A Combined Ridging and Cover Crop Tillage System for Sustainable Coffee Plantation in Kenya
Round 1
Reviewer 1 Report
This is an interesting and highly applied study, comparing the impact of cover crops and ridge tillage, individually and collectively, on soil quality metrics. The experimental design appears to be sufficiently robust to allow conclusions. However, several major revisions need to be addressed, particularly in terms of the statistical treatment of data and the overall presentation.
1. The same experimental data is given in both figures and in Table 1. This is unacceptable. I find that Table 1 is sufficient for providing the data (after several modifications are made, per my comments below). Therefore, Figs 3.1-4 are not necessary and should be omitted.
2. A statistical framework (e.g., Tukey) should be applied to determine if apparent differences in soil properties between treatments is statistically significant. For example, even though the moisture content for T1-T3 appears significantly higher than TC (per Table 1), it is not clear if the difference is statistically significant.
3. Table 1.1 needs to be modified to include A) error bars for each entry, B) a footnote to explain the meaning of the error bars (e.g., standard deviation or standard error), and C) letters to indicate statistically significant differences.
4. Abstract. The numerical values given therein need to be accompanied by error bars. The conclusion given in the second-to-last sentence is weak for two reasons: A) "significant" is not supported by a statistical analysis and B) "significant effects on soil properties" is a rather vague phrase. The overall hypothesis (as given in the Introduction) should be provided in the Abstract and the Abstract should describe if the hypothesis was supported or not.
5. Please provide the rationale for selecting beans as the cover crop, rather than another cover crop.
6. For the 2.2 section heading, please use "pH", not"PH"
7. The nomenclature used for figures and tables is rather unorthodox (e.g., Table 1.1, Figure 3.1). For most journals, tables and figures are numbered sequentially in a single list (e.g., Table 1, Figure 3, etc.). Please check with the journal's Author Guidelines for the correct presentation. Also, when a specific table or figure is mentioned in the text, "Figure" and "Table" are capitalized (e.g., "Figure 3", not "figure 3").
8. In the experimental design, each treatment (T1-T4) had three replicates, correct? Please provide the specific number of replicates in the Experimental section narrative. Was the plot layout for the 4 x 3 experimental design randomized?
9. Table 1.1 How were these measurements obtained .. by the authors (in which case the methods should be provided) or another source (in which case the source should be cited)? Can error bars be provided?
10. Fig 2.2: scales need to be provided for both maps and should be legible. Text boxes appear to be small and are difficult to read.
11. L 182: "Water infiltration into was measured" is awkward. Please revise.
12. Given that this paper contains a small amount of data, the use of separate Results and Discussion sections is unwarranted. Please revise your paper to consist of a single, combined "Results and Discussions" section, to provide a more concise presentation.
13. I find that the two Fig 2.2's are not necessary, since the methods used for soil analysis are routine.
14. Note that there are two different Fig. 2.1's and 2.2's. Each figure should be uniquely numbered. Please see comment 7 above.
Author Response
Dear Reviewer
Please find herein my response to the comments. Corrections have been included in the paper. Thanks for your valued comments on our work which has made it even better. I hope it will be useful to the society.
Author Response File: 
Author Response.docx
Reviewer 2 Report
I have reviewed with interest your manuscript entitled „A combined ridging and cover crop Tillage system for sustain-2 able coffee plantation in Kenya” submitted to the future number of Agronomy.
This study assumes that the introduction of cover crop and ridging conservation tillage practices will increase the soil organic matter (SOM) within the coffee farms and also reduce compaction. This increase in SOM is assumed to have a direct influence on the porosity and thus on the groundwater holding capacity.
In my opinion the current version of your manuscript is suitable for publication in Agronomy, but after small revisions. The quality of the presentation should be improved. In general, manuscript is well written.
There are same grammar mistakes and awkward sentences, that have to be improved.
The article suffers from a number of small mistakes, ranging from misspellings to incorrectly phrased sentences.
Some adjustments are suggested to qualify the paper:
Issues include:
The Abstract should not exceed 200 words. Now there is 279 words. In my opinion it can be the shorten.
General comment to the Introduction section: The content of the literature review chapter is related to the research topic.
In the chapter "Materials and Methods", the methodology is adequate.
But I have same question:
1.Please provide other chemical parameters (N, P, K, Mg, C, humus,…), and references on the methods used to provide the soil properties.
2.What’s the sowing density of coffee
3 Why the Authors include only two years of the study. Three years of the study are more proper for statistical analysis.
4 What were the years of the study?
5 Provide the details of geographical localization of the study point, please.
6There is also lack of information of the size of experimental plots. Could You complete it?
In the chapter "Results", the results are displayed correctly.
The “Discussion” is informative. Moreover, the Authors attempt to discuss their important results and the rest is a quotation of literature. The Discussion should consider citing more latest and relevant papers to back up the arguments.
The Conclusions are correctly. Limitations of this study or suggestions for further study should be given here.
The list of References is enough, but a lot of Literature are not actual, before 2010 (42%). I suggest that Authors have to correct it and add same actual literature.
Same information are including in the text of Manuscript.
I hope that these comments help you to make an improved version of the manuscript.
Comments for author File: 
Comments.pdf
Author Response
Dear Reviewer
Please find herein my response to the comments. Corrections have been included in the paper. Thanks for your valued comments on our work which has made it even better. I hope it will be useful to the society.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper provides data on soil quality impacted by ridging and cover crops in Kenyan coffee plantation but has serious flaws. Authors claimed that erosion, low profitability and soil fertility must be curtailed by conservation practices but did not show whether the size of soil quality change enables any improvement in addressing such specific problems. On the other hand, the experimental design and statistical analysis are poorly described. The wording is often redundant and inconsistent. Apparently, the experiment was conducted during a single season, which is insufficient to reach solid conclusions.
Abstract
l. 17-19 : not clear what was the experimental design. Description provided l. 170-175 seems to differ. Ridges or terraces?
l. 19 : define CK before reporting it as acronym
l. 23-24: g cm-3. Correct across text.
l. 26: 0.25 mm in diameter
Introduction
Introduction is too long and repetitive. Please focus on the main idea then elaborate hypotheses.
l. 41: the production chain
l. 43: define GDP before using the acronym
l. 52-53: can you provide or approximate some numbers for extra fertilizer applications that contributed to the decline of planted area by 30% in Kenya? What would be the potential payback for smallholders to implement beneficial soil conservation practices such as ridging and cover crops?
l. 59: what would be the limiting soil fertility factors attributable to mismanagement?
l. 64-66: check for spacings
l. 85: are nutrient exports from farms compensated by extra fertilization? What nutrients appear problematic the most? If you use legumes as cover crops, I suppose that nitrogen is the main limiting factor?
l. 91-92: water pollution by phosphorus and GHG emissions as N2O, CO2?
l. 93: in inadequate amounts? Is the problem due to nutrient imbalance or extra fertilization?
l. 115-128: what would be the impact of those practices on soil fertility and the useless and costly extra fertilization?
l. 121: ‘decides’?
Figure 2.1: improve readability. There are two Figures 2.1 and two Figures 2.2 but no Figure 1. Renumerate Figures.
l. 145-150: too long sentence.
Table 1.1: g cm-3. Total porosity and moisture content as % in volume. What would be the variation across the experimental area for the values provided in the Table.
l. 142: pH and fertility
Material and methods
Provide methodologies and computations to quantify the variables in Table 1.1. There were several experimental fields, hence, means and standard deviations should be shown for each variable in Table 1.1.
l. 170-175: There were 12 fields and four treatments, hence 48 experimental plots. Treatments are terracing, cover crop, combination of terracing and cover crop, and three controls for side-by-side comparison. Terracing or ridges (l. 17-19)? For how many seasons did you conduct the experiment? In field experiments, you need at least three seasons to make results reliable due to variation in climatic conditions among seasons. Present meteorological data for every season. What is plot size for each treatment? How many replications? What were the grain yields of coffee trees to show the profitability of the proposed practices. You use ‘field’ and ‘plot’ interchangeably in the text (e.g., l. 221). How did you check for similarity of the baseline variables as required by quasi-experimental design? What are the possible sources of bias in site selection? Please clarify this.
Section 2.2: at what period did you sample the soil for physical analyses?
l. 184-185: provide ring sizes in cm.
l. 192-206: why did you dry the soil before conducting aggregate analysis? Rewetting may explode dry aggregates, leading to considerable bias in quantifying aggregation state.
l. 207: how did you measure sand content in aggregates? By sieving soil to less than 2 mm after crushing the aggregates?
Table 3.1: provide means and standard deviations as well as multiple comparison tests. Present soil fertility data, because soil fertility is of concern in the paper (see introduction).
l. 220, 229, 236, 243, 255, 262, 278, 281, 293…: ridging, ridge tillage or terracing? Please be consistent across text.
l. 230-231: will or would?
l. 246-48: moved to the M&M section. Define MS and indicate the version.
