A Seagrass Mapping Toolbox for South Pacific Environments

Round 1

Reviewer 1 Report

This manuscript designs a monitoring toolbox for seagrass in South Pacific environments. The main function of the toolbox is a decision-tree for selecting monitoring tools for monitoring seagrass at regional, national, local or community-based and emergency response scales, i.e. in-situ equipment or remote sensing technology or their combinations. And the decision-tree considers the six criteria. They are the purpose of the mapping, the size and accessibility of the area to be mapped, the speed with which the information is needed, the accuracy and level of detail required, the skills and experience of the monitoring team and the finances available for monitoring. 

The main issues confuse me as followings.

1) What is the motivation of designing the monitoring toolbox? There is a vague or lack of discussion in Introduction Section.

2) The key point of this manuscript is to build the decision-tree for selecting monitoring tools. This manuscript considers six criteria, however, how to build this decision-tree on the six criteria, i.e. how to draw the Figure 3 from Table 2? 

In Figure 3, how to determine Yes or No? Most of the discriminations are based on qualitative analysis rather than quantitative index. The qualitative toolbox would reduce the scientific metric of this manuscript, also limits its popularization and application.

3) What is the relationship between the topic of this manuscript with remote sensing domain or Remote Sensing Journal.

4) The toolbox freely shared or not should be addressed in discussion.

 

Author Response

This manuscript designs a monitoring toolbox for seagrass in South Pacific environments. The main function of the toolbox is a decision-tree for selecting monitoring tools for monitoring seagrass at regional, national, local or community-based and emergency response scales, i.e. in-situ equipment or remote sensing technology or their combinations. And the decision-tree considers the six criteria. They are the purpose of the mapping, the size and accessibility of the area to be mapped, the speed with which the information is needed, the accuracy and level of detail required, the skills and experience of the monitoring team and the finances available for monitoring. 

The main issues confuse me as followings.

1) What is the motivation of designing the monitoring toolbox? There is a vague or lack of discussion in Introduction Section. We have re-ordered the introduction, moved some text from Section 3 and inserted new text (lines 99-122) to better describe the context and motivation for the study.

2) The key point of this manuscript is to build the decision-tree for selecting monitoring tools. This manuscript considers six criteria, however, how to build this decision-tree on the six criteria, i.e. how to draw the Figure 3 from Table 2? The five criteria are used to populate the toolbox itself, while the decision tree is used to decide which tool from the toolbox to use in a given situation - so it focusses on slightly different questions (6 drivers, lines 392-400). We have edited the manuscript to better explain this (Introduction lines 101-107, Section 3.2 paragraph 1, new figure to illustrate the toolbox-Figure 4).

In Figure 3, how to determine Yes or No? Most of the discriminations are based on qualitative analysis rather than quantitative index. The qualitative toolbox would reduce the scientific metric of this manuscript, also limits its popularization and application. That is correct, the decision tree is focussed on yes/no questions that allow decision makers, particularly those who may not have extensive technical knowledge, to select remote sensing tools (or in situ if that is more appropriate for their circumstances or for ground-truthing). We argue that a qualitative toolbox, rather than one based on quantitative indices, will increase the popularisation and application of the toolbox because it will be more accessible than one requiring a high level of technical skill to navigate. The paper aims to simplify many of the complexities associated with remote sensing applications providing a pathway for South Pacific environmental managers to access seagrass data beyond in-situ monitoring. We have added text to reflect this in Section 3.3 (lines 402-410).

3) What is the relationship between the topic of this manuscript with remote sensing domain or Remote Sensing Journal. The topic of the manuscript is to provide a framework for how to decide which remote sensing tool to use for mapping and monitoring seagrass in different circumstances. This fits with the scope of Remote Sensing because the majority of the tools encompassed in the toolbox are remote sensing tools (satellite, remotely piloted aircraft, kites and underwater camera systems). It thus fits under the journal scope of ‘remote sensing applications’. We have re-organised Section 2 to make remote sensing the first focus of the text and have clarified text in the title, abstract and throughout that the focus is mapping for the purposes of monitoring.

Again, as above, we believe that the paper fits within the scope of the special issue, “Earth Observation technologies and data has greatly advanced our capability to map and monitor such ecosystems, providing essential information to support and evaluate management and conservation strategies. Submissions are thus invited on recent advances in remote sensing of the Great Barrier Reef and other reef ecosystems in Oceania, from novel methodological approaches (sensors, algorithms development and data management) to applications for environmental monitoring and management. “ Remote sensing journal often has very specific special issues, drawing on many aspects of the use and application of remote sensing that go beyond the mechanics and development. This special issue, as with many others, is also about how we can operationalise the process, allowing remote sensed data to be part of our environmental monitoring and assessment. This is particularly relevant to environmental managers in remote areas across the South Pacific, and this paper provides such a pathway.

