Assessment of Aquatic Reed Stands from Airborne Photogrammetric 3K Data

Round 1

Reviewer 1 Report

This study proposes an approach of assessing aquatic reed stand status such as heights, area, density, etc. from the DLR 3K-system. There is a useful contribution to researchers and engineers to evaluate such status, and I recommend the manuscript to be acceptable with minor revision.

General: From the total RMSE of 70 cm, the reviewer is wondering if the effect of light refraction would appear in the DSM under the water surface? Did the authors consider any effect of light refraction in the process of Agisoft Photoscan? If so, please write it in detail.

Line #69: Please explain what “OA” is.

Line #204: The word “DSM” appears here for the first time, so “digital surface model (DSM)” should be written here.

Figure 3: Same as line #204. It is better to write “OHM (object height model)”.

Table 3: Please check the value of “User’s accuracy [%]” like 0.71%.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper uses photogrammetric tools for assessment of aquatic reed stands. The paper is well written and presented. Some comments are:

1) the computational cost should be given

2) the effect of the parameters of the photogrammetric tools should be given.

3) some papers for aquatic photogrammetric  analysis should be cited

Skarlatos, Dimitrios, et al. "Project iMARECULTURE: advanced VR, iMmersive serious games and augmented REality as tools to raise awareness and access to European underwater CULTURal heritagE." Euro-Mediterranean Conference. Springer, Cham, 2016.

JOHNSON, LUCINDA, and STUART GAGE. "Landscape approaches to the analysis of aquatic ecosystems." Freshwater Biology 37.1 (1997): 113-132.

Clarke, Sharon, and Kelly Burnett. "Comparison of digital elevation models for aquatic data development." Photogrammetric Engineering & Remote Sensing 69.12 (2003): 1367-1375.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper is well designed and properly presented. Nevertheless some critical points still persist, suggesting a Major Revision (actual a Minor with constraints) They have been addressed as comments within the pdf.  I expect that they are carefully taken into consideration to properly finalize the paper, that is greatly promising and already close to its final state.

Biggest criticalities concern photogrammetric aspects related to image processing and alignment and ground measure positioning. An evident weakness is also the one related to the accuracy assessment of classification that is based on an auto-referential approach.

I greatly encourage also the deepening of the description of the propsed index aRSI making evident all therationals behind its defiition.

Please provide answer to all my comments reported within the pdf

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

Abstract

Line 13 replace “By a” with “Using”.

Introduction 7

After reading this entire section, I was left with one fundamental question. What is Reed status? This is unclear in the introduction, and this shouldn’t be the case considering that is the main objective of this assessment. The authors need to better explain why photogrammetry is important for measuring the biophysical attributes of wetlands and coastal floral management, the basis of 3-Dimension modelling on critical assessment of aquatic landscapes. Most of the required information are already presented, for instance, L59 – L74, however, it needs little refinement to clearly explain to the potential readers of RS Journal why Reed status assessment is important and why very high-resolution aerial photos are suitable alternative to very high-resolution satellite images.   

Line 76 – 77 Authors may want to highlight the fact that different angles for the cameras aid in getting a stereoscopic view of the landscape. I think they have inferred to this by saying 3D models could be derived but a more technical description is likely to be helpful to readers of Remote Sensing Journal.

Method

Overall, this section is very well written.

L172 How are you able to distinguish abiotic damages due to driftwood from other abiotic influences (e.g., wind or waves)?

L232 – 233 An excellent way of describing reed status and should probably be mentioned in the introduction.

L303 – 305 Why will the threshold values be adjustable? It seems that the equation on L295 requires several metrics to estimate an index which the authors stated on L302 to be equal to a threshold.  If the authors mean a partition of the estimated index values into thresholds, then this needs to be clearly stated, otherwise the entire sentence becomes very confusing.

Results

Figure 7 I suggest that the authors change the figure labels for the “reed density” layer to high (for dense) and low (for sparse). That way it makes sense to interpret the information as “high reed density” and not “dense reed density” and vice versa.

Section 3.5 The authors should in addition to the accuracy assessment provide the proportion of the area of interest that can be attributed to be healthy or stressed.

Discussion

L466  - 467 Is the water level measured during data capture (517.96m asl) significantly different from average historical measures of 518.20 m asl? If not so, there is no need to attribute the influence of this factor on reed extent.

Overall, the discussion section is lacking to critical aspect. One is the implication of this assessment for aquatic and coastal landscape management. Does this mean less time spent monitoring coastal areas? Can the modelling framework improve decision making? Through accurate identification of areas under stress or those undergoing fast deterioration. The other aspect missing in the discussion relates to the accuracy reported. For the different metrics and one indicator derived, authors report varying accuracy ranging from 60 – 90%. What this means is an uncertainty of 10 – 40%. This needs to be addressed and broader context provided to help attribute the value of remote sensing technology for coastal features management and intervention.   

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

None of my comments have been addressed. 

The revision would take less than one hour for the authors. 

Reviewer 3 Report

Thank you for having provided answers to all my comments. I still see some criticalities concerning the level of detail about image processing, but your considerations that moved the point to the final results, are enough for making the paper acceptable. I suggest, for your future works, to increase consciousness in software and sensors features and avoid a too large use of references to fill this gap.