Identifying Variables to Discriminate between Conserved and Degraded Forest and to Quantify the Differences in Biomass
Round 1
Reviewer 1 Report
Review report
Identifying variables to discriminate between conserved and degraded forest and to quantify the differences in biomass
Overall
The manuscript titled "Identifying variables to discriminate between conserved and degraded forest and to quantify the differences in biomass" is an interesting article that focuses on determining the variables that can help in distinguishing between degraded and conserved forests. I believe it is a good fit for the journal. However, a number of issues will need to be addressed before it should be considered for publication.
Abstract
Line 12. The authors stated, "This paper presents the results of a statistical study…" But it did not include the study's objective in the abstract. I suggest that the first sentence should present the aim of the study. for example, "this work focuses on …." Then state the objective. It is essential to state the aim of the work in the abstract.
Introduction
Line 46: I preferer going with the latest citations. The study that was cited was presented in 2013. It would be good to add a newer citation discussing the changes in forest degradation.
Line 92-99. The author discussed the remote sensing techniques and sensors used to assess canopy cover. However, the work did not include any remote sensing analysis. I believe that the remote sensing part is making the introduction a little less tight. I think cutting a little down there and just making it shorter and denser will better represent the massage of this work.
I also suggest focusing more on studies discussing statistical analysis used in similar work. This is important since the work mainly focused on comparing the two types of forests using different statistical methods.
Materials and methods
Line 113: I prefer explaining a bit more about what is ejidos. Readers from other countries and regions may not be familiar with it.
Since the paper is mainly depending on the statistical analysis, I believe that the methodology section needs to be more substantial, especially the statistical methods.
- The experimental design was unclear. What is the number of samples, replicates, etc..
- The authors mentioned that they selected 43 plots, 25 for the conserved, and 18 for degraded. Why are they different? I can tell that these are the number of replicates. I think the number of replicates needs to be similar. Increasing the replicates at one site may impact the results as the number of samples at the conserved will be larger than the degraded. Therefore, I suggest excluding plots from the conserved forest to make them 18 plots for both. Then, explain in the methodology that each forest type has 18 plots as replicates.
- Also, the size of the plot was not mentioned. What was the size of each plot?
- What is the number of samples in each plot? It is also unclear how structural parameters were measured at each plot. I believe more details need to be added to this section. For example, how was the number of trees measured at each plot? Did they use satellite imagery, field survey, and how they were carried out? How were the heights measured, did they use remote sensing techniques, or they were collected from elsewhere? They need to explain how each of these parameters was collected from the plots. This is super important since the results mainly depend on the data collection of these parameters.
Line 141. the following sentence, "Of the total 43 plots, 25 plots were of conserved forest and 18 degraded" needs to be moved to line 130. immediately after the number of plots. It will confuse the reader how it is written because it says in line 130, the total number of 43 plots, but it is unclear if it is 43 for each site or both. Therefore, moving the number of plots at each forest type will make it clearer for the reader.
1ine 164-178. This section covers the statistical analysis. Since more than one statistical method was utilized, I suggest including the hypothesis in this section and the statistical techniques used for each hypothesis. Then the results section will follow these hypotheses. This will improve the flow of the results and discussion of the paper.
Line 175. How was the threshold determined, and the accuracy calculated? Please provide more details.
Results and discussion
Line 232. Why was Kendall tau correlation applied? More justification needs to be discussed. For example, why didn't the authors use stepwise regression analysis, since they are dealing with multiple variables?
Line 251. The authors mentioned that the "Density of trees and density of branches is also highly correlated (τ = 0.57, p < 0.001)" this makes it a bit confusing. It seems from the objective that the work focus on understanding the variables that are influencing the AGB (dependent variable). This means that the authors are testing more than one hypothesis. Therefore, I suggest adding the hypothesis in the methodology section, as I mentioned in my previous comment. Then, follow the hypothesis in the results section. This will improve the statistical section, which is a core section of this paper. It will make it easier for the reader to understand the dependent and in-depended variables for each hypothesis.
Line 256. the following statement is unclear "There is also a good correlation between AGB and canopy cover (τ = 0.4, p < 0.05)" for which forest type this result refers?
The discussion section requires more citations to support the findings.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Paper deals with assessment of variables to discriminate conserved and degraded tropical dry forest in Mexico. Field data from 43 plots (0,05 ha) were used. I consider this topic as eligible, but presented work is by my opinion little big insufficient with no important scientific or practical addition. Authors found that “tree height and number of trees correlate best with biomass and that change in these variables would better estimate changes in aboveground biomass than canopy cover“. This fact is elementar and trivial. Authors mentioned about using of Remote sensing by quantify of forest degradation, but they presented only field assessment without interconnection. In Methods I lack for specification of degraded and conserved forests and methods for their classification. Where is limit between conserved and degraded, “although none of the area has been completely free from human disturbance “
Sentence “Plots were selected ..in collaboration with local forest officers who are very familiar with the territory and could clearly identify which areas were degraded” I consider as unsuitable in methods description in scientific journal.
In each plots were collected tree parameters, it is clear. How was estimated canopy cover?
Values of Standard deviations in Table 1 for conserved forests are very low, are they really correct? Variability of BA then was only 7% and by Mean Height only 3,5%. How was calculated mean height? Presented values are very low! 5 meter is limit of FAO definitions of forest.
Presented Figure 1 and 2 I consider as same, they express similar results. Chapter Result is by my opinion with their volume and condition weak and insufficient and required more changes and supplements for publication in Forestry journal. Too chapter Discussion can be massive improved.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Dear authors,
Firstly, I would like to thank you for presenting your results. The manuscript provides an interesting analysis of forest characteristics in two types of forest: conserved and degraded. So, I think you did a good job. The study design setup, the applied methods are adequate and allow to obtain reliable results. The article is understandably written and well- organised.
However, I have editorial comments.
- Citations in the text and literature do not comply with the requirements of Forests [see instruction for authors]
- Two reference from the list of references are missing in the text:
- Joseph, S.; Murthy, M.S.R.; Thomas, A.P. The progress on remote sensing technology in identifying tropical 426 forest degradation: a synthesis of the present knowledge and future perspectives. Environmental Earth 427 Sciences 2011 64 731-741.
- Rodríguez-Veiga, P.; Saatchi, S.; Tansey, K.; Balzter, H. Magnitude, spatial distribution and uncertainty of 486 forest biomass stocks in Mexico. Remote Sensing Environment 2016 183 265-281
Therefore, the term for my review is "Reconsider after minor revision”.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I want to thank the authors for the hard work. The comments were appropriately covered, and the manuscript is improved. The manuscript is tighter, and the quality of the figures is good and readable. I only have one minor comment:
The quality of figure 4 is quite low. I suggest enhancing the figure quality, especially the y and x-axes.
Author Response
Thank you very much. We have enhanced the quality of figure 4, and made sure all figures have at least 300 dpi.
Reviewer 2 Report
Authors have made changes and relevant improved delivered paper. I continually believe, that whole paper can be anymore improved. But I have at the moment no other concrete comments and suggestions.
Author Response
Thanks. Based on the foundation laid in this paper, in the future work, we will incorporate both optical and radar data in the estimation of emissions from forest degradation.
