Estimation of Potential Suitable Habitats for the Relict Plant Euptelea pleiosperma in China via Comparison of Three Niche Models

Round 1

Reviewer 1 Report

2023, May 1st

 

I have completed my review of the manuscript entitled “Prediction of potential suitable habitats for the relict plant Euptelea Pleiospermum in China via comparison of three niche models” submitted to the Journal Sustainability. I found this research important and timely, well conducted, and to fall within the scope of Sustainability. The authors developed species distribution models to estimate the principal climatic and environmental factors influencing the distribution of a near extinct relict tree species in China ; and investigated the effects of the projected climate change in its potential suitable habitats. Overall I enjoyed reading it and suggest to accept it with minor revisions. I offer a few suggestions for considerations. Please review carefully the points 1 to 7, others (a to h1) are mostly detail and typo. Best.

 

(1) In climate change modeling, uncertainties are inherent to climate models, bias-corrections and resolution used to represent the complex and chaotic nature of the climate variability. The present study used the medium spatial resolution BCC-CMS2-MR climate model from the Beijing Climate Center. Why a higher resolution version (BCC-CSM2-HR available at 0.25x0.25°) or another downscaled climate model has not been used instead should be discussed. A major limitation of the present study also relies on the use of a single climate model which hampers the ability to draw confidence intervals around the future climate projections since it results from the same atmospheric engine. Usually future climate projections are assessed using a climate model ensembles of 5 to 10 Global Climate Models using the same simulations parameters (realization, initialization and physics) to assess the uncertainties in our ability to model the future climate variability.Theses two major limitations (coarse spatial resolution + single climate model) should be emphasized in the manuscript to draw the limit of the study before its publication (and maybe state that further research is needed to quantify those uncertainties, and that this study constitutes a first step in understanding the projected trajectories of Eupetelea pleiospermum toward the 2100 horizon)

 

A following concern is the sudden change in the projected centroid of potential habitats presented in Figure 8 and compared to Figure 7. How to explain, for the 2100 horizon,  the dissimilarities between the sharp migration of the potential projected habitats under the SSP585 climate scenario vs the similar potential distribution of Eupetelea p.projected by DOMAIN ? This should be discussed somewhere in the manuscript. 

This sudden change in centroid as compared to the grouped centroid projected using the other climate scenarios may also be the result of a climate-model-specific spurious behavior no?

 

(2) The responses of Eupetelea pleiospermum to the selected environmental and climatic predictors in the MaxEnt model demonstrate to be non-linear interactions (Fig. 4). How in this context, justify the sequential removal of covariate predictors using the highest linear correlation coefficient (Pearson) ? A more conceptual approach investigating the effect of linear, non-linear and covariant predictors could have been implemented to improve the modeling of the system.  For instance, a Random Forest framework trained with the Boruta algorithm (B-RF) would have been more indicated. Please at least state this limitation in the discussion. BRF might also help to classify predictors according to their relative importance in the modeling, as compared to (1) other predictors and (2) to randomized versions of the predictors. Which leads me to the following concern :

 

(3) Emphasis should be made in the manuscript, and to a higher extent than at lines 234-237, that only one SDM model has been used to describe the relative importance of the selected climatic and environmental predictors; and that, since the MaxEnt model demonstrated the lowest values of model accuracy evaluation compared to BIOCLIM and DOMAIN, this representation of each predictor contribution might be biased. I’m thinking of the first lines of the conclusion for instance, stating that temperature was a dominant factor in the distribution of the tree species. The conclusion especially might benefit from adding something like : "L 576... x and y were the most critical factors, although additional studies are needed to determine the specific contribution of each factor in more than a unique model, and particularly in an approach using a multi-SDM-model ensemble." Same for future work needed to determine if the distribution of suitable habitat issued using the BCC-CSM2-MR is comparable to that projected by a more robust ensemble of (and / or downscaled?) climate models. This should put in perspective and stated clearly in the manuscript;

 

(4) In the introduction, L. 83 and following, the paper might benefit from highlighting the traditional medical uses associated to Eupetelea pleiospermum as well as that, delimiting the habitat of Eupetelea pleiospermum might also benefit to address conservation issues related to the endangered fauna & flora associated to those forests and habitats. Another related point (maybe for the conclusion or discussion ?) is that, habitat modeling, even using SSP scenarios does investigate the potential distribution of suitable habitats, but does NOT reflect any change in, for instance, urbanization trajectories. SO, the shrinkage projected using the climate + environmental factors might be even higher considering the future societal needs for land and related changes in land uses in China; thus, further highlighting the need to identify asap long-term conservation areas.

 

(5) The discussion could benefit from some cleaning as it is mixed with results (e.g., L.433-442; L501-511) and some presentations of other studies (such as paragraph L.406-430) that I think can be more adequately framed to address and contrast the current results and methodological limitations. The raised concerns (1), (2) and (3) should also be discussed and placed in the perspective of other research that has studied the modeling of tree species distribution. 

