Dear Mr Deneu,

Thank you very much for submitting your manuscript "Convolutional neural networks improve species distribution modelling by capturing the spatial structure of the environment" for consideration at PLOS Computational Biology.

As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Arne Elofsson

Deputy Editor

PLOS Computational Biology

Arne Elofsson

Deputy Editor

PLOS Computational Biology

***********************

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: This paper describes a deep learning approach to species distribution prediction that uses spatial environmental information as input to a convolutional neural network (CNN). The study is well carried out with ablation studies and comparison to other methods. The two main problems with the paper are 1) the exposition could be more clear (see below) and 2) other works [19,21] have already applied CNNs to this problem. To address 1) it is suggested that the authors work more on the writing (perhaps consult a native English speaker). On 2) it should be made more clear what the novelty is of this work.

Minor comments:

1. the later -> the latter

2. Define punctual model properly first time used.

3. For practical reasons 6 (data required), most Species Distribution Models (SDMs) are correlative methods 7 relating known species occurrence data to potential environmental predictors [2–7]. 8 Popular examples of such methods include MAXENT [8–10], random forest [11] and 9 boosted regression trees [12–14].

MaxEnt etc are general regression method that can be applied to SDM.

4. Eq (1) needs more explanation.

5. The paragraph Predictions is unclear. What is exactly being predicted? I understand a softmax is being used. So it is multinomial classification. But the data is presumably occurrence of each species. What is the conversion here?

6. There are also other examples similar to 4. and 5. later in the paper in the same spirit that it should be possible for the authors to spot without be pointed to by a referee. ;-)

Reviewer #2: In this paper, the authors compared convolutional neural networks (CNNs), deep but non-convolutional neural network models (DNNs), boosted trees (BT) and Random Forest (RF) for predicting species distributions. They found that CNNs outperformed the other models for rare species. This demonstrated the usefulness of CNNs. The authors used top-k accuracy to characterize the performance of the models. However, this approach of model evaluation is difficult for species distribution modelling practitioners to understand because they always use the conventional model accuracy measures (including the area under the receiver’s operating characteristic curve, true skill statistic, sensitivity and specificity, etc.). It’s better also to give the model evaluation using these measures.

Spefics:

P2 L71: Table ??

P7 L193: “is apply at” may be changed to “is applied at”?

P8 L214: “are uses” may be changed to “are used”?

P10 L298-299: “the environmental neighborhood more than the punctual environment matters for prediction”?

P10 L309: “it’s”?

P11 L356: change “parcimonious” to “parsimonious”

Some information is lost for this reference:

52. Botella C, Joly A, Monestiez P, Munoz F, Bonnet P. Bias in presence-only niche models related to sampling effort and species niches: lessons for background point selection. 2020;.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Computational Biology data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example in PLOS Biology see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions, please see http://journals.plos.org/compbiol/s/submission-guidelines#loc-materials-and-methods

Dear Mr Deneu,

We are pleased to inform you that your manuscript 'Convolutional neural networks improve species distribution modelling by capturing the spatial structure of the environment' has been provisionally accepted for publication in PLOS Computational Biology.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Computational Biology. 

Best regards,

Arne Elofsson

Deputy Editor

PLOS Computational Biology

Arne Elofsson

Deputy Editor

PLOS Computational Biology

***********************************************************

Please make sure that the minor changes suggested by Reviewr #2 are corrected in the proofs

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The authors have substantially improved the paper and addressed the concerns of the reviewers in good manner as far I can see. I leave it to the other reviewer to address whether the evaluation metrics question has been satisfactory answered.

Detailed comments:

1. The figure should be made higher resolution.

2. Add zoomed in plots.

Reviewer #2: Comments to the authors

I am satisfied with the authors’ response. Now, I only have a minor concern. In P11 L352-353: The authors said: “It is important to note here that 36% of the species in the test set only have 1 occurrence. For such species the AUC can only be 0:0 or 1:0 and thus has a very high variance.” This is not correct. Even though there is only one occurrence, you have many (pseudo-)absences. Assume the prediction for the occurrence is p1 (between 0 and 1), if the predictions for all the (pseudo-)absences are less than p1, AUC = 1; if the predictions for all the (pseudo-)absences are larger than p1, AUC = 0; if the predictions are less than p1 for some (pseudo-)absences and are larger than p1 for the other (pseudo-)absences, 0 < AUC < 1. For example, if p1 = 0.8 (for the occurrence) and p0 = c(0, 0.1, 0.2, …, 0.9, 1) (for 11 (pseudo-)absences), AUC = 0.2273. The AUCs may have a very high variance, but they can be any value between 0 and 1. This is also true for TSS.

Specifics:

P1: (in Abstract) “Convolutional Neural Networks (CNNs) are statistical models suited for learning

complex visual patterns.” They are machine learning methods.

P3 L71: In “We use 33 environmental raster variables”, may use “used”?

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Computational Biology data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Ole Winther

Reviewer #2: No