Site Index Modeling of Larch Using a Mixed-Effects Model across Regional Site Types in Northern China
Round 1
Reviewer 1 Report
The work proposed by the Authors is interesting and methodologically correct. I believe that the work is pretty good and just few adjustments are necessary, in my opinion. Here the suggestions:
1) In the ABSTRACT I suggest to check the L11-13 that are quite obvious and L13-15 are unclear. Then I also suggest to add the Site index classes also here (see comment 5)
2) The Introduction is fine, I just suggest to check L77 replacing the word “material” with “wood” or “timber” and to replace “resistance to bad climate” with something like “plasticity”
3) I suggest to modify the sentence on L179 because it is very unlikely (and Tables 4 confirms it) that all the 4 indicators could be optimal all together. Actually you can have a model with the lowest MAE but a slightly higher RMSE, so I suggest to use just one of them… I also suggest to check Table 4 since the first 2 bu also the 5th MAE are zero, so a perfect FIT?
4) On L221 “were converge” should be “were converging” or “converged” or what else?
5) Although the M8.7 with 8 classes was the best I believe a trade-off between science and usage is mandatory. In my idea 5 classes could be enough (and R2 is not really different between 5 and 8 classes) ans I also suggest you, in the end, to calculate the Site Index value for each of them as the height at age 50. So the new names could be (as for example SI9, SI12.5 etc…)
I hope it helps
Author Response
Please see the attachment
Author Response File: 
Author Response.doc
Reviewer 2 Report
the study uses mixed-effects modelling, a very popular approach recently
the paper is well designed, data and results presented clearly
abstract needs rephrasing as it seems to be a bit chaotic and some language errors can be setfound
Author Response
Please see the attachment
Author Response File: Author Response.doc
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
Review of manuscript forests-1632199 “Site index modeling of larch using a random-effects model across regional site types in northern China” by Guangshuang Duan et al.
Site index is modeled for 394 Larix principis-rupprechtii plantation sample plots, clustered to 8 site type groups, using a logistic mixed-effects polymorphic site index model. The elaborate three stage model selection includes (i) comparison of 9 different base models, (ii) comparison of 7 different random effect combinations and (iii) comparison of 3 different cluster approaches. The final model yields a remarkable high coefficient of determination (R²=0.87) and enables an accurate prediction of site productivity.
The manuscript offers a solid contribution to the field and is interesting to wide audience. The applied methodology is appropriate, and the results justify the conclusions. The manuscript is clearly structured and generally easy to read. Introduction, Discussion and Conclusions are principally flawless; however, a little extra effort must be spent on a clearer presentation of the methods and results (please see my detailed comments attached).
Summarizing, I recommend MINOR REVISIONS.
----
Title: Please consider changing "random-effects" to "mixed-effects"
L86: “third largest conifer species” please clarify, if you refer to the size of single individuals or to the area of distribution.
L113 & L120/Tab1: Is “Neimenggu” a synonym for “Inner Mongolia”? If so, please stick to one name. If not, please state the number of plots in Inner Mongolia in Tab1 and add Neimenggu to the description of the study region (L103-118).
L121-130: Please add a sentence or two, clarifying if any agents (fertilizer, herbicides, or insecticides) that might influence the results were applied on the plantations.
Tab3: I guess it should be “model” instead of “molde” (8 times).
Tab3: Please use a consistent notation in the formulas, i.e. do either use × or · for multiplication and ex or exp(x) for the exponential function, but do not mix different types of notations.
L166 Please provide a link or reference to that Forstat software, the reference to a rather old course book in Chinese language is not sufficient to find the software.
L182 Please define TRE in section 2.5, just like you did with R² (eq 2) and MAE (eq 3)
L183-186 Please add the p-values (or significance stars) to the parameter estimates in Tab4
L195 Please add the p-values (or significance stars) to the parameter estimates in Tab5
L198: „significant improvement“ In case you tested for significance of the improvements, please specify the used test and provide the corresponding p-values. Otherwise, please rephrase to „noticeable improvement“
L199-200: “the AIC and BIC were the lowest” please add “and R² was highest”
L204-206 I think something went wrong with the labels and caption of Fig 2: Shouldn’t it be Model 8 and Model 8.7? In Fig2b it actually says Mode 11, and in the caption it says model 4.9 – this is confusing, please clarify. Moreover, what do you mean by “at the level of k-means 8”? At this point in the manuscript, k-means clustering has not been introduced, so please provide a residual diagram for M8 and M8.7 at this point, and present the residual diagram for M8.7 with 8 groups in section 3.4
L210 “M4.7” Shouldn’t it be “M8.7”? Please clarify.
L216-217 The R2 increase from 0.5773 for M8 to 0.8683 for M8.7 was already shown in Tab 5. Please avoid repetition. Regarding the usage of the word “significantly”, please see my comment on L198
Author Response
Please see the attachment
Author Response File: Author Response.doc
Reviewer 2 Report
Dear Authors,
Please find my comments in the attached file
Regards
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.doc
Round 2
Reviewer 2 Report
Dear authors,
Thank you for taking my comments into account.
Yet, i strongly suggest to add Figure 3 presented in the coverletter to the manuscript. It shows nice inproval in your results.
Regards
