Nitrogenous and Phosphorus Soil Contents in Tierra del Fuego Forests: Relationships with Soil Organic Carbon, Climate, Vegetation and Landscape Metrics

Round 1

Reviewer 1 Report

The idea of using vegetation type to better understand soil chemistry particularly in areas where soil sampling can be difficult due to terrain and accessibility can be a great tool for better understanding economic and cultural importance of areas in native vegetation. 

It is unfortunate that the on the ground sampling of two of the forest types wound up being quite small, but I think looking at things "globally" and "individually" is a reasonable way to address that issue.  

The language, particularly in the Results and Discussion section needs to be cleaned up to improve the clarity of those sections there are some places where the intent gets lost the way it is worded.  See the attached file for specifics.

Comments for author File: Comments.pdf

Author Response

Extensive editing of English language and style was made, especially in Results and Discussion. We hope these sections have been enough improved, as was requested. Thanks for the detailed suggestions in the text, all of them were added to the new version. Two specific comments: (a) Yes, we agree that the main weakness of the draft were the lower number of one of the forest types; however, we discussed and highlighted this detail in the text; (b) You suggest to be more striking the colour scale in the Figure of the maps; however, the scale is already extreme, although it is not noticeable because the pixels with extreme values are not very evident in the maps (due to very low pixel occurrence in the extreme values).

Reviewer 2 Report

The manuscript “Nitrogenous and Phosphorous Soil Contents in Tierra del Fuego Forests: Relationships with Soil Organic Carbon, Climate, Vegetation and Landscape Metrics”, is a well planned and executed work that model SN and SP contents on soils within 30 cm depth in Tierra del Fuego forests (Argentina) using two modelling based on climatic, topographic and vegetation variables.

 The topic is not original, because similar research has been carried out by other researchers, but under different conditions. Therefore, I believe that the results of the conducted research provide a lot of new data, significantly enriching the knowledge in this area. I also believe that the research has been conducted correctly and the results are well documented. Conclusions resulting from the conducted research are consistent with the presented evidence and arguments and relate to the main research objective. The discussion of the obtained results against the results of other researchers was also correct. Tables and figures are complete. I believe that the article can be published after taking into account the comments in the table below.

 In line 26, The models captured variability well for SP (R²-adj. >70%), please check the linear regression models suit for SP.

 

In line 152, In Figure 1. the sampled stands (red dots) and others maybe add legends in the figure.

 In 3. Results, Linear regression models of SN, SP, why not use Non-Linear Models???

  In line 351, Table 5, 141.3(22.3) , is means that 141.3±22.3, is it right? Please clarify. 

Author Response

Reviewer 2

Extensive editing of English language and style was made, as also was suggested by Reviewer 1. We improved the methods according to the comments and suggestions made by the reviewers. We hope, now, these are clearer than in the first draft. Regarding the specific suggestions:

(a) “In line 26, The models captured variability well for SP (R²-adj. >70%), please check the linear regression models suit for SP.” The numbers are OK, but we include the full range variability to clarify the sentence.

(b) “In line 152, In Figure 1. the sampled stands (red dots) and others maybe add legends in the figure”. We prefer to maintain the figure as simple as we can, and left the explanation of the dots and colours in the legend. We believe that it is clear in the present form (less text inside the figure is much better).

(c) “In 3. Results, Linear regression models of SN, SP, why not use Non-Linear Models???”. This is a good question. When we use linear regression models, all the variables have the same weight and chance to be selected, and it is possible to implement stepwise multiple regressions. In the case of non-linear regressions, it is necessary to start with some “model” or “premises”, and was not the idea for this study in particular. I use the non-linear regressions in many papers, but mainly in ecology and eco-physiology studies.

(d) “In line 351, Table 5, 141.3(22.3), is means that 141.3±22.3, is it right? Please clarify.” We clarify the Table as was suggested.

Reviewer 3 Report

Please address the queries given in the sticky notes embedded in the manuscript itself.

Comments for author File: Comments.pdf

Author Response

We improved the methods according to the comments and suggestions made by the reviewers. We hope, that this section is more clearly presented now. Regarding the specific comments in the text:

1) “The authors have provided excellent background of their work with logical gap in knowledge. I feel it would be better the authors also provide the specific research questions along with their hypothesis based on which this study was planned.”

