Effects of Climate Change and Environmental Factors on Bamboo (Ferrocalamus strictus), a PSESP Unique to China
Round 1
Reviewer 1 Report
-This study assessed the effects of extreme climate and environmental changes on the extremely samll population species Ferrocalamus strictus in China. I have the following suggestions:
-A robust methodological approach is needed. For instance: As sample size increases, model accuracy should increase until achieving its maximum accuracy potential (Stockwell and Peterson 2002; McPherson et al. 2004). The authors only used 25 occurrence records, such a small sample size may lead to poor model accuracy. This should be discussed.
-As far as I am concerned, some Ferrocalamus strictus are cultivated. The authors should explain among the 25 occurrence records, how many records are cultivated. Because for a cultivated species, human activities play a decisive role in the diffusion and migration of this species. For example, Ferrocalamus strictus can be cultivated in cold areas by greenhouse, or in arid areas by irrigation. Further, 'climate tolerance and trait choice hypothesis' was used to explain the distribution of cultivated tree species. It is believed that the distribution of tree species is mainly determined by aesthetic characteristics (such as beautiful flowers) (mainly depends on decision-makers). In the discussion, the uncertainty and applicable conditions of the research should be discussed and explained (See Jenerette et al. 2016; Booth, 2017; Kendal et al. 2018)
-There is no mention of the background dataset used for the models. This information is critically important for evaluating the approach used by the authors. I assume that background was sampled from the entire grid, yet this is typically inappropriate (e.g., Elith et al. 2010; VanDerWal et al. 2009), and can lead to inflated performance metrics.
-Why the default settings of MaxEnt is used without proper explanations should be discussed (see Merow et al. 2013 "A practical guide to MaxEnt for modeling species' distributions: what it does, and why inputs and settings matter").
-Authors used AUC to evaluate the models. The use of AUC to evaluate presence-only SDMs has been highly criticized (e.g. Lobo et al. 2008, and much more recent literature). I highly recommend the authors to use an additional evaluation metric (e.g. TSS, pAUC, etc.) in addition to AUC to support their results.
-Splitting the data into training and testing dataset does not maintain independent evaluation due to spatial autocorrelation, as training and testing datasets can be very close to each other. This can be one of the reason for the very high AUC value reported in the manuscript. I recommend if the authors evaluate the model using spatially-independent data (preferably from bias-free presence-absence data from new field surveys) or, in the case that such data does not exist, to evaluate the model using spatial-block cross-validation. Please see: Radosavljevic & Anderson (2014); Fithian et al. (2015); El-Gabbas & Dormann (2018).
-Finally, the authors use a very subjective reclassification of Maxent's outputs; authors must explain why they used this strategy.
-The results and discussion parts, author need to discuss the ecological processes that influence the distribution of the studied species from your modelling results. However, what would be most powerful with these potential niche models would be adding the species dispersal constraint to, not only see how climate change will shift the species' niche, but to see how much the species could actually shift given species dispersal limitations…
Here's a link to the MigClim package for modeling species dispersal constraints:
https://onlinelibrary.wiley.com/doi/full/10.1111/j.1600-0587.2012.07608.x
https://cran.r-project.org/web/packages/MigClim/index.html
-In addition, modelling limitations are not discussed in details. Suggest writing a 'Limitation of modelling approach and future research directions' section in the discussion.
-In many places I found the language to be imprecise or vague, and there are numerous grammatical issues throughout (e.g. incomplete/meaningless sentences, poor word choice leading to apparently unintended meaning). I understand the challenges of writing in a non-native language, and this isn't intended to be malicious criticism. However, the submission requires careful proof-reading by a fluent English speaker with experience in the field.
Author Response
我们非常感谢您对我们的研究和手稿的认可和认可。我们也非常钦佩你严格细致的工作态度。我们已尽最大努力逐一回复您的意见和建议，希望修改后的内容得到您的认可。
Author Response File: 
Author Response.docx
Reviewer 2 Report
This study was conducted to examine the potential distribution patterns of one PSESP bamboo at different climate scenarios, with the expectation that this would provide conservation strategies for this species. The results of the study are instructive for the conservation of the species.
