Conservation and Conventional Vegetable Cultivation Increase Soil Organic Matter and Nutrients in the Ethiopian Highlands

Round 1

Reviewer 1 Report

The article is nicely designed and timely submission can be potentially accepted for publication after incorporation of the following suggestions 1. The abstract should be reframed by focusing on more qualitative results. The message of the study must be included in the last paragraph of the abstract. 2. Introduction, in general, is well designed but it should be reframed by adding the research gap in a more comprehensive manner. The hypothesis should be more clear. 3. Methodology is ok but needs restructuring especially production estimation and soil analysis part. 4. Results well presented 5. Discussion is fairly good but needs citations of recently published reports from the same line 6. The conclusion must be client-oriented. it should not be another abstract. Kindly reframed. 7. Language of the article, in general, is good but critical grammar improvement throughout the manuscript is suggested. 8. Authors are requested to be strict with the pattern and style of the journal.

Author Response

Dear Reviewer

Thank you for your comments. Our detailed response and the manuscript with tracked changes is attached

Best regards

Author Response File: Author Response.pdf

Reviewer 2 Report

The effects of vegetable production on soil organic matter is a good topic, however, some serious issues were concerned:

1. The knowledge gap is missing, the manuscript focused on  the effect of CA and CT on SOM, How to define CA and CT?
2. How to compare the effects of CA and CT without non-conservative agriculture and less or non-tillage control?
3. Onion, pepper and garlic were involved in this study, there are the inputs of organic materials, i.e., 4 tone/ha grass much and 5 tone/ha cow manure, The logic of the experimental setup is not clear. The inorganic fertilizer only considered urea, why not P and K?
4. The available K content in Table 3 is questionable. What are the yields of each vegetable? 
5. The conclusions almost made no sense, which almost were unreliable ,applicable or in the range of common knowledge. 
6. The experiment involved many factors, and  I strongly suggested that the author need summarize the results and raised counterpart knowledge gap, such as long term effect of vegetable production, vegetable rotation, organic materials input,  etc. 
7. In addition the topic seemed to be out the scope of this journal.

Author Response

Dear Reviewer

Attached is our response to the commnets and manuscript with tracked changes. 

Thank you so much for your time in reviewing the manuscript.

Best regards

Seifu, Tammo, Vera and Sisay

Author Response File: Author Response.pdf

Reviewer 3 Report

This is an excellent paper. Just minor changes 

Comments for author File: Comments.pdf

Author Response

Dear Reviewer

Thank you for your comments. In the attached response, we detail the changes made in the manuscript.  The original manuscript with tracked changes is appended to the response.

Thank you again for helping us to improve the manuscript

Seifu, Tammo Vera and Sisay

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The reponses to former comments were not satisfied, still the knowledge gap was missing, and and the setup of the experiments were really not reasonable enough as scientific research.

Author Response

Dear Reviewer 2

Thank you for your time reading the manuscript and making comments

The comment in round 2 was as follows:

“The reponses to former comments were not satisfied, still the knowledge gap was missing, and and the setup of the experiments were really not reasonable enough as scientific research.”

Our apologies for not satisfactorily answering all comments.  Reading over what we submitted in round 1, we misinterpreted the comment on the knowledge gap.  In this response, we have attempted to better respond on how we addressed the knowledge gap and comment on the scientific basis of the performed experiments.

RESPONSE  on KNOWLEDGE GAP

The knowledge gap is presented in the introduction as follows:

We introduce first the general problem that the soils have become degraded and it affects the livelihoods of the people, viz,

“The quality and health of soils are the basis for improved livelihoods and poverty alleviation worldwide [1].  Soils of sub-Saharan Africa (SSA), particularly in the Ethiopian Highlands, are largely degraded due to high pressure on land that increases nutrient mining, land degradation and limited organic or inorganic resupply[2].  Conventional tillage (CT) practices with the Marisha plow are common in Ethiopia under irrigated and rainfed conditions and, thus, have increased soil erosion and declined soil quality.”

