Soil and Land Cover Interrelationships: An Analysis Based on the Jenny’s Equation

Round 1

Reviewer 1 Report

Manuscript number: 2018051

Manuscript Title: Soil diagnostic horizons and land use relationships: an analysis in the context of Jenny’s equation

Article Type: Research article

The study by Rodríguez-Rastrero & Ortega-Martos investigated the analysis of the relationships between "soil" and "organisms" as components of a fundamental expression in soil science, the Jenny equation. After surveying a total of 424 soil profiles in northern Spain, authors, through different statistical dependency relationships between both variables, confirmed that the method proposed in their study contributed in an affordable way to the development of quantitative models of the spatial distribution of soils. However, the manuscript requires significant improvement as indicated below.

Research Title

Lines 2-3: The research title requires improvement as the title needs to reflect the objectives' clarity. Remove the full-stop at the end of the title

Abstract

Lines 11-22: Require improvement to state briefly the existing problem, objective of the study, methodologies, results and conclusion. 

  1. Introduction

Lines 27-159: The introduction section needs to be shorter and begins with soil forming factors instead of focusing on the description of soil horizon diagnostics and land use relationship. I recommend that the authors revise this section and include information starting from line 130 and borrowing some points very briefly between lines 27-129. 

 2. Materials and Methods

Lines 161-163: Require a supporting reference. 

Lines 167-171: Any evidence of these conditions (acidity, rockiness, salinity, granulometry) associated with lith…and Geom… factors, This para fit to be in the results section and also as a supporting statement to the discussion 

Line 172: Start this para with a statement, "A survey was conducted during the year…..covering a total of 424 ….. points" 

Lines 174-178: What was the size of the soil profile length, width and depth?

Line 187: will it be used, or was it used? 

NB: I did not read in the materials and methods on methodologies, or the instruments used for opening up a soil profile, a tape for measuring the depth of different horizons, usage of soil colour kit etc that concluded ochric, umbric and mollic epipedons. How was soil pH determined? Furthermore, there should be a subsection under materials and methods showing the statistical software used for analyzing the data. How were the data analyzed for this study? 

3. Results

The results could be presented better! Authors should provide the results first, supported by their Table results.

Line 211: What are these two variables regarding Tables 1 and 2 at P<0.05? 

Lines 221-232: These results are derived from which Table? Please indicate a relevant Table in line 224 after the word 'higher.' 

Lines 233-234: Please provide the results first and give a reference Table 4 later. The authors have not presented the results.

Lines 238-241: Where are the results for Table 5? It would help if you consistently numbered tables based on the results logically. Discussion for Table 3 and Table 5 need to be in one section, and these tables should appear below the results

Lines 251-260: Which Table is under these results? 

  4. Discussion

Lines 263-265: Is this a discussion or description of statistical analysis? 

Lines 263-284: The Authors should discuss the results logically with reasons supporting the changes in the Tables, then finally provide references to compare or contradict results with other authors by quoting a references

Lines 287-335: Where are these discussions related to which Tables? These are not discussions but rather a part of the literature review. 

Lines 336-349; 352-376: Please discuss the results concerning relevant Tables, followed by the supporting results from other authors. 

5. Conclusion

Line 375: The authors start mentioning Jenny's equation developed, of which I have yet to see either materials and methods, results or discussion. The authors should summarizes the essential findings without repeating the results and discussion.

Author Response

Manuscript number: 2018051

Manuscript Title: Soil diagnostic horizons and land use relationships: an analysis in the context of Jenny’s equation

Article Type: Research article

The study by Rodríguez-Rastrero & Ortega-Martos investigated the analysis of the relationships between "soil" and "organisms" as components of a fundamental expression in soil science, the Jenny equation. After surveying a total of 424 soil profiles in northern Spain, authors, through different statistical dependency relationships between both variables, confirmed that the method proposed in their study contributed in an affordable way to the development of quantitative models of the spatial distribution of soils. However, the manuscript requires significant improvement as indicated below.

We sincerely appreciate your review. First of all, we would like to clarify that the manuscript has undergone a profound modification, in all its sections, derived from both your review and the one carried out by another reviewer. Added to that, are numerous corrections from our own review. A new author has participated in such improvements. We sincerely hope that all this has allowed a significant improvement of the manuscript in all its aspects.

Despite the great extension and importance of the corrections, we have marked in red type the texts that have been rewritten or are new for a better follow-up of the content of the previous version. I apologize for the delay in the review and the inconvenience caused.

All references to lines in this document refer, unless expressly indicated, to the new numbering of the lines of the manuscript.

Research Title

Lines 2-3: The research title requires improvement as the title needs to reflect the objectives' clarity. Remove the full-stop at the end of the title

Thanks for your comment. We propose a new title for the manuscript, perhaps more specific with respect to what, in our opinion, this work contributes:

“Soil and land cover relationships in the context of Jenny’s equation: an approach based on statistical dependency between categorical variables”

Abstract

Lines 11-22: Require improvement to state briefly the existing problem, objective of the study, methodologies, results and conclusion.