However, to my great disappointment, it was very difficult for me to read this article because of the very rough English, unclear expressions and the "difficult to see" figures. Here are some comments are as follows:
1. The word “extreme” used in the title and in the article does not reasonable. This is because the future climate results only present the climate under different scenarios, not necessarily the extremes, or at least not all of them.
2. Please carefully read the instructions for inserting literature into the manuscript of the journal. When there are multiple articles, they are sorted by author's letter or year?. eg. Line33, 40, 42.....
3. Line 45: “(Hetem.，2014; ” to Hetem., 2014. Comma under English characters. Please check the full text.
4. Line 33 “2012).. And attracted international attention”. there are two full points.
5. Line 63 Literature citations for Maxent model construction should be added.
6. Line 66-67, grammatical problem.
7. Line 69-78 This paragraph is invalid information relative to this study and it is recommended to delete this paragraph.
8. Line 83 120CM should be 120 cm.
Line 84,88 ”Green and Environmental Protection material” should be “green environmental protection material” or “environmental-friendly material”
9. Line 98 what is the mean of “ habitat needs”?
10. Line 104,105 The second point needs to be rephrased.
11. Line 108 this study will ......
12. On the whole, the Introduction part needs to be changed a lot. a.English language correction; b.logic problem in and between each paragraph.
13. Line 138, 230-238 F. Strictus need fontItalic.
14. Line 141 “In order to understand the spatial distribution pattern of F. strictus” should be “In order to understand the spatial distribution pattern changes of F. strictus”
15. In table 1 the line between bio1 and bio2 need to be deleted.
16. Line 153 “....into Maxent” to” ....into Maxent (Version 3.4.1, available at https:// .......)”
17. Line 155 The knife cutting... may be the jacknife cutting.....
18. Line 160 Which version of ArcGIS?
19. Line 160-161 Describe which function or method in ArcGIS was used to classify the habitat suitability level. Is there any evidence for such a classification.
20. Line 165 should be “detect the change of spatial distribution pattern.....”
21. In method part: a. Many parameters run by Maxent are not described ; b. Research on species distribution models(SDM) has developed rapidly in recent years, and studies using Maxent's analysis have gradually been replaced by new methods, especially for R packages that integrate multiple SDM models, for example” Biomod2”, “SDM”,”httflexsdm”.... Here, considering the urgent need for the species in this study to be conserved, we suggest the authors to re-run the analysis with a more comprehensive approach.
22. Line 169, 170 It is recommended to delete the first sentence.
23. Line 173 19985M to 1985 m.
24. Line 174 5.92Km2 to 5.92 km2
25. Figure 4 The species in this study are only distributed in Yunnan Province, China, but the author mapping the potential distribution in different periods is based on the scale of China, so it is difficult to see the distribution differences on the map. Although the author has made some analysis on the map, it is difficult for readers with good eyesight to see the differences. Therefore, we suggest that only the regions of Yunnan Province should be used when displaying the distribution patterns of different periods. Therefore, it is recommended to redraw Figure 4.
26. Figure 6 The Unsuitable area is not shown in Figure 6, but it is listed in the legend.
27. About the result part: I think there are many contents here that are more suitable for the discussion part. For example, Line177-181, 212-218, 225-228, 251-275. It is worth noting that according to the prediction results, this bamboo was distributed in Hainan during the LGM period(figure 1d). However, the author lost the description and explanation in the results and discussion.
28. The author analyzed the distribution area change of species under different climate scenarios, and expected to discuss species protection. However, the author neglected the loss of original habitat, which should be paid more attention to compared with the change of area data. Therefore, I suggest that the author compare the spatial changes of distribution areas in different periods.
29. It is suggested to reorganize the results and discussion parts of this manuscript.
30. Too many reference formatting issues. eg., 4,7,8,9,10,11,12,13,14...........too much.
In summary, I believe that this manuscript needs to be reprepared for the sake of responsibility for the quality of the papers published in Forests.
Author Response
We are very grateful for your approval and recognition of our research and manuscript. We also greatly admire your strict and careful working attitude. We have done our best to respond to your comments and suggestions individually, and we hope that the revised content will meet with your approval.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I have no further questions
Reviewer 2 Report
The author has made great changes according to the previous suggestions. I think the current version can be accepted. Please consider it.
