Mapping Forest Stock Volume Based on Growth Characteristics of Crown Using Multi-Temporal Landsat 8 OLI and ZY-3 Stereo Images in Planted Eucalyptus Forest

Round 1

Reviewer 1 Report

Summary:  Thank you for the opportunity to review this great manuscript. This manuscript assesses the improvements of Forest Stock Volume (FSV) estimates of a Planted Eucalyptus Forest employing vegetation indices, band reflect aces and structural metrics representing growth characteristics. The study was carried out in using Landsat 8 OLI and ZY-3 Stereo Images and have used different machine learning and regression models to estimate the model performance in estimating FSV. The results shows that inclusion of composite images from before and after canopy closure improve the accuracy of FSV estimates. Further, the study shows that the saturation problem in the optical images due to canopy close after a certain period of plant growth can overcome into some extent by adding structure metrics such as canopy height derived from the stereo images. The study brings some values by analyzing images at different growth episodes separately and adding more structural metrics from stereo images to improve the accuracy of FSV estimates. However, there are a lot of inconsistencies in the methods, results and discussion which need to be fixed before publishing. Please read the below section to see detailed comments to revise the manuscript.  In addition, detailed captions for figures are required.

 Broader comments: Throughout the manuscript, there a lot of abbreviations used. Please introduce them in the first place they are using in the manuscript. There are a lots of inconsistent terminology used (composite, variable sets, data sets, difference images etc. ). It is required to be consistent in the terminology, so the reader does not get confused.  In addition, there are many pieces missing in the methods (see below comments) and in the discussion. In addition to these, all the figures require axes labels, detailed caption with explanations on all the pieces displaying in the figures.

 L 18: what growth characteristics (age, height etc.? )

L 18: new spectral variables?? Did not see any new variables (please add what that is here so the readers know before getting into details)

L20: what is CHM?

L21: use the extended version of these (RF, SVM, KNN, and MLR) instead of abbreviations in the abstract.

L25: what is RRMSE? re-write the sentence please. What does mean by increase or decrease in RRMSE (accuracy of estimation of FSV increased/decreased??). I think readers are more interested in that than how much RRMSE changed.

L27: overestimation of what??

L31: what is CCHM?

L30-L32: not sure what it is meant in this sentence. Please re-write.

L57: LAI is a structural Metrix and is not measured as other common vegetation indices are measured/estimated using reflectance. and I am not sure why the next sentence was about therefore and then reflectance..

L83: What is DEM stands for?

L104: reference for these estimates?

L115: replace “samples” with “plots”

L122-123: reference?

L124: Sample or plot?

L124-125: reference?

L143-146: put this sentence before the DEM sentence.

L155: Add standard band numbers in parenthesis

L167: reference for this operation?

L168: what are the four. Can see only three. Composite image of before closure, composite image of after closure, and difference image. Please be explicit so readers know what you are referencing here..

L173: any reference for what is forward, backward image means or just explain what that is.. simply how did you collect these images?

L179-181: how did you do this? used any software, programming language. please provide with reference.

L182: One is 2.1 m resolution and other one is 12.5 m resolution. Did you do any resampling? how did you subtract these images with mismatching resolutions?

L184: add any reference for these models and please provide some examples, use cases of these models and their values in the introduction....

L184: “with the following variable set”

L186: how did you got the 30x30? the plots are 20 x 20 m.

L230: Why are these four regions and how they were selected?

L250: are these indices and the bands are different mentioned in the previous section (table 1)? if so, say what indices or bands used here...

L264: BCC is also a composite image, right?? so this is average before and average after canopy closure??? Be consistent with terminology.

L272: Positive/negative?

L283: Are these from all dates single images or only from the reference image taken on 2018?

L297: what is this?

L362: what underestimated samples..

L414: composite images/ band combinations, BCC, DCC all are confusing..please be consistent with the terminology...

L426: what is DTM?

L427: what is deciduous season???

L432: Never talked about seasons before in the study other than growth episodes. please re-write this section and tell only what you did or change the terminology in the method. Do not introduce anything that you did not do as you did or considered here if they are not in previous sections, or they are from other research those help you for this discussion.

L441-442: RMSE increased or decreased???

L465: what seasons you are talking here.... growth episodes like premature, mature, near mature or phenological seasons like growing season, leaf mature or senescence?? be explicit based on what you studied...

L466: over- or underestimated what???

Figure 2: scatterplots between a) stand age and FSV and stand age and AHF etc. Also mention how these FSV was measured or if you got from a reference what is it? and its accuracy.

Figure 4: What is this red dash line and how that was derived?

Figure 5: What is this red dashed box, and red solid outline explain them in the figure caption.

Figure 6: label Y axis

Figure 10: what are those red and black lines in each plot. explain all features explicitly in those plots.

Figure 12: dataset or variable sets?? please be consistent.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

minor revisions according the attached pdf file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

the work is interesting, and well written,but need further improve is needed.

1.Differences between 4 different models ( (RF, SVM, KNN, and MLR) and lack of discussion on applicability。 This is the shortage the this work? Innovation is not obvious. The author needs to  highlight the innovation of this article.

2.need higher resolution images, e.g. Fig 1, 2, 5 etc

3.Formula 2 is not standardized, Every parameter needs to be defined.

The definition of ????? does not give clear context and lacks necessary explanation. Hard to follow, why need RRMS?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thanks for revising the manuscript and addressing all the reviewer comments. The manuscript has been significantly improved and can be accepted.

Reviewer 3 Report

The author has replied to all my comments, it now can be published in the current form.