Thanks for your comment. Indeed, the previous abstract was incomplete in these respects. It has been modified and perhaps can now be more adjusted to what you are suggesting:

Regarding the existing problem and objective of the study:

“This research develops an approach to the analysis of the relationships between "soil" and "organisms", as components of the Jenny equation, a fundamental expression in soil science, which is the theoretical basis to modeling the complex occurrence of soils over landscapes.”

Regarding methodologies and corresponding results:

“This approach is based on the interpretation of the indeterminate function "f" of the equation, as "statistical dependence between categorical variables”. The categories of the “soil” component of the equation have been defined as “diagnostic horizons”, and those of the “organisms” factor, as synthetic types of “land cover”. After the application of this methodology on 424 soil profiles surveyed in an oceanic climate region of northern Spain, a multiple correspondence analysis showed well differentiated and pedologically consistent groupings between land cover categories and diagnostic horizons, as well as climate, relief and parent material forming factors. Likewise, bivariate analysis demonstrated coherent dependency relationships between Ochric, Mollic, Umbric, Cambic, Argillic and Spodic diagnostic horizons, and land cover categories as crops, prairies, shrublands, or pine plantations.”

Regarding Conclusions:

The results of this work provide objective and easy-to-apply criteria on assessing soils and land cover relationships, which can contribute to addressing the challenges involved in the development of models for the spatial representation of soils.

On the other hand, the keywords have also been substantially modified to better reflect aspects not expressly included in the title of the manuscript and that may facilitate future accessibility of the paper.

1. Introduction

Lines 27-159: The introduction section needs to be shorter and begins with soil forming factors instead of focusing on the description of soil horizon diagnostics and land use relationship. I recommend that the authors revise this section and include information starting from line 130 and borrowing some points very briefly between lines 27-129.

I sincerely appreciate your comment, which has undoubtedly contributed to a substantial improvement in this section. The introduction has been redrafted, and noticeably reduced in length (lines 28 to 101). Numerous paragraphs that were probably not necessary with respect to the objectives of this study, have been deleted and the contents and objectives have been redrafted, I hope in a clarifying way. Numerous references have been changed.

What you indicated regarding starting this section with the diagnostic horizons was initially developed by us. However, after a detailed reading of the resulting text, in our opinion, an introduction based first on the development of Jenny's equation (now reduced and perhaps better focused) was better suited to the objectives of this work. And then, develop everything related to the "soil" as a product (that is, the diagnostic horizons) and the factors that condition it. In any case, the modification of the introduction has also taken into account the comments made by another reviewer. I hope the result could be appropriate.

2. Materials and Methods

Lines 161-163: Require a supporting reference.

The Materials and Methods Section has been greatly modified. The information that you mention has been referenced in Line 108.

Lines 167-171: Any evidence of these conditions (acidity, rockiness, salinity, granulometry) associated with lith…and Geom… factors, This para fit to be in the results section and also as a supporting statement to the discussion

The soil properties and, of course, the lithological and geomorphological conditions have served as the basis for the Discussion throughout the entire section, supported in some cases by new references.

Line 172: Start this para with a statement, "A survey was conducted during the year…..covering a total of 424 ….. points"

Thanks for your comment. The paragraph now begins as follows:

“A soil survey was conducted in various field campaigns carried out in the years 2007, 2010, 2011 and 2012 with the participation of the corresponding author of this research. Various additional field tasks were carried out occasionally until 2017, covering a total of 424 sampling points [21], as shown in Figure 1.”

Lines 174-178: What was the size of the soil profile length, width and depth?

The minimum width of the soil profile was around 80 cm, enough to allow access to the surveyor, and its maximum depth was around 1.80 m, although logically this depended on the depth at which coherent rock or water table appeared (in fact, in most cases, the parent material appears at a clearly shallower depth). The length of the soil pit was therefore variable, but it could be assigned an average of about two meters (in general, a minimum occupation an impact of the plot on which the soil pit was carried out was intended). In general, the soil pits were made by machinery, unless the conditions of the place (as frequently happened in such a steep territory) did not allow their access, for which manual pits and road cuts were used, as long as they guaranteed access to the parent material (C or R depending on the coherence of the lithological substrate).

Line 187: will it be used, or was it used?

Sorry, indeed in a few cases the future was used by mistake; this has been corrected and the past simple has generally been used.

NB: I did not read in the materials and methods on methodologies, or the instruments used for opening up a soil profile, a tape for measuring the depth of different horizons, usage of soil colour kit etc that concluded ochric, umbric and mollic epipedons. How was soil pH determined?

In relation to the tools used, these included hoes and spades of various sizes, along with small tools for cleaning the profile and, indeed, a tape to measure the depth of the horizons. On each profile, a sampling of the different horizons was carried out; soil color measurement was carried out using Munsell Soil Color Charts (frequently in a wet condition, later completed dry in the laboratory), photographs of the profiles were taken, etc.; that is, the set of tasks typical of the soil survey.