 

(6) Throughout the manuscript : ‘prediction’ is conflated with ‘projection’. ‘Projection’ should be kept for estimating the future trajectories of the species in the context of climate change ; while ‘prediction’ should be replaced by ‘estimation’ or ‘modeling’ in the context of explaining the model’s ability to represent the system. This also applies to the title of the manuscript…

 

(7) Throughout the manuscript, readers are left with no details on how the DOMAIN, BIOCLIM, and MaxEnt species distribution models were calculated? Was this study done using the R package dismo and what version? Or was it done using specific software? This should be added to the article, with the appropriate reference(s) before publication.

 

(8) at some point of the paper, in discussion maybe, it should be stated that the potential habitat distribution modeling most likely reflect the overall 'general' population signal of the species based on 'general' environmental and cliamte forcings. For considering future conservation scenarios, further analysis of specific response of tree individuals, or regional tree sub-populations to environmental and climate might help to capture a broader spectrum of specific responses to environmental and cliamte constraints that may (or may not) differ from the overall population 'general' response. For instance, genetic or phenotypic adaptations may allow some individuals to better cope with some specific forcings on tree growth. I think this will bring perspective of the work done in this paper, and on the extent to which the resutls could be used in active conservation :)

 

 

Other concerns and details :

 

(a) Eupetelea pleiospermum is sometime referred to Eupetelea pleiosperma:  which one is the good one ? 

 

(b) the use of “various studies”, “numerous studies”, “many studies, many species” … is annoying throughout the text. Please develop the research / methodological ideas instead, since bringing the argument of the amount of research papers associated with a given topic is irrelevant. 

 

(c) L. 32. which has AND WILL pose …

 

(d)  L. 36 tree species distribution ? L. 42 tree species ? Please specify since the modeling of animal vs plant species is quite different.

 

(e) L. 44 the next few years ? what is the timeline ? 2050 horizon ?

 

(f) L. 51 Please rephrase toward: the most abundant species diversity, ancient and complete succession sequence …  to many ‘most’

 

(g) L. 70 ? Common how ? for who ? L. 62 to L.82 are mostly methodological aspects … I suggest to move this paragraph into method, or at least, to interchange it with the Euptelea p. paragraph L.83-102

 

(g1) L. 83. Please state here clearly that this is the definition of a ‘relict plant’.

 

(h) L. 86 I suggest not using the ‘climate change’ terminology here to avoid any confusion

 

(i) L.92 “Therefore, it is of certain value to perform studies … palaeoflora and palaeogeography” this sentence is irrelevant, please develop the idea or remove it.

 

(j) L. 100 What about the classic UICN Red List of Endangered Species ? 

 

(k) L. 122 Line introduced by mistake maybe ? : “Ethical approval code” 

 

(l) Figure 1. Please state in caption that this is Central China. What does the insert here refer to? Are these islands ? 

 

(m) L.136 This should be NASA SRTM, not STRM

 

(n) L. 148-L153 this paragraph is unclear, how many predictors were introduced before this linear correlation screening ? So the screening was done on a correlation matrix of the complete predictors, removing sequentially each covariate, keeping the one covariate demonstrating the higher r coefficient right ?

 

(o) L. 158 remove space in cumu___lation

 

(p) Figure 2. Please specify : The AUC (left) and Kappa (right) values of the ..; Why the values below 0.55 are not presented ? ?

 

(q) L.238 and after. Please name the variables in the text instead : minimum temperature of the coldest month (bio6), annual total precipitation (bio2) … 

 

(r) Figure 5. What is with the lower inset presenting islands ?  Is this Philippines ? Please delete or state clearly what this is.

 

(s) L. 298, L.305 not predicted, estimated

 

(t) Figure 6, Figure 7, remove inset. Figure 8 Then on the inset : where is located the main map presenting the trajectories of change in centroid ? 

 

(u) L.406-430. This paragraph brings little to the discussion since it is mostly results from other studies, with no perspective to the current results ? Could this paragraph be rephrased to better highlight the current limitations of the modeling, instead of bringing limitations of other sutdies ?

 

(v) L.433-442 and L501-511 are repetitions of the results with no perspective from similar research ? I would reduce them significantly.

 

(w) L.452 “E. pleiospermum prefers …. resistance and drought susceptibility.” miss a reference for that statement. Same for L.457 : “based on the previous studies” ? which references ?

 

(x) L. 458 delete “extremely” and delete or reword ,‘revealed that our predicted results were of great reference value”. Instead I would probably say something like : The robustness of our results were confirmed by x and y studies.

 

(y) L. 465. Should Taxus Linn be italicized ?