Table 3.2. On which variable was ANOVA conducted? What do you mean by rows and columns as sources of variation (what are the variables: sites, treatments?). There are 15 df as total, hence 16 experimental plots analyzed statistically. If there were 12 fields and four treatments, there were 48 experimental plots in total. Please clarify this.
l. 297: did you observe earthworm channels?
l. 305: were measurements taken during the drought period?
Author Response
Dear Reviewer
Please find herein my response to the comments. Corrections have been included in the paper. Thanks for your valued comments on our work which has made it even better. I hope it will be useful to the society.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors have addressed my concerns satisfactorily. Yet, a few additional minor revisions are recommended. Once these issues are addressed, I recommend acceptance of the paper.
1. There are 2 tables that are labeled Table 4 in the paper. Please review and fix the table numbering.
2. Figures 6 and 7 are essentially tables since they contain an organized array of numbers. They should be converted into tables. Please review the significant digits of the tabulated values. It appears that most entries contain too many significant digits.
3. Please address the following minor spelling or grammatical errors: A) L 202: "prormance"; 118-121: "beans .. it" (grammar); L 132: "dtermines"
Author Response
please find attached copy of my reply to review comments
Author Response File: Author Response.docx
Reviewer 3 Report
Authors answered to several questions but did a lousy job in answering some others. Hypotheses should be listed as 1, 2, 3… Objectives should be formulated in relation to the future of the Kenyan coffee production. The experimental design is better explained in answers to l. 170-175 than in the new version. But were plots permanent or changed place every year? This would impact the ANOVA procedure (repeated or yearly). While statistical analyses show small numbers of degrees of freedom for the error term, significant results were shown for three interrelated soil properties. The authors must explain why crop yield was not taken into consideration (see their pending answer). Conclusions should be made in relation to hypotheses. Future areas of research could be formulated, but crop productivity is of prime importance for the farmers.
Abstract : correct several typos and units (g cm-3, add spacing)
l. 13-16 : You mention four treatments, of which there are three with ridge tillage, three with cover crop, three combined, and three controls (hence 12 treatments)? I guess there were 4 treatments replicated thrice for each treatment on a single site during two consecutive years on permanent plots, hence 12 plots monitored each year and 24 in total. Or each replication was represented by a single site receiving four treatments, and each site was as similar as possible as others to support the ceteris paribus assumption (assuming that all factors but the ones being varied are equal)? In the ANOVA presented in Tables 3 and 4, there are only 12 plots in total. Clarify.
l. 18: no significant differences between years?
l. 27-28: the farmers would also be interested in higher profitability.
l. 104: levels?
l. 124: groundwater or water?
l. 121-134: just making assumptions does not justify lone the study; clearly list hypotheses as (1), (2), … then formulate the objectives.
Section 2.1: there would be 28 sites (figure 1). Hence, 12 plots appeared to have been selected among the 28 sites. If so, say so. This has been stated more explicitly in their answer. Authors should mention that in the M&M section as well as the abstract. Otherwise, if there are only a few sites, authors should mention how representative there are of the Kenyan coffee producing region in terms of soil and climate.
Table 1: provide confidence interval for sand, silt, and clay. What is the significance level of confidence intervals (0.05?) about means? Correct units to g cm-3.
l. 180: list the baseline characteristics used in your study.
l. 181-183: obscure sentence, please make understandable statements
l. 193: prormance?
l. 205-206: How was your sampling procedure? What is a W transformation? Only treated field were sampled, not the orchard? If you make ridges and bean alleys, then soil properties changed on ridges but how does it apply to future orchard itself? Are significant differences between treatments sufficiently large to impact crop productivity and tackle erosion? Is there literature on this for tropical soils?
Section 2.5: if plots were permanent plot, then the ANOVA should consider time as repeated experiment. If the setup changed place every year, then ANOVA is conducted every year. Why then do you used ANOVA followed by One-way ANOVA. What is the difference between both?
Table 2: if you used ANOVA and Tukey, then the standard deviations should be the same for the same variable. This is not the case in Table 2. Why?
l. 254: where are the error bars. In general, we use error bars in figures.
l. 399: mention treatments and parameters.
l. 400: clear or significant?
l. 401: superior or significantly improved soil properties?
l. 402: moderate or intermediate?
l. 404: better of significantly higher performance
l. 405: may maintain or maintained? Compared to control?
l. 407, 410: tends to increase or increased significantly (p < 0.05)?
l. 411-413: move to discussion and expand on this.
Author Response
Please find attached my reply to the review comments
Author Response File: Author Response.docx