Regarding the method for soil survey, a new text was added (lines 128-131):

“Whenever possible, soil profiles were prepared on soil pits made with machinery, unless the conditions of the place (as frequently happened in such steep territory) did not allow access. In that case, sufficiently deep cuts in the ground along roads, paths and other excavations were also used (Buol, 2011).”

Schoeneberger et al. (2002) was indicated as a reference for the soil survey carried out (Line 123).

Although the measurement of pH is not a requirement in most taxonomic criteria (except pH 1:1 for foreseeable spodic materials), it was carried out in a large number of samples due to its simplicity and informative capacity, by direct measurement with a lab pH meter on a suspension soil:water 1:2.5.

Furthermore, there should be a subsection under materials and methods showing the statistical software used for analyzing the data. How were the data analyzed for this study?

Indeed, the previous section of Material and Methods was markedly insufficient in the terms that you propose. This section has been modified and increased in its entirety, both in relation to references for the description of the soil profiles and for the taxonomical classification of the profiles, as well as for the definition of the categorical variables, and in general the statistical methods, paying particular attention to performing a multivariate analysis (multiple component analysis), as required by another reviewer. We believe that the inclusion of an exploratory multivariate analysis has contributed to improving the statistical treatment of information and the interpretation of land-cover / soils relationships.

Section 2.3 “Statistical analyses: definition of the components of the clorpt equation” (lines 149-211) and Section 2.4. “Statistical analyses: methods” (lines 212-222), were added to this manuscript.

Additionally, the review of the document has allowed us to see that the term "land use" could probably lead to confusion and it has been preferred to use the term "land cover", perhaps more appropriate to describe the land units subjected to different agriculture and livestock uses, or natural or semi-natural vegetation, in such a way that is more consistent with the objective of our work. Therefore, it has been modified in the manuscript.

3. Results

The results could be presented better! Authors should provide the results first, supported by their Table results.

The Results section has also been completely modified, presenting first the variables and categories for soil and soil forming factors (Table 2) and then the results of multivariate analysis (Figure 2 and lines 236 to 274), and finally the results of the analysis bivariate (Chi-square test; Tables 3 and 4), all in accordance with the objectives established in the introduction.

Line 211: What are these two variables regarding Tables 1 and 2 at P<0.05?

Lines 221-232: These results are derived from which Table? Please indicate a relevant Table in line 224 after the word 'higher.'

Lines 233-234: Please provide the results first and give a reference Table 4 later. The authors have not presented the results.

In order to improve the Results section, significant modifications have been made:

Former Line 211; and lines 233-234:

Table 2 referred to the variables "surface diagnostic horizons" and "land cover". Table 4 referred to the variables "subsurface diagnostic horizons" and "land cover". In both cases it was an attempt to indicate that there was a dependency relationship (p < 0.05). However, tables 2 and 4 have been deleted and the content has simply been indicated in the text.

Former Lines 221-232:

These results were derived from former Table 3. These comments have been considered more appropriate in the Discussion section, so they have been removed from Results section and integrated throughout that section.

Lines 238-241: Where are the results for Table 5? It would help if you consistently numbered tables based on the results logically. Discussion for Table 3 and Table 5 need to be in one section, and these tables should appear below the results.

Lines 251-260: Which Table is under these results?

Results for Table 5 were described in former lines 243-260, but indeed the section had deficiencies in the information. In this sense, it has been considered to include in the Results section only the relation of the data obtained as tables and to transfer the comments to the Discussion section for a better understanding of the results.

4. Discussion

The Discussion section (lines 292-451) has been largely rewritten and reorganized in its entirety, since indeed, it presented notable deficiencies when discussing and evaluating the results obtained.

Lines 263-265: Is this a discussion or description of statistical analysis?

Lines 263-284: The Authors should discuss the results logically with reasons supporting the changes in the Tables, then finally provide references to compare or contradict results with other authors by quoting a references

Lines 287-335: Where are these discussions related to which Tables? These are not discussions but rather a part of the literature review.

Lines 336-349; 352-376: Please discuss the results concerning relevant Tables, followed by the supporting results from other authors.

Possibly, the review carried out answers all the questions raised, for which we are sincerely grateful since they have served as a reference for this improved new version. Various aspects have been complemented with new references and tables and new Figure 2 have been conveniently referenced in the text.

5. Conclusion

Line 375: The authors start mentioning Jenny's equation developed, of which I have yet to see either materials and methods, results or discussion. The authors should summarizes the essential findings without repeating the results and discussion.

The Conclusions section has been completely rewritten and we appreciate this comment for its usefulness. We think that with this new version, the basic findings are highlighted and redundancies or undefinitions contained in the previous version of this section are purged.

Thank you very much.

Original Submission Date: 24 October 2022

Date of this review: 30 Jan 2023

Author Response File: Author Response.docx

Reviewer 2 Report

Review of the manuscript “Soil diagnostic horizons and land use relationships: an analysis in the context of Jenny’s equation”

The manuscript analyses the relations between present land use types and diagnostic soil horizons of Soil Taxonomy by means of contingency tables and chi-squared tests.

1.- The manuscript requires a complete revision based on a multivariate analysis of the data that the authors seem to have collected rather than on the bivariate analysis that they present in this manuscript.

While the authors discuss Jenny’s equation at length in the Introduction (may be a bit too much) they actually fail to apply it. The authors state in lines 38-41 that the development of a statistical equation with “soil” as the dependent variable and the factors of soil formation as explanatory variables requires the latter to be “independent variables, which is not generally acceptable”. But the independent character, or otherwise, of the explanatory variables requires a statistical analysis for each case study.

The authors are overrating the explanatory power of land use by ignoring the rest of the factors of soil formation. Jenny’s equation is clear about the joint effect of these factors and therefore the authors cannot pretend that soil (or diagnostic horizon) variability may be analysed through one of the factors (land use in this case). The authors implicitly recognise as much during their Discussion when they frequently state how the presence of the various land use types is influenced by soil parent material (lines 274-275), climate and geomorphology (lines 283-284; lines 345-346), climate (line 347), geomorphology and parent material (lines 353-354), and in their final statement in lines 371-372. While land use is clearly influenced by all those other factors, it does not “bring together the influence of all the soil-forming factors” (lines 375-376), particularly when some of these land use types may have only been performed for relatively short periods of time. The present land use type on a certain field may be relatively ‘young’ as some land use types have expanded and others have shrunk, not only in recent decades but throughout History (i.e., the time factor of soil formation).

Furthermore, the authors seem to have collected the basic information to introduce the other factors of soil formation in their analysis, and therefore they should change to a multivariate analysis of their data, including information about the climatic conditions, geomorphology, and parent material of their sampling plots. Various statistical procedures may be used to determine whether land use is really the main variable explaining diagnostic horizon variability or whether it is some other/s of the factors.

2.- The Materials and Methods section is quite poor:

- the climatic data should include at least some data on temperature, and about the

  spatial variability of temperature and rainfall,

- lines 167-171 are too general and do not provide any specific information about the

  study area,

- information about parent materials and geomorphology is missing,

- I think that data about diagnostic horizons (lines 187-198) may be better placed in the

  Results section,

- I do not agree with the inclusion of orchards, which have a specific class in the

  CORINE classification, in the arable class,

- there is no information at all about the statistical procedures used.

3.- The manuscript requires a very thorough revision in terms of written English

4.- Other issues:

- The authors seem to convey the idea that diagnostic horizons can be a surrogate for soils (e.g. lines 130-134, 393-394). But the parts (or even the sum of the parts) do not make the whole. While it is a fair objective to analyse the relation between vegetation and soil diagnostic horizons, this is not the same as analysing the relations between vegetation and soils.

- I do not agree with the statement in lines 52-53 that pH is a “direct consequence of the biotic factors”. Climate (particularly in an area with annual rainfall varying between 900 and 1800 mm) and parent material are also very significant in this respect.

- I do not agree with the statement in line 77 that “soil evolution itself favours the development of plant cover”. The formation of Albic, Argillic, or Petrocalcic horizons, for example, may involve a severe limitation for root development and plant growth.

- I do not agree with the statement in lines 100-102 that the differences between Soil Taxonomy and WRBSR “make it difficult to interpret the available soil information…and limit the application of soil knowledge”. Any soil map provides much more information about the various soil types than just their names in any of the taxonomic systems

- I do not agree with the sentence in lines 123-125 that vegetation “is a main criterion to define and delimitate of soil cartographic units”. But except in areas where vegetation may be considered to be ‘natural’, disturbances, particularly those produced by humans, have change the vegetation patterns so profoundly that these may not be of much use for soil mapping. In fact, the authors state in lines 128-129 that “relationships between vegetation-land uses and diagnostic horizons have not been statistically established”.

The authors support their statement with a reference to the book by Porta et al. (2005, p.541). But this page does not exist in this book, and the chapter in the book that refers to soil maps does not support the statement by the authors at all.

5.- Minor issues:

- The authors should stick to one single way of writing the names of diagnostic horizons. For example, they use, for example, “argillic”, argillic, Argillic, and argillic. I suggest they always use Argillic because it is a proper noun.

- The sentence in lines 139-140 is incomplete

Author Response

Manuscript number: 2018051

Manuscript Title: Soil diagnostic horizons and land use relationships: an analysis in the context of Jenny’s equation

Article Type: Research article

Despite the great extension and importance of the corrections, we have marked in red type the texts that have been rewritten or which are new, for a better follow-up of the content of the previous version. I apologize for the delay in the review and the inconvenience caused.

The manuscript analyses the relations between present land use types and diagnostic soil horizons of Soil Taxonomy by means of contingency tables and chi-squared tests.

1.- The manuscript requires a complete revision based on a multivariate analysis of the data that the authors seem to have collected rather than on the bivariate analysis that they present in this manuscript.

Indeed, the manuscript has undergone a necessary rewriting in all sections. The new text has been conditioned by your own review as well as and the one carried out by another reviewer. Additionally, numerous corrections from our own review have been considered necessary. A new author has participated in such improvements, as indicated above.

A new title for the manuscript has been proposed according to the comments of another reviewer. In our opinion, it is perhaps more specific with respect to what this work contributes:

“Soil and land cover relationships in the context of Jenny’s equation: an approach based on statistical dependency between categorical variables”

We sincerely hope that all this has allowed a significant improvement of the manuscript in all its aspects.

All references to lines corrected in this document refer, unless expressly indicated, to the new numbering of the lines of the manuscript.

While the authors discuss Jenny’s equation at length in the Introduction (may be a bit too much) they actually fail to apply it.

Thank you very much for your observation. In fact, another reviewer has requested to significantly reduce the Introduction and restructure it. The introduction section has been significantly reduced in length (lines 28 to 101). The paragraphs that were considered not necessary or undefined with respect to the objectives of this study, have been deleted, and all contents and objectives have been redrafted. We hope in a clarifying way. Numerous references have been changed.

The authors state in lines 38-41 that the development of a statistical equation with “soil” as the dependent variable and the factors of soil formation as explanatory variables requires the latter to be “independent variables, which is not generally acceptable”. But the independent character, or otherwise, of the explanatory variables requires a statistical analysis for each case study.

The authors are overrating the explanatory power of land use by ignoring the rest of the factors of soil formation. Jenny’s equation is clear about the joint effect of these factors and therefore the authors cannot pretend that soil (or diagnostic horizon) variability may be analysed through one of the factors (land use in this case).

Thank you very much for these comments; actually, the multivariate analysis was necessary, and we have carried it out on the variables that make up Jenny's equation, with the exception, justified in the text (Lines 160-165) of the time factor. The application of this analysis has significantly improved the study and has implied the modification of the manuscript as a whole. The results of the multivariate analysis are summarized in the new Figure 2.

Indeed, the soil survey carried out provided information on the formation factors associated with each of the 424 soil profiles. Such information was compartmentalized into qualitative variables and their corresponding categories, in a way specifically developed in the doctoral thesis of Rodríguez-Rastrero (2016), from which this part of the methodology derives.

The authors implicitly recognize as much during their Discussion when they frequently state how the presence of the various land use types is influenced by soil parent material (lines 274-275), climate and geomorphology (lines 283-284; lines 345-346), climate (line 347), geomorphology and parent material (lines 353-354), and in their final statement in lines 371-372.

Indeed, the results of the multivariate analysis have been incorporated into the Discussion section and we believe that such incorporation is largely consistent with the results of the previously performed bivariate analysis, which allows us to elaborate on some aspects of the role of vegetation. Probably, the discussion has been improved with comments supported more consistently by the rest of the forming factors.

While land use is clearly influenced by all those other factors, it does not “bring together the influence of all the soil-forming factors” (lines 375-376), particularly when some of these land use types may have only been performed for relatively short periods of time. The present land use type on a certain field may be relatively ‘young’ as some land use types have expanded and others have shrunk, not only in recent decades but throughout History (i.e., the time factor of soil formation).

In the rewrite of the Discussion section, the sentence in lines 375-376 was indeed inappropriate in the terms so expressed and the sentence has been removed. Thank you very much for the comment.

Furthermore, the authors seem to have collected the basic information to introduce the other factors of soil formation in their analysis, and therefore they should change to a multivariate analysis of their data, including information about the climatic conditions, geomorphology, and parent material of their sampling plots. Various statistical procedures may be used to determine whether land use is really the main variable explaining diagnostic horizon variability or whether it is some other/s of the factors.

Indeed, the information collected in each of the 424 soil profiles and their immediate environment has been the basis of the statistical treatment. The multiple component analysis has made it possible to verify the proximity between most of the types of land cover.

In this sense, the review of the document has allowed us to see that the term "land use" could probably lead to confusion and it has been preferred to use the term "land cover", perhaps more appropriate to describe the units of land subject to different agricultural uses. and ranchers, or natural or semi-natural vegetation, in a way more in line with the objective of our work. Therefore, it has been modified in the manuscript.

2.- The Materials and Methods section is quite poor:

The Materials and Methods Section has been greatly modified and re-structured, both in relation to numerous references and for the taxonomical classification of the profiles, as well as for the definition of the categorical variables, and in general the statistical methods, paying particular attention to performing the multivariate analysis by means of a Multiple Component Analysis. New sub-sections were added to this manuscript and included:

2.1. Study area: summary of soil forming factors (lines 104-116),

2.2. Sampling points: soil profiles and on-ground information on soil forming factors (lines 118-148),

2.3 “Statistical analyses: definition of the components of the clorpt equation” (lines 149-211),

2.4. “Statistical analyses: methods” (lines 213-222).

- the climatic data should include at least some data on temperature, and about the spatial variability of temperature and rainfall,

- lines 167-171 are too general and do not provide any specific information about the study area,

- information about parent materials and geomorphology is missing,

The information contained in subsection 2.1 has been expanded; in any case, a detailed description of the great physiographical complexity of the study area may be outside the objectives of the work; however, the manuscript has now obvious improvements in this regard: an assessment of the variables "climate", "relief" and "parent material" has been made between lines 174-211. Relevant information regarding these factors have been included in the Discussion section, supported, when necessary, by new references.

- I think that data about diagnostic horizons (lines 187-198) may be better placed in the   Results section,

Thanks for the comment. In this work, we have considered the information from the 424 soil profiles as a "material" rather than a "result", since this information comes from the doctoral thesis of the corresponding author. In this sense, we have thought of keeping it in the Material and Methods section, where, in our opinion, it fits properly, particularly after the improvement made to it.

- I do not agree with the inclusion of orchards, which have a specific class in the CORINE classification, in the arable class,

Thanks for the observation. The inclusion of the term "orchards" in the arable class soils really raised doubts: on the one hand, apple orchards are a crop of high economic interest in certain areas of the Asturias region, where their expansion has occurred at the expense of surfaces previously dedicated to meadows. However, these areas were not included in our study, that is, in these new orchard areas we do not have soil pits. On the other hand, from our field experience in the area where the soil survey has been carried out, the surfaces that can be included in the definition of orchards "plantation of trees or shrubs that is maintained for food production" correspond fundamentally to kiwifruit orchards, in intimate mix with fodder crops, or vegetable plots, on which landuses are very recently expanding, depending on the farmer's particular criteria. This circumstance and, above all, the lack of direct information from the soil pits, did not allow us to include "orchards" as a specific landuse variable, being included, for the sole purpose of our study, as one more component of the description of the types of use that are part of the arable lands". In any case, an indication regarding this was included in Table 1.

- there is no information at all about the statistical procedures used.

Subsections 2.3 and 2.4 (lines 149-222) describe the statistical procedures developed in this work.

3.- The manuscript requires a very thorough revision in terms of written English

Thank you. We hope that the revision carried out may have sufficiently improved the level of written English.

4.- Other issues:

- The authors seem to convey the idea that diagnostic horizons can be a surrogate for soils (e.g. lines 130-134, 393-394). But the parts (or even the sum of the parts) do not make the whole. While it is a fair objective to analyse the relation between vegetation and soil diagnostic horizons, this is not the same as analysing the relations between vegetation and soils.

Thank you for the comment. I think that raises an interesting question. Indeed, the text could lead one to think that our work proposes a kind of replacement of the concept of soil by the "diagnostic horizons". For this reason, we have rewritten both paragraphs or phrases, since it should clarify better that what is being proposed is effectively analyzing relationships between land cover and diagnostic horizons, but not so much as an alternative to soil relationships. taxon-land cover, but rather as an essay to address the application of the Jenny equation by greatly minimizing the number of categories. We consider this to be of great interest for assessing soil-land cover relationships. In this sense, we can point out that in the original study approximately 60 soil taxa were determined at the subgroup level, compared to only 3 superficial and 5 subsurface horizons. This approximation therefore allows an enormous simplification of the statistical analysis, without leaving aside the fact that the diagnostic horizons are also a product of the interaction of the soil-forming factors and can be mapped, as Bockheim points out in the corresponding reference [20].

- I do not agree with the statement in lines 52-53 that pH is a “direct consequence of the biotic factors”. Climate (particularly in an area with annual rainfall varying between 900 and 1800 mm) and parent material are also very significant in this respect.

Thank you for this observation. The idea was not correctly expressed and has been rewritten in lines 42-48.

- I do not agree with the statement in line 77 that “soil evolution itself favours the development of plant cover”. The formation of Albic, Argillic, or Petrocalcic horizons, for example, may involve a severe limitation for root development and plant growth.

Thank you for the correction; totally agree. This idea has been put into context and is described in lines 39-46.

- I do not agree with the statement in lines 100-102 that the differences between Soil Taxonomy and WRBSR “make it difficult to interpret the available soil information…and limit the application of soil knowledge”. Any soil map provides much more information about the various soil types than just their names in any of the taxonomic systems

Yes, it really is poorly phrased and has been rewritten and explained in lines 75-83. What was intended to be said is that the differences in criteria between the maps that use Soil Taxonomy and those that use WRB make it difficult to jointly use the information contained in the different national or even regional soil maps. Soil information in Spain is a notable example of this problem. And it was intended to affirm that, on the other hand, diagnostic horizons are in themselves a particularly valuable tool since, unlike taxa, they are defined in a similar way in both classification systems and are undoubtedly key to any attempt at a universally accepted classification system.

- I do not agree with the sentence in lines 123-125 that vegetation “is a main criterion to define and delimitate of soil cartographic units”. But except in areas where vegetation may be considered to be ‘natural’, disturbances, particularly those produced by humans, have change the vegetation patterns so profoundly that these may not be of much use for soil mapping. In fact, the authors state in lines 128-129 that “relationships between vegetation-land uses and diagnostic horizons have not been statistically established”.