Next, we indicate what agronomic practice (i.e., conservation agriculture) can improve the soils after they are  degraded:

“The use of organic mulch cover and no-tillage practice with proper crop rotations is called conservation agriculture [3].  It increases soil organic matter and consequently soil fertility through biological processes [12], enhances water and crop productivity [3, 13, 14], and encourages environmental sustainability by reducing soil erosion, groundwater contamination, and greenhouse gas emissions [4, 15].  Consequently, conservation agriculture (CA) is proposed as a practical solution in the northern Ethiopian Highlands to improve crop productivity and enhance environmental sustainability [16].”

It is followed by a section that implementation of  conservation agriculture has not been successful in Ethiopia (this is the knowledge gap), and we provide the reasons that it is the case

While CA practice effectively increases soil fertility and crop productivity using chemical fertilizers [3, 13, 17], successful implementation requires that the practice is integrated with local indigenous practices [18]. ……………..Consequently, the impact of CA on soil organic matter and macro-nutrients should be quantified over extended periods by farmers in their fields to understand its benefits.”

Finally, we state that our objective is to find answers to the  knowledge gap as

“The objective of this study was, therefore, to evaluate the effects of CA and CT practices on soil organic matter and nutrients under irrigated and rainfed vegetable on-farm productions.  Ten vegetable farms in the Dengeshita Kebele  of the northern Ethiopian Highlands were selected.  CA and CT plots were established on each farm, monitored,  evaluated for four years, and their impacts quantified….”.

RESPONSE  on SCIENTIFIC BASIS of EXPERIMENTAL SETUP

We respectfully disagree with the second comment that the experimental setup was not scientifically sound.  The experiment consisted of 10 replicates and was carried over a four-year-plus period.  It is the most extensive experiment on the effect of conservation tillage and conventional tillage on soil properties carried out on the effect in Ethiopia.  Few experiments with conservation agriculture and conservation tillage have that many replications and are carried out over four years.  Despite the ten replicates, we did not find a statistical difference between conservation agriculture and conventional tillage. 

Thank you again for the review

. 

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 for Water, MDPI

Belay et al, nr. 1354514

This paper presents the effect of conservation agriculture (CA) vs conventional tillage (CT) on 10 sites in the highlands of Ethiopia over 4 years. The authors added a substantial amount of hay as mulch in the CA system, and also a substantial amount of manure in both CA and CT. As a consequence they measured an increase in SOM, which was more pronounced in CA than in CT. This was accompanied with increased total N and available P.

Though the results are not very surprising and the general idea around CA is laudable, I have practical questions about the implementation of this system on a larger scale, with the following points (which are major reasons why CA is not more widely implemented, according to Giller et al. who you cite in your manuscript):

• How to implement CA with an increase of organic matter input (mulching, etc.) when the human pressure is so great, i.e. the need for fuel, fodder etc? how to create a virtuous circle again?
• Is this system practicable for any farmer or can it only be replicated by few farmers that have the money to purchase (or produce) these inputs in the amounts required?
• Would there be enough hay, and manure available to replicate this system on large areas? Or would it have to be focused on vegetable production only, due to shortage of input on the long term?
• If the hay is imported from other fields, which I suppose it was in your trial (intensive vegetable production), are you not depleting other fields from their organic matter? The same holds for the manure, which is likely produced after feeding animals with fodder grown on other fields.

The discussion should address these issues in reference to e.g. Giller et al

Below are my detailed comments:

Address 5 should be “Colombo”

Introduction

L40 has increased soil erosion?

L40 ha^-1 a^-1

L40 be more precise: nutrient content(?) and availability(?) have decreased

L45 decreases plant nutrient content?

L53 “increases SOM content” instead of “improve SOM”?

L62: 40%: is there a reference for this statement in the literature?

