include in the previous comment

Line 472: The name of the first author for [13] is lacking

Lines 491-494: This paragraph is a little bit confusing, please rephrase

Line 501: Although they have certain mobility when settled... I would consider mussels as sessile animals, not semi-sessile

Lines 503-506 I would use the oyster example first and then finalize the discussion with the potential of mussels and drone monitoring to replicate those efforts, as a good end for the discussion.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

This paper deals with the topic within image processing for remote sensing. The work can help fish farmer or mussel’s farmer to optimize the production and environment protection and preservation. The study proves the use of the drones and the information and communication technology to help decision making.

The obtained results painted the possible high differences in the total occupied zone between rocky shores notwithstanding being near in space. The higher mussel’s covered area on the East shore, where there was a shore-specific mussel preference for occupying south-oriented microhabitats suggests the role of topographic features in conditioning the distribution of mussels.

The most likely cause of this higher mussels’ covered area in the East is the relationship between substrate orientation and slope and particular microclimate conditions. Such effects of topographic features could also play a role in the dynamics of occupied areas, these must be justified and clarified

The authors are invited to give the other species with similar color properties.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 5 Report

This is very interesting study and the paper is written well.

In this study, they used orthomosaics and a Digital Surface Model (DSM) generated from drone surveys to characterize the distribution of mussels  aggregations at high resolution (centimeters) on two rocky shores oriented differently on both sides of a beach, and to evaluate the role of topographic features, intertidal height, slope, and orientation angle in determining the mussels’ distribution.

I have a small question in this study.

In Section 2.2, the final cell resolution was decreased to 20 x 20 cm before deriving terrain attributes to facilitate processing and avoid very local roughness effects when evaluating the effect of topographic features.

How was 20cm determined? We know that this is a critical parameter that closely related to the size of the targets and could caused the different slope and orientation statistics.

LAT’s full name should be given in the abstract where it first used.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Overall, I found the manuscript tremendously improved. However, just a few comments to make this manuscript:
Line 89-90 in the clean version- what is mean by orientation attributes? It is confusing for the terrestrial vs aquatic. I would argue the orientation for  a terrestrial habitat would be the sun. And do not think that is necessarily applicable here. I think the authors need to explicitly state why orientation is important for mussels. Because in Lines 168-169, they discuss that it is looking at the orientation to due north. Is that based on the how the mussels open? Or the patch structure? This also needs to be clarified here.  Because in the results (Figure 3) it seems that it is location within the rocky shore.

I think just adding a little more info in that intro and methods to make in explicitly clear what you are wanting the reader to follow, will go a long way. As of now, it is a little unclear. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf