Depth Effects of Long-term Organic Residue Application on Soil Organic Carbon Stocks in Central Kenya

Müller, Claude Raoul; Six, Johan; Njiru, Daniel Mugendi; Vanlauwe, Bernard; Van de Broek, Marijn

doi:https://doi.org/10.5194/egusphere-2024-2796

Preprints

https://doi.org/10.5194/egusphere-2024-2796

Preprints

23 Sep 2024

| 23 Sep 2024

Depth Effects of Long-term Organic Residue Application on Soil Organic Carbon Stocks in Central Kenya

Claude Raoul Müller, Johan Six, Daniel Mugendi Njiru, Bernard Vanlauwe, and Marijn Van de Broek

Abstract. In arable soils, a substantial portion of soil organic carbon (SOC) is stored below the plough layer. To develop sustainable soil management strategies, it is important to assess how they affect the quantity of SOC stored in the subsoil. Therefore, we investigated the impact of organic and inorganic nutrient inputs on SOC stocks down to 70 cm depth in a long-term field trial in Embu, Kenya. There were 3 organic input treatments (manure, Tithonia diversifolia residues, and maize stover) and a control treatment, each with and without the application of mineral nitrogen. These different treatments were applied to a maize monoculture over 38 growing seasons (19 years). Our results show that manure application had the largest positive impact on SOC stocks compared to the control, which was observed down to 60 cm depth. In contrast, Tithonia diversifoliaand maize stover significantly increased SOC compared to the control only within the top 20 cm and 40 cm, respectively. Among the three organic residue treatments, only the application of manure had a significant effect on the SOC stock of the subsoil (i.e., the 30–70 cm depth layer). However, when considering the whole measured profile (i.e., 0–70 cm), all treatments led to significantly higher SOC stocks compared to the 91 ± 12 t C ha^-1 of the control: manure had the highest stocks (120 ± 24 t C ha^-1), followed by maize stover (112 ± 17 t C ha^-1) and tithonia diversifolia (105 ± 11 C t ha^-1). Mineral nitrogen application did not have a significant impact on SOC stocks down to 70 cm depth. Overall, our findings indicate that the subsoil in the studied agroecosystems is affected by the type of added organic amendments. Our results imply that gathering knowledge on the soil below the typically studied 0–30 cm depth layer will improve the overall assessment of agroecosystem properties, which is necessary to optimize soil system resilience, limit organic matter losses and improve crop productivity.

Received: 06 Sep 2024 – Discussion started: 23 Sep 2024

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 1352 KB)

Supplement (246 KB)

Download & links

Claude Raoul Müller, Johan Six, Daniel Mugendi Njiru, Bernard Vanlauwe, and Marijn Van de Broek

Status: final response (author comments only)