Materials and Methods

L103-104: are shown in Table 1

L113ff: was the hay coming from the plot itself or imported? I suppose imported. If so, how much area is needed to harvest 4t/ha twice a year? That should be considered in the discussion: is there enough hay production to possibly extend the implementation of such a CA system? I think one major hurdle to CA adoption is the availability of organic matter. The same applies to 5t/ha manure. Would there be a competition between mulch production and cattle/manure production at some point if CA was extended?

L115-116: maybe better to mention the cattle manure fertilization before describing the nutrient analysis?

L117: 4 tons?

L153: does this mean that you made the calibration standards with KH2PO4 salt? Or you used the salt as a quality control/reference only?

L154-155 please rephrase: if you used the same Kjeldahl method you can write “Total N in plant, manure, and soil was measured by Kjeldahl...”

L157-158: “So, we measured...”: is this sentence necessary? If you describe Kjeldahl digestion it is clear that this is your measure of total N I think.

L163: remove “a salt such as”?

L159 and 163: why water as blank? Did you take your reagents into account? You analyzed solid samples, so a reaction vessel going through your procedure with the same reagents (and water for dilution) as with the samples should be enough?

L165: CEC: did you measure remaining NH4 in the solution? Did you exchange the NH4 back with another ion and then measure NH4? Or did you measure Ca, K, Mg, Na in the solution?

2.4 Data analysis: what software / package / procedure did you use?

In the regressions figures (Fig 5) you show only the averages. Did you perform your regression analysis with the means or with the individual data of all plots? Because that could influence the results of the regression in terms of p-value

L177: “trend analysis t-test” do you have a reference or can you give more explanation on this method?

L181: 3.3.1 Subsection?

L188: Table 2 is missing

L189: agriculture  on (double spacing?)

L192-194: the statement is confusing because time and treatment are mixed and it is not clear what you want to tell. It is obvious that CEC increased in CA and CT after 3 and 4 years compared to the beginning, but it does not clearly appear in Fig 4 that CEC of CA > CEC of CT. At best it is a marginal difference of a few percent. Some effect size and p-value would help assess the difference.

L194-196: what reference have you used to judge that K was sufficient? National recommendations? Published research article?

L201: it would be better to show the p-values (not only p<0.05)

L201: “paired t-test” (not pair t analysis)

L202: in Fig 5a and in Table 4, I see slope of 0.62 for CA and 0.42 for CT instead of 0.43 and 0.28 as here in the text

Fig4: please indicate in the caption n=10 and the meaning of error bars (SD). Error bars are missing for CT (0-30) in (b)

Fig 4 (c): in 3. and 4. year there are large error bars, indicating that the distribution was not normal. Maybe a boxplot would be more suitable?

Fig5: to me it reads “Availanle P”. And the scaling (x and y) should ideally also be the same in left (0-30) and right (30-60) panel. For visual purpose, please try to better align the panels.

Table 3: How did you perform paired t-test on the Mean and the increase over four years? For the mean did you use the mean of each plot over 4 years and then do the test between CT and CA? And for the increase, did you make the paired t-test between the 4-year average of each plot and the starting value of each plot? I think that some more explanation would be needed in the Data Analysis section.

L227-228: “paired t-test”. Or don’t write anything, because you already described your data analysis in the methods

L 231 maybe omit “using t-test analysis”

L232: should be 3.5

L239 and at (p<0.10) – BTW is there a specific reason why 0.1 was used and not 0.05? Also please check my comment on “significance” below (referring to L285-287 but valid for all the text).

Table 5: I don’t really follow this table: why is the Trend slope difference not always within the upper and lower confidence limits (0.19 is not between 1.5 and 1.37)? And when 0 is not contained in the confidence interval, then should the p-value not be <0.1? Also please explain the significant level of 1.94: Is it the t-value, i.e. the number of standard errors to build the confidence interval? Looking at the regressions in Fig 5 you seem to have different numbers of data points, so the t value should vary accordingly?