Thank you very much for the annotation. It is probably poorly expressed in the text and can lead to confusion. I think it's an interesting topic; Actually, our objective was to highlight that land cover is one of the tools that allows the delineation of polygons of soil cartographic units, and based on remote sensing techniques. I think that this could even be independent of the degree of conservation of the original soil; For example, if the problem arises of delimiting certain soils associated with alluvial deposits, from others associated with colluvial deposits, irrigated agricultural use, for example, is usually a good indicator. This indicator character is helped precisely by the fact that the changes in the types of agricultural use are "spatially sharp" which facilitates the delineation. In the case of different formations of natural vegetation, indeed the profound alteration of such formations makes it difficult to use this criterion as a delimiter of soil types. On the other hand, our affirmation of former lines 128-129, was rather intended to justify the convenience or usefulness of this research.

The authors support their statement with a reference to the book by Porta et al. (2005, p.541). But this page does not exist in this book, and the chapter in the book that refers to soil maps does not support the statement by the authors at all.

I am sorry about the mistake. It is a reference included in the doctoral thesis of Rodríguez-Rastrero (2016) that has been wrongly added here. The number p. 541 corresponds to the number of pages in the book, not the page on which the reference is found. The correct reference actually corresponds to D.G. Rossiter, in his works regarding Soil Mapping, of which we have chosen the following:

[17] Rossiter, D.G. Soil mapping today: computer-generated predictive soil maps – their role in soil survey and land evaluation. Agriculture for Development, (2021), 44.

Specifically, in a brief Introduction, the author mentions, among the fundamental factors for the current development of predictive soil mapping, the “free availability of a vast store of geographic gridded coverages related to the factors of soil formation”, including “organisms”, as “vegetative cover from satellite imagery”.

5.- Minor issues:

- The authors should stick to one single way of writing the names of diagnostic horizons. For example, they use, for example, “argillic”, argillic, Argillic, and argillic. I suggest they always use Argillic because it is a proper noun.

I agree. All diagnostic horizon names in the new text have been corrected as indicated.

- The sentence in lines 139-140 is incomplete

Yes, indeed it was incomplete, and the idea has been redrafted in the lines 85-89. He was trying to highlight the usefulness of using diagnostic horizons as tools in the conceptual application of the Jenny equation.

Thank you very much.

Original Submission Date: 24 October 2022

Date of this correction: 30-Jan 2023

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Research Title

Lines 1: I propose the title read “Soil and Land Cover Relationships under different pedological factors” 

Abstract: Okay 

Introduction: Okay 

Materials and Methods: Okay

Results

Lines 233-234: Describe Figure 2 to show an explanation for part a) three groups or clusters and b) the association of soil and soil-forming factors 

Discussion

Lines 297-299: Any supporting Table or reference to justify the results? 

Conclusion

Lines 455-459: Is the developed new and straightforward equation also shown in the discussion section? What explicitly shows the difference with the former one? 

Author Response

First of all, I really appreciate your positive evaluation of our manuscript. However, we must tell you that Reviewer 2 requested a "major revision" of the manuscript, which included a profound modification, especially of the Results, Discussion and Conclusions sections. We sincerely believe that the required modification has resulted in a better and broader explanation of the findings of the work. Therefore, the changes in the aforementioned sections, and in particular Discussion and Conclusions, have been profound, even in the Abstract. I'm sorry for the inconvenience this may cause you.

We respond below to the questions you have raised:

Research Title

Lines 1: I propose the title read “Soil and Land Cover Relationships under different pedological factors”

Thank you for your proposed title for this manuscript. However, I must adjust it fundamentally due to the profound transformation of the document in this second phase of revision. We have come to the conclusion that a much more synthetic title, along the lines that you propose, and suitable for a better definition of objectives and results, could be the following:

“Soil and land cover interrelationships: an analysis based on the Jenny equation.”

Abstract: Okay

Introduction: Okay

Materials and Methods: Okay

Results

Lines 233-234: Describe Figure 2 to show an explanation for part a) three groups or clusters and b) the association of soil and soil-forming factors

Thank you. It was corrected:

“Figure 2. Cluster factor map for soil profiles (a), and categories of “soil” and “soil forming factors” in the plane of dimensions 1 and 2 (b). Illustrative variables are soil diagnostic horizons and land cover.”

Discussion

Lines 297-299: Any supporting Table or reference to justify the results?

“These are forest uses (wooded or shrubby), high altitudes, lithologies that favor the development of acid and stony soils, and landforms associated with erosion and deposition processes.”

Thank you. It was corrected. The new paragraph is in lines 325 to 329:

“The presence of Umbric epipedons and Spodic endopedons in Cluster 1, as well as the absence of any subsurface diagnostic horizon, are pedologically consistent with factors like forest uses (wooded or shrubby), high altitudes, lithologies that favor the development of acid and stony soils [36], and landforms associated with erosion and deposition processes, as steep slopes, or high and convex slopes [37].”

It includes two new cites:

[36] Carballas, T.; Rodríguez-Rastrero, M.; Artieda, O.; Gumuzzio, J.; Díaz-Raviña, M.; Martin, Á. Soils of the temperate humid zone. In: Gallardo, J. F. (eds.) The soils of Spain. 2016. Cham, Switzerland: Springer International Publishing. https://doi.org/10.1007/978-3-319-20541-0_1

[37] Sun, L.; Guo, H.; Liu, B.; Wu; S.; Weckler, P.R., Yang, J. Characterizing erosion processes on a convex slope based on 3D reconstruction method. Geoderma, 2021, 402, November, 115364. https://doi.org/10.1016/j.geoderma.2021.115364.

Conclusion

Lines 455-459: Is the developed new and straightforward equation also shown in the discussion section? What explicitly shows the difference with the former one?

Results, Discussion and Conclusions sections have been redone to a great extent due to the indications of the other reviewer. After intense work, the new section can clarify and improve the manuscript. I understand that the question that you raise here refers to what this work explicitly contributes in relation to the Jenny equation. Basically, the work provides a way of approaching a solution to such equation, rather than a new equation in the strict sense. We do consider this way to be new, and to the best of our knowledge, it has not been previously developed. We think that in this current version of the manuscript, both the objectives and the contributions of this work have been better defined and, perhaps now, they could better answer the question that you posed here.

Thank you very much for your revision work.

Best regards

Reviewer 2 Report

- I do not think that the change in the title of the manuscript is appropriate. The fact that a statistical analysis has been performed is not important. In fact, it may be argued that statistical analyses are almost a ‘must’ these days. And I certainly argue that, in this case, the fact that the authors have studied such a high number of sampling plots (424) is much more important than any statistical analysis or than the fact that variables are categorical.

- Precisely for this reason, I think it is a pity that the authors are not making the most out of such a wealth of information.

- In this new version the authors have simply added the basic graphs of an MCA to their previous version. But they have failed to extract all the information that a multivariate analysis can provide, and have therefore failed to study the significance of the relation between land use and diagnostic horizons considering the influence of the other factors of soil formation.

- The objective of the multivariate analysis should be to define which of the explanatory variables used (the factors of soil formation in this case), if any, is significant explaining the presence of, in this case, certain diagnostic horizons.

-  And ONLY if this is proved for land use may the authors proceed further to analyse the relation between certain diagnostic horizons and certain land uses. And this last analysis should be done on the basis of the groups of land uses produced by the previous multivariate analysis. Right now, the chi-square test is not correct

Author Response

Dear reviewer:

In accordance with your important suggestions, this new version of the manuscript has undergone a profound modification, especially of the Results, Discussion and Conclusions sections, and includes a new Abstract that, we understand, better responds to the objectives and obtained results. The writing of new texts has been very significant, which has delayed the delivery of this new version. I'm sorry for the inconvenience.

We respond below to the questions you have raised:

"- I do not think that the change in the title of the manuscript is appropriate. The fact that a statistical analysis has been performed is not important. In fact, it may be argued that statistical analyses are almost a ‘must’ these days. And I certainly argue that, in this case, the fact that the authors have studied such a high number of sampling plots (424) is much more important than any statistical analysis or than the fact that variables are categorical."

Thank you. The title has been simplified and mentions to methods of statistical analysis have been avoided. We trust that the new title adequately responds to the improvement in objectives and approach, that we have tried to carry out in this new version of the manuscript.

“Soil and land cover interrelationships: an analysis based on the Jenny equation.”

- "Precisely for this reason, I think it is a pity that the authors are not making the most out of such a wealth of information."

We really appreciate your suggestion; We believe that this new version better responds to the potential offered by the significant amount of information that resulted from our field work. The new version includes a profound modification of the text of the first reviewed version, with special emphasis on Results, Discussion and Conclusions sections. We sincerely hope that these important changes improve the explanation on the findings of this work.

- "In this new version the authors have simply added the basic graphs of an MCA to their previous version. But they have failed to extract all the information that a multivariate analysis can provide, and have therefore failed to study the significance of the relation between land use and diagnostic horizons considering the influence of the other factors of soil formation."

Lines 230 to 232 include the improvements applied in the sense of your comment, that is, in order to improve the extraction of information from the multivariate analysis, and thus, they serve as a basis for more clarified and adequate results. And, therefore, as a base for the Discussion.

- "The objective of the multivariate analysis should be to define which of the explanatory variables used (the factors of soil formation in this case), if any, is significant explaining the presence of, in this case, certain diagnostic horizons."

A new figure (Figure 3), in lines 271-272, shows the results of the application of the selection criteria used (Cos2 values for all the categories of the variables used in the MCA).

-  "And ONLY if this is proved for land use may the authors proceed further to analyse the relation between certain diagnostic horizons and certain land uses. And this last analysis should be done on the basis of the groups of land uses produced by the previous multivariate analysis. Right now, the chi-square test is not correct".

Lines 277-288 of the new version, in the Results section, explain the selection made for diagnostic categories of horizon-land cover soils, whose interaction was relevant for the subsequent bivariate analysis.

These interrelationships are later explained and referenced in the Discussion section, which has therefore been profoundly modified.

Thank you very much for your review work.

Best regards.