RC1:
'Comment on egusphere-2024-2796', Anonymous Referee #1, 18 Oct 2024

This paper represents a nice study of the impacts of different forms of agronomic management (w/ and w/o N amendment & 3 different organic residues) on deep carbon stocks. As the paper highlights, this is an area that is in need of more study because most studies focus on the top layers of soil, whereas we know that much carbon, and especially stable carbon is in deeper soils. My biggest comment is on the statistical analysis, and potential interpretations. I applaud the authors for conducting a power analysis and being quite clear about where there are limitations on sample numbers. That said, I think there may need to be a little more caveating of the results earlier on in the study, and also an indication that there is also the potential for type I errors as well (I believe...). I'm also slightly concerned about model over-fitting with the inclusion of all of the interactions, given the number of samples present in the study. Especially given that almost none are significant, and also I don't think there is sufficient justification for inclusion of all of the interactions - ideally, the linear model should include interactions with sound theoretical justification and hypotheses around why these interactions might reasonable shift overall carbon levels, and specifically deep carbon. An additional analysis that only subsets deep carbon may also be interesting to show, given that this is the piece that is particularly novel for this study. I presume that much of the presented model is driven by the carbon in the top soil. Some level of sensitivity analysis would also help make me more confident in the results of your model - e.g. do model results stand when you include fewer variables/interactions. This could be presented in supplements. Finally, I think this study should be published, but the language around your findings should be slightly less 'confident' given the results of your statistical analysis - this could just be done earlier on.
I'm a bit confused on the calculation of stocks given that I think you only measured a subsets of the soils at depth - in line 165 you say that you look at layers of 5cm increments, but these do not add up to the whole soil profile - do you assume that 45-50 is representative of 40-50? Perhaps I missed this explanation, but I think there should be some statement that makes clear what assumptions you are making on the calculation of stocks.
The intro covers some good ground but in its current form read a bit choppy. The isotope paragraph feels like its just floating and is not needed in the intro. The flow between paragraphs could also use a bit of work. A bit more context for the specific system could be useful in the final paragraph with the research questions. I also think that these questions still read somewhat broad, and contextualizing them to your system can add some useful specificity.I personally also think its nice, even if briefly, to state what you expected out of these different schemes given the current literature.
I'm not totally convinced that the isotope work fits in currently in a way that feels meaningful. The final data is somewhat mixed in its results and while I think its interesting, I'm not sure that in its current form that it is adding to the manuscript or primary findings. I think the paragraph in the intro feels quite plopped in, and while I'm not against its inclusion, the authors should work to make its inclusion feel less like an add-on.
The following are more specific/smaller comments that I think would strengthen the manuscript:
33: citation for ...groundwater and the atmosphere; "it" should be clarified - I presume it is something like "intensive agriculture" or something?
41: OC is not defined in its first use.
42: "This is clear from the fact" is unnecessarily wordy
49: Are there key functions other than nutrient management that you think are relevant to bring up? What are the key functions
66-67: "However" not "Nevertheless" works better; I don't think you have defined OR - I presume organic residue.
92: Not sure about this paragraph and how it fits in.
115: I'm not experienced in this region, but I'm a bit confused about the rain seasons - is it raning all year round? the given months span the entire year.
117: passive voice not needed "having been" => " the site was originally covered... before conversion to" would be clearer.
125: Would be good to clarify that these treatments are the same as in this paper - this whole paragraph maybe feels like it should be in the intro justifying the gap in the literature of deeper depth investigations. This paper seems like a fundamental work that you build off of.
139: Are these typical application rates for the region? Also some justification of Thitonia, corn stover, and maize (which I presume has happened in previous papers) would be good to include in brief somewhere. Perhaps in Study Area. Jumping back to "The organic resources" is a little confusing after discussing other amendments. Also, are organic resources = organic residues as you've referred to them before. Consistency in terminology would be good.
Para at 146 is not needed.
171: no need to repeat. Decide where and how you want to organize without repetition from above paragraph.
182: HCl DID or DID NOT show a reaction? If there is a reaction, I would think this means there ARE carbonates?
205-207: Extra line?
256: "THis was only the case for...: and 267 "The results indicated..." should be in results, not methods.
273-274:Cohen's d - "d" should be ital. "Significance" not "Significant"
287: MIneral N fertilizer - spelling inconsistency
Figure 3: the overlapping error bars make it pretty hard to read... any ways to clean this up or make it more legible would be good.
387: Unclear when you are/aren't italicizing
392: Given the immediately preceding statement on your power analysis, I wonder if you could rephrase here to bring your conclusion in better alignment with your actual statistical tests.
449: "Very" would be better than "much responsive"
476-481: "Therefore, although there might be..." This sentence is too long - break it up. I'm also a bit weary of the sudden introduction of the co2 sequestration piece. Might be clearer to just keep it to your context of increasing SOC to offset losses in this cropping system. Some statement on whether you think there's any ways to actually stabilize or rebuild SOC such that there isn't continual losses would be interesting to me as well... how much of this is the continual maize?