Discussion

L258 It is not clear here that manure was also added to the CT. I can read it in the methods and further below in the discussion, but I think the paragraph should be recast to make clear that manure was added to CT in the same amount as for CA.

L259 “an upward trend”: “an increasing trend over time”?

L266: too short to show a significant trend

L266: How would longer experiment duration influence the “significance” of the difference? Do you mean that there would be more data, which would reduce the standard error and reduce the p-value? As my comment on Fig 5, maybe you could use all the datapoints in the regression instead of the means?

L267 and throughout the discussion: you have made the effort to review these studies, but are there not already review/meta-analysis available in the literature to show the variables of interest? You use the data for general statements, like “SOM and CEC correlate”, or “TN and SOM correlate”. But this is textbook knowledge and there are probably many studies showing that on very large sample sets. And I don’t think it is needed to select studies on CA specifically.

L279-280: Summing up aboveground OM input to explain the measured SOM Stock does not seem meaningful to me for 2 reasons: (1) There should be only a small portion of the OM added remaining in the soil, i.e. maybe a few percent of the OM added in the first year are remaining in the soil after 4 years. And (2) the root input to SOC is substantial but is challenging to estimate. So It can serve for an educated guess, at best.

L285-287: something is wrong in this sentence: (1) “not significant” is mentioned twice, and (2) R² as such does not inform on “significance”. In general, p-value is referred to when mentioning the “statistical significance” of a parameter. However, “statistically significant” is a term that is not appropriate according to recent recommendations of many statisticians themselves. Please have a look at these papers and try to adapt the text throughout, preferably refering to actual p-values rather than tagging a difference as “significant” or not: https://doi.org/10.1038/d41586-019-00857-9 and https://doi.org/10.1080/00031305.2019.1583913.

L292: What could be the mechanisms behind the relationship between pH and SOM content?

Section 4.3 - Available P: Did all studies you compare with use Bray 2 as an extractant as in your study? If not, the comparison is invalid.

L309: P content of clay...? It is about the relationship of P content with clay content

L311: the comparison is invalid if not all studies used the same extractant. Please remove Fig 8. Also, P availability depends on so many factors, not only OM, TN and clay but primarily fertilizer use (present or legacy), calcium, and iron/aluminium oxides.

Fig9: “OM changes (%)” are you referring to absolute or relative? e.g. from 1% to 31%, or from 1% to 1.3% after 10 years? The caption does not indicate that you show changes at all and rather suggest that you just show the content. Same remark about P. And also remove the P panel if all studies did not use the same extraction method (it is not comparable).

Conclusions

See my general comment at the beginning and please address that in the discussion and in the conclusions.

L350 – 351: remove

Reviewer 2 Report

The paper needs improvement. I reported my suggestions in the attached file. In addition, the quality of figures must be improved.

Comments for author File: Comments.pdf

Reviewer 3 Report

The manuscript deals with the effects of conservation agriculture (no tillage and grass mulch) on soil properties in a sub-Sahara area. The topic may fit the scope of Soil Systems, Agronomy or Sustainability rather than Water. Anyway, I leave this issue to Editor.

I further raise several issues that the authors may consider.

1. It would be much useful to add the results of vegetable production.
2. Soil nutrients increased in both treatments after four years (Table 3), denoting that the hypothesis, i.e. tillage leads to soil degradation, stands nowhere. The authors may reconsider the scientific question that they aims to answer.
3. The Discussion must be improved. The correlations between soil properties are too general to provide any useful and interesting information.
4. The authors confused with soil organic carbon and organic matter.

Other things:

1. In Fig. 4, the results on 3rd year appeared not useful. Indeed they were not well presented and discussed in the main text.
2. The authors may note the units such as g kg-1 and % for TN.
3. The data in Fig. 5 and Table 3 are the same?
4. Table 2 is missing.