Comparing a Random Forest Based Prediction of Winter Wheat Yield to Historical Yield Potential

Round 1

Reviewer 1 Report

Certainly a valuable piece of work. Worth publishing after correcting the indicated errors and addressing the specified issues. Several major and minor grammatical errors are in the text. I indicated them in the PDF with sticky notes. Tenses are used haphazardly. Quite a few details should be clarified… indicated in the PDF. Some references are indicated with numbers some are cited by the name(s) of the author(s). I believe the first method is required.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I found this paper to be extremely interesting and generally well written.  There are places in the manuscript where "trials" is spelled "trails", so the authors need to manually check the entire ms for spelling.

My concerns:  I am a bit surprised that the overall R2 was not higher for the new model when compared to the soil only model.  Related to that, I think it would be instructive to quantify, if possible, the contribution of each of the components listed below (those with an importance score of >50) to the overall R2 of the Random forest model, much as one might do in  a multiple regression model.  

Other than that, I have no concerns and will recommend publication with these minor revisions.

1. Temperature 100.0 +

2. Clay (0 - 30 cm) 91.6 +

3. Grow Days 77.8 +

4. pH (60 - 100 cm) 72.5 +

5. Precipitation 70.5 -

6. Clay (100 - 200 cm) 63.4 +

7. Geology 58.7 C

8. Drainage Class 57.5 +

9. FAO Soil Group 52.9 C

Author Response

Please see the attachment.

Author Response File: Author Response.docx