Citation: https://doi.org/10.5194/egusphere-2024-2796-RC1
- AC1: 'Response to Referees 1 and 2', Claude Müller, 12 Dec 2024
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-2796/egusphere-2024-2796-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2024-2796-AC1
RC2:
'Comment on egusphere-2024-2796', Anonymous Referee #2, 19 Nov 2024

Dear Authors, thank you for the opportunity to review your interesting manuscript. I hope my recommendations prove helpful in improving the quality of your work. I have attached detailed comments in the PDF for some major and minor revisions.
Overall, the paper is well-written, but there are several instances where shorter, more concise phrasing could enhance readability. The figures and tables generally complement the findings. However, many appear unnecessary for addressing the proposed research questions. For instance, Fig 1 and Fig 4 were neither introduced in the introduction nor used to discuss the findings, raising questions about the rationale for their inclusion.
I found the focus of the text to be inconsistent, oscillating between topics on the subsoil, the entire soil profile, and comparisons of the top and subsoil. For instance, the title emphasizes depth-specific analysis, yet the abstract lean toward only subsoil dynamics, leading to some confusion. Moreover, the inclusion of mineral N fertilization in an experiment centered on SOC stocks was inadequately justified, even if such N fertilization didn’t seem to impact the SOC stocks.
I find the introduction to be the most troubling part of the paper. The argumentation is often meandering, with unclear research gaps and no compelling new directions this this study seeks to address. Here also, C isotopes are introduced abruptly, without sufficient context or connection to the broader question of subsoil C dynamics. The questions are rather weak and fail to promise any novel or relevant insights, yet they remained only partly answered. Question 1 on how do organic residue application and mineral nitrogen fertilization affect SOC stock along a soil profile, was not answered. It seems self-evident, even without data, that organic residue application and mineral N fertilization would influence SOC stocks as was concluded in this study. I propose to rephrase the original questions into hypothesis-driven statements based on underlying mechanisms and processes to be investigated.
Unfortunately, while the study a range of interesting soil analyses, the paper remains too superficial, offerring limited insights and failing to highlight exciting research directions. In addition to my PDF comments, I have also made some detailed comments below.
Abstract
The abstract proposes a study on SOC dynamics in the subsoil, yet, the boundary of the subsoil remained vague and largely undefined throughout the abstract. The research problem was also not mentioned, despite using the space to express the findings hat manure had the greatest effect on SOC at deeper depths. Nothing was mentioned about C isotopes and key soil properties such as pH and CECeff although they seem to be key to the findings of this study. This raises the question of whether measuring C isotopes and all the other soil properties were justified for this study. I suggest to use the first 3 lines of the abstract to indicate the research problem succinctly, and report on all measured parameters so that the reader can have a complete overview of what is to be expected in the paper.
Introduction
The paragraphs in the introduction lack coherence and fail to provide an adequate background to clarify the actual scientific problem. The initial two paragraphs primarily restate textbook knowledge, emphasizing the importance of SOC stocks for maintaining productive soils. While this is a fundamental concept, devoting two whole paragraphs to reiterate this point is an inefficient use of space.
An attempt to address the research problem appears in the third paragraph, but it concludes without proposing any exciting new research directions or unresolved questions. For a study focused on subsoil dynamics, depth-specific differences in soil nutrient dynamics are only introduced in the fourth paragraph. Even here, no explanation is offered regarding the specific processes dominating subsoils that might influence topsoil functions or plant nutrient acquisition. While the statement that 50% of SOC stocks lie below 30 cm is noteworthy, it lacks context—such stocks may be significant for long-term carbon storage but could be irrelevant for plant nutrition.
Overall, the introduction fails to convincingly establish the novelty of the study. The proposed research questions appear to have already been addressed in standard soil science textbooks, making it difficult to discern what new insights this study aims to contribute.
Methods
Including a map showing the location of the study area in central Kenya would greatly assist readers unfamiliar with the geography, providing a quick and clear overview. Section 2.1.1 is presented as though readers already have detailed prior knowledge of the experimental setup, including the findings of Laub et al. (2023b). This creates the impression that the results of this study are merely a subset of data from Laub et al., leaving little room for novelty or original contribution as it seems everything significant has already been reported in the prior work.