 

(z) L.473 limiting the distribution range of what species ? E. pleiospermum ? and following L. 482.”... had negative impacts of the growth of subalpine trees”, such x and y trees growing in a similar environment as E. pleiospermum in this study right ? 

 

(z) L. 487-490 not sure which results are these ? PLease specify

 

(a1) L. 496 In reference to which timeline ? which reference period and horizon ? 

 

(b1) L. 501 and following I would replace increasing trend and decreasing trend by ‘expand’ and ‘shrink’ or some other word meaningful in terms of change in habitat surface.

 

(c1) “varied to some extent” could be more specific…

 

(d1) L.552 “based on the results of this study” : which ones ?

 

(e1) L554 and following. I would reword this section to try to place other conservation efforts in light of the current findings to support the need for action. I mean, instead of using this and that SHOULD be done. It would be better to say: conservation efforts x and y have proven effective for a and b reasons and in a similar context. The establishment of a seed bank or other project… could thus be beneficial for the protection of the species in the context where projected climate change envisions a shrinking habitat for E. pleiospermum… 

 

(f1) L.563 Typo. remove extra space

 

(g2) there is no acknowledgments section. Is that ok ?

 

(h1) Data availability statement. This is bold since all the data are not reported in the manuscript (SDM calibration outputs and projection ?) … better to say that all links to input data are reported in the manuscript and all output data are available upon request to the corresponding author. Or something alike.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

A very valuable research has been done that deserves to be accepted. Such research is needed to help preserve endangered species. It can be accepted after major revision and making all the following corrections:

 

-          L38-40; Check the sentence and rewrite it: “It has been indicated that….”

-          Add ref for this sentence: L83: “Euptelea pleiospermum Hook. f. et Thoms is a small…..”

-          Add ref:

L38: Yang, Y., Li, T., Pokharel, P., Liu, L., Qiao, J., Wang, Y., An, S., Chang, S.X. (2022). Global effects on soil respiration and its temperature sensitivity depend on nitrogen addition rate. Soil Biology and Biochemistry, 174, 108814. https://doi.org/10.1016/j.soilbio.2022.108814

L44: Yang, Y., Dou, Y., Wang, B., Xue, Z., Wang, Y., An, S., Chang, S.X. (2023). Deciphering factors driving soil microbial life-history strategies in restored grasslands. iMeta 2: e66. https://doi.org/10.1002/imt2.66

L53: Wang X, Wang T, Xu J, Shen Z, Yang Y, Chen A, Wang S, Liang E, Piao S. 2022. Enhanced habitat loss of the Himalayan endemic flora driven by warming-forced upslope tree expansion. Nat Ecol Evol. 6(7):890-899. doi: 10.1038/s41559-022-01774-3

L59: Chen, D., Wang, Q., Li, Y., Li, Y., Zhou, H., Fan, Y. (2020). A general linear free energy relationship for predicting partition coefficients of neutral organic compounds. Chemosphere, 247, 125869. https://doi.org/10.1016/j.chemosphere.2020.125869

-          Although the discussion is well organized, it needs to be improved and focus on the main innovations and achievements of the current research.

 

-          MS needs to be completely checked by an expert in English so that spelling and grammar errors are completely removed.

MS needs to be completely checked by an expert in English so that spelling and grammar errors are completely removed.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Initially, I would like to congratulate the authors for the excellent work submitted to the journal. The importance of discussing the discussed topic is of unique relevance to environmental science as a whole. However, the work is lacking in some aspects, which end up weakening it too much.

In the introduction, it would be interesting to insert a map of the distribution of the “relics of the Tertiary” species (lines 44-49), as well as photos of the plants, especially those listed in lines 53 to 57, and lines 82 and 100, 107-112.

The work is very well structured. Congratulations to the authors.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

Manuscript "Prediction of Potential Suitable Habitats for the Relict Plant Euptelea Pleiospermum in China via Comparison of Three Niche Models" study has been planned and performed in accurate way, data generated is also relivent to the study. Quality of figure 1, 2 and 8 must should be increased. Remaining figure and tables are fine. Reference formatting should be checked to adhere with the journal guidelines.

Manuscript "Prediction of Potential Suitable Habitats for the Relict Plant Euptelea Pleiospermum in China via Comparison of Three Niche Models" study has been planned and performed in accurate way, data generated is also relivent to the study. Quality of figure 1, 2 and 8 must should be increased. Remaining figure and tables are fine. Reference formatting should be checked to adhere with the journal guidelines.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

MS is well-edited and can be accepted in its current format.

Author Response

Thanks for the comments sincerely. Your comments have been of great help to us.

Reviewer 3 Report

 

The authors made the corrections and suggestions given to improve the article. The presentation of the results and the quality of the graphic material has been adjusted and improved. Therefore, I recommend publishing the manuscript. Congratulations to the authors.

Author Response

Thanks for the comments sincerely. Your comments have been of great help to us.