4) The toolbox freely shared or not should be addressed in discussion. The toolbox and decision tree are presented in the paper and are thus freely available for anyone to use.

Reviewer 2 Report

Please find my comments in the attached pdf file

Comments for author File: Comments.pdf

Author Response

Line 46. Vague, please be more specific Done

Line 128: recommended reference We thank the reviewer for the reference and have included it.

Table 1: What about bathymetric LiDAR and its potential to measure full wave-form? AND Line 48: In addition, it seems that bathymetric LiDAR can be a good solution for bedform and seagrass mapping. Besides bathymetry, it allows measurements of full wave form that can be interpreted similarly to backscatter Thank you for pointing out we had omitted LiDAR. Thanks. We have incorporated both airborne and terrestrial LiDAR; see Table 1 and lines 194-206. Note that both approaches are ruled out of our toolbox when appraised against the criteria (cost and complexity).

Line 148: good summary but what about underwater photogrammetry? Added, see lines 212-214.

Line 157: Less than 0.1 m for multibeam echosounder in shallow waters . Noted. Have deleted statement as resolution is dealt with in Table 2.

Line 172: typo Done

Figure 2: geographic coordinates are missed. Recommend marking the maps with A-C as well as adding a general map with locations of all smaller maps. Figure updated, thanks.

Table 2: Why satellite data has lower max spatial resolution than crewed aircraft? We have edited the column in Table 2 to reflect that it refers to maximum spatial resolution.

Line 265: What kind of risk? How to compare it with crewed aircraft? Edits in line 350-351

Line 396: Please provide more details of the implementation of Random Forest Additional information on the mapping process has been given in lines 513-536 and Appendix 1.

Reviewer 3 Report

Dear colleagues,

 

the paper „A seagrass monitoring toolbox for South Pacific environments” suggest how seagrass meadows might be monitored. I think the authors have chosen the wrong Journal.

It is one of many papers that are currently published promoting seascapes mapping with a combination of various sensors (optics, acoustics), which is meaningful, but not innovative. The paper does not present new knowledge that might be valuable for a Journal like this. Everyone firm with remote sensing knows what is suggested in this paper already. (e.g. Table 1, good for someone new to the field). The work is a nice recommendation for local politicians, e.g. with Fig. 3. It remains in the dark what is the scientific base for such recommendations.

I think most authors are new to the field of remote sensing, and the ones affiliated with i-Sea promote their commercial product with this paper. The methods how Fig. 4 was classified remains unclear, atmospheric correction? Water column correction? By the way, all figures come without co-ordinates…

The authors seem not at all up to date with the state-of the art literature regarding remote sensing of submerged aquatic vegetation neither in optics nor in acoustics.

 

Overall, it is a recommendation paper for the local politicians, but not suited for this high JIF remote sensing Journal. 

Author Response

Please see attachment

Author Response File: Author Response.docx

Reviewer 4 Report

This manuscript seems to be very close to a review paper than a research paper. I don't feel the need to modify it since the contents are rigidly structured by the authors. This manuscript will be interesting for the readers, but the novelty as a paper is weak. 

Author Response

Reviewer 4

This manuscript seems to be very close to a review paper than a research paper. I don't feel the need to modify it since the contents are rigidly structured by the authors. This manuscript will be interesting for the readers, but the novelty as a paper is weak. We agree the paper does not produce a novel remote sensing method per se; the emphasis of the paper is on expanding a previous framework to include a range of remote sensing tools and consider a range of scenarios relevant to the South Pacific context, so is aimed at an ‘improved approach’. There is novelty in the production of a simplified, objective, evidence-based toolbox and decision tree specifically for those who have less expertise in remote sensing but who have a need to map seagrass resources; the paper is submitted to the special issue on remote sensing for reef ecosystems in Oceania which considers papers that “applications for environmental monitoring and management. We do not set out to produce a review of seagrass mapping remote sensing tools, because this has been addressed in several other journal articles elsewhere; the paper may be better considered a “Communication” than a “Research Paper” or a “Review”.

Round 2

Reviewer 1 Report

Thanks for the authors’ responses.

All my questions and comments, exception of the mapping toolbox based on qualitative analysis not on quantitative index, have been addressed, and the responses would be promising. I agree that the qualitative toolbox is more accessible than the quantitative one requiring a high level of technical skill to navigate. However, What is the basis of the qualitative analysis? Also, as a toolbox, how to evaluate the qualitative toolbox?

 

 

Author Response

Hi there, thanks for the response. The two questions are: 

1. What is the basis of the qualitative analysis? The toolbox is for selecting remote sensing tools for mapping seagrass, it does not extend to methods for analysing the resulting data. Is this not clear from the manuscript, so perhaps we need to add a sentence to explain? Or have I misunderstood your question?
2. as a toolbox, how to evaluate the qualitative toolbox. I don't quite understand the question, can you elaborate on what you mean by evaluating the toolbox?

Thanks, Julie