A key concern with the experimental design is the potential for unconstrained parameters to influence the results. For instance, while P and K are reportedly evenly supplied across treatments, N is included as a treatment variable (±N). This setup could lead to cascading effects that influence the reported findings, but this critical aspect is largely unexplained. This raises doubts about whether the experimental design is adequately optimized to address the research questions posed.
The statistical analysis is somewhat unclear and inconsistent. The mixed models approach to account for site-specific variability is reasonable, but it might be worth considering whether a generalized linear mixed model (GLMM) would be a better choice, given the diverse data types among the covariates (numeric and categorical). Moreover, it is unclear whether all statistical assumptions were verified on the residuals or the raw data.
A fundamental question remains: on what theoretical basis were the selected covariates expected to influence SOC stocks? The paper does not clarify whether a stepwise selection process was employed to exclude less explanatory variables or whether the model is purely theoretical. For instance, the removal of soil pH and CEC from the model specifications is unexplained. It is of course inadequate to simply state that they had a strong relationship with the SOC stocks. Additionally, how soil depth is treated in the model—whether as ordered categories or independent variables—has significant implications for interpreting the outcomes. Clarification on this point is essential. In the end, how model parameters were validated and where necessary fine-tuned was not reported.
The use of a t-test over an ANOVA for comparing treatments and controls is particularly concerning. This choice suggests a lack of clarity on how to handle the soil depth variable, which further complicates interpretation. It is well known that multiple t-tests increase the risk of Type I errors (false positives), yet this issue is not addressed. Moreover, the assumptions underlying the t-tests were neither tested nor reported. Given these shortcomings, the justification for conducting a power analysis is unclear.
I strongly recommend a thorough re-evaluation of the statistical methods to ensure that each test is theoretically grounded and suitable for combining data from different treatments. This re-evaluation should include a clear explanation of why certain variables were included or excluded and how soil depth was treated within the model.
Results
The results section does not adequately address the research questions and instead shifts focus to soil properties that were not introduced earlier in the paper. For example, it is unclear why Section 3.1 is included, particularly as it is not integrated into the discussion. While these soil properties may be important, they are not referenced in the abstract or introduction, raising questions about their relevance and utility in opening the results section.
The use of vague phrases such as “non-significant but... the tendency” is problematic, as statistical outputs are binary (significant or not significant). Such wording undermines the credibility of the statistical interpretation. Although Fig. 1 highlights some interesting trends in soil pH and CECeff, these trends are largely ignored in the results section. For example, while manure treatment shows no change in soil pH across depth layers, there is a clear decreasing trend in CECeff with depth. Conversely, for the control, as well as the stover and Tithonia treatments, soil pH shows a somewhat increasing trend with depth, while CECeff remains relatively stable. What mechanisms might explain these contrasting patterns? This is a missed opportunity to link observed trends with meaningful interpretations.
Furthermore, the entirety of Fig. 4 could be removed or moved to the supplementary material without any loss of key information.
Discussion
A major premise in the discussion is that adding organic residues under the various treatments limits SOC loss, as these treatments consistently showed higher SOC stocks compared to the control. This argument is perplexing because “SOC losses” are neither defined nor measured as part of the experiment. Moreover, instead of offering a logical interpretation of their own results, the authors rely heavily on finding confirmation in other studies. This approach weakens the discussion, as it fails to provide plausible explanations for the observed outcomes. Consequently, the discussion lacks depth, giving the impression of a superficial analysis that does not fully explore the implications or mechanisms underlying the findings.
Conclusion
One would expect the two questions posed in the introduction to be revisited here, reflecting on how the findings contribute to a broader understanding of SOC stocks in the subsoil. However, this was not the case. Instead, the speculations presented in the discussion were validated as final conclusions without critically examining the insights provided by the results.

Citation: https://doi.org/10.5194/egusphere-2024-2796-RC2
- AC1: 'Response to Referees 1 and 2', Claude Müller, 12 Dec 2024
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-2796/egusphere-2024-2796-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2024-2796-AC1

Claude Raoul Müller, Johan Six, Daniel Mugendi Njiru, Bernard Vanlauwe, and Marijn Van de Broek

Supplement

https://doi.org/10.5194/egusphere-2024-2796-supplement

Data sets

Depth Effects of Long-term Organic Residue Application on Soil Organic Carbon Stocks in Central Kenya Claude Müller https://figshare.com/s/c2f2787b7a56ef7ad656?file=49027291

Claude Raoul Müller, Johan Six, Daniel Mugendi Njiru, Bernard Vanlauