Multi-Species Inference of Exotic Annual and Native Perennial Grasses in Rangelands of the Western United States Using Harmonized Landsat and Sentinel-2 Data

Round 1

Reviewer 1 Report

Review:

The paper describes an ambitious project to measure abundance and distribution of a range of exotic annual grasses and native grasses. Grasses are difficult to identify and discriminate on the ground, so undertaking this task with remote sensing is very difficult. Despite this, the authors have been successful.

The paper is very well written with few errors.

I have the following concerns / comments:

Table 2 is not cited.

Supplementary file 2 appears incomplete. Is this because of the conversion to PDF? I think its supposed to be an animation?

There are only confidence intervals shown for TACA8 in Figure 4. Have the others been omitted or are they smaller than the blue line, and thus can’t be seen?

Lines 364 to 365 talks about minor positive deviations for Medusahead in 2019 (figure 5). I can’t see these deviations in figure 5. Is figure 5 too small, or is the wrong map inserted?

Lines 371 to 373, and elsewhere in the paper, reports changes in in abundance to two decimal places of a percent.

 

If I have read your paper corrected ‘abundance’ represents percent cover (Fig 4 and Fig 5). Given the margin of scatter around the predicted vs true values in figure 4 is often at least 20%, it seems inappropriate that you report changes in abundance to two decimal places of a single percent (lines 371 to 373, and elsewhere in the paper). I don’t think this is necessary or realistic. Further, you need to add commentary somewhere in the paper about the precision of the cover (abundance) estimates. I realise that the value of the model is in the average prediction, but particular plots could differ in abundance by a substantial margin. Figure 5 and associated commentary indicates that substantial changes in abundance occur within the scale of a +/- 10 (I assume %, but its not stipulated in the figure or caption) deviation from the five year median. From a biological point of view a change in cover of 10% is very minor. Is the output of your model not really an abundance metric of some sort rather than a percent cover? Some discussion about this also needs to be included in the paper.

Line 391 to 393 contain “(lowest)” after the word highest, and “(greater)” after the word little. I do not understand what these mean or how they fit into the sentence. What are is the sentence trying to say?

Lines 416 to 424. Likelihood of predictions down to two decimal points of a single percent seems unnecessary.

Last sentence in caption for Figure 7 need some work. Do you mean: “Values in boxes are percent accuracy, values inside () are number of observations”.

Lines 425 to 442. I can see the lack of correlation referred to for Sandberg bluegrass (lines 441 to 442) in Figure 8, but I cannot see the prior two correlations (cheat grass and EAG with total hydrologic year precipitation, and higher abundance of EAG for the wet years). I think Figure 8 needs reworking somehow to show these correlations.

Supplementary 3. POSE is missing from the annual abundance maps.

 

Author Response

Dear Reviewer 1,

We are submitting the revisions for the manuscript titled: “Multi-species inference of exotic annual and native perennial grasses in rangelands of the western United States using Harmonized Landsat and Sentinel-2 data”. Taking into account the comments made by reviewers, we have made updates to the manuscript. They are noted in track changes in the manuscript document. As a result, we believe the manuscript has been improved from the previous version that was submitted. Your comments and recommendations have been addressed and our responses are provided in red font in the attached document. The line numbers may have changed from the previous version, but we provided the latest numbers in our responses to the manuscript document with track changes.

We are grateful for the comments and feedback and hope you will agree the manuscript reflects a contribution to the field of Remote Sensing, particularly to the Ecological Remote Sensing issue.

 

All best,

Devendra Dahal

and co-authors

 

Author Response File: Author Response.docx

Reviewer 2 Report

see attached file

Comments for author File: Comments.pdf

Author Response

Dear Reviewer 2,

We are submitting the revisions for the manuscript titled: “Multi-species inference of exotic annual and native perennial grasses in rangelands of the western United States using Harmonized Landsat and Sentinel-2 data”. Taking into account the comments made by reviewers, we have made updates to the manuscript. They are noted in track changes in the manuscript document. As a result, we believe the manuscript has been improved from the previous version that was submitted. Your comments and recommendations have been addressed and our responses are provided in red font in the attached document. The line numbers have changed from the previous version, but we provided the latest numbers in our responses to the manuscript document with track changes.

We are grateful for the comments and feedback and hope you will agree the manuscript reflects a contribution to the field of Remote Sensing, particularly to the Ecological Remote Sensing.

 

All best,

Devendra Dahal

and co-authors

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript submitted for review includes an analysis of the feasibility of using Harmonized Landsat Sentinel-2 for grass density analysis of the Western US. The authors focused on analyzing the occurrence of species of conservation interest or a suite of species and used available public data for the analysis. The method of analysis, results and conclusions are correct. The text is clear and well written. I have no comments on the submitted manuscript. 

Author Response

Dear Reviewer,

We are submitting the revisions for the manuscript titled: “Multi-species inference of exotic annual and native perennial grasses in rangelands of the western United States using Harmonized Landsat and Sentinel-2 data”.

We are thankful for reviewing the manuscript and suggesting for approval. However, we made substantial updates to the manuscript, and we hope you will agree with the revised version of manuscript and its contribution to the field of Remote Sensing, particularly to the Ecological Remote Sensing.

 

All best,

Devendra Dahal

and co-authors