We not stated a hypothesis for this study, mainly due to it is based on modelling of variables at landscape level. However, as was proposed, we defined two hypotheses to better support the specific research questions. The specific questions are listed between lines 132 to 137 as specific objectives. We can present them as questions, but we believe that they are clear enough, and were answered in results and discussed in the text. We can change the style, but it will not change the “content.”

2) “Will it not be better to give the baseline information of climatic and edaphic conditions of your study area? I feel that providing the information on silvicultural management adopted for the forests is imperative.”

We add some lines about the climate and soils for the study area as was suggested.

3) Don't you feel that giving the historical trend of climatic data of you study area (at least for past thirty years) graphically will provide better insight to you readers?

We did not test the climate change effects here specifically. For this, we do not agree to amplify this aspect in the database and text. Besides, there are many different conceptions about the climate change here at these high latitudes, and not all of them follow the same path.

4) What exactly is adequate sampling for your case? What was the principle of your sampling?

If I can choose, a systematic grid is the best one. However, it was not possible according to the accessibility of sites. Also, a balanced distribution among the studied factors is the best for any kind of scientific studies. However, if you have more samples in the most valuable forests (e.g. timber values), the models will be more accurate there. In our study, we have strengths (e.g. number of samples) and weaknesses (e.g. under- and over-sampling in different forest types). We just want to highlight these bad and good things to the potential readers and users of the models.

5) What was the basis of your soil sampling from a sampled plot?

We don´t understand the question. These samplings were conducted in the framework of bigger projects, where we also measured forest structure, biodiversity and animal uses. What we do in the soil sampling was described in lines 178-186. The idea was to characterize the first 30 cm layer of the forest floor. Nothofagus forests usually concentrated the roots in this first soil layer.

6) What method did you apply for estimating bulk density?

As was described in the text, we extracted soil samples of a known volume. The soil samples were air-dried and sieved to 2 mm to remove roots, stones and wood. The remnant soil weight was related to the sampling volume, and this is the bulk density. This was described in lines 178-186. We added a reference about the method.

7) Can you elaborate for better insight of your readers?

The necessary information was in the text. We tried to clarify the text and to improve it.

8) less robust

It was changed.

9) What about silvicultural managements adopted? They do influence soil NP of forests, isn't they?

As was described in the methods, for this study, we do not include recently managed stands (<50 years). For sure, the management, as a human related impact and use, influenced on soil properties, as was published in recent papers. We clarified in the text the conducted sampling.

10) Does this under-sampling have any effect on your study? If any please discuss and how you managed this effect?

Does means that some models can present better adjustment than others, or represent better the landscape. However, it is impossible to measure this effect to the present data. Most of the authors prefer to hide the weaknesses, we prefer to point these limitations to the readers and potential users. Some models can be more accurate or trusted (e.g. Nothofagus pumilio) than others (e.g. mixed forests); however, they are good enough compared to those models informed in the literature. Besides, if you don´t want the under-sampling by forests type, you can choose the global modelling (as was traditionally modelled) with a huge sampling effort (1 plot every 1000 hectares). We believe that our text is well presented in this topic.

11) Why do you say so? Will you please elaborate?

We clarify the sentence.

12) You mean to say that we need to employ a combination of the methods? As a case you have different type of forests in terms of species composition, what was your strategy?

Many of the cited studies compared different methods, and none is presented as the best for all the landscapes and research objectives. That means that all the methods presented advantages according to the study site, forest types, history of uses, etc., as was explained in the text. In our case, as was explained before, we followed a different objective than usual: we try to show if splitting the forest types can allowed to improve the modelling. For this, the stepwise linear regression was the chosen method.

13) I agree but how do you make it user friendly?

In GIS platforms or web platforms, e.g. we made friendly web platforms for public and national Agencies (see https://silvis.forest.wisc.edu/webmaps/forest_structure_maps_for_argentina/).

14) You are too general on your conclusion, please be specific based on your work. Also give some quantitative information. Please mention the limitation of your work also with any suggestion and recommendation.

Maybe this is just a style difference in the writing. We consider the contents of conclusions as adequate, where the main contributions of the paper were included. We believe that here data must not be included (this is not a result summary). If we include more specific or data results, the title must be changed for “final remarks” or something similar. We prefer the conclusion section as a new one with the summary of the main general new findings.

15) None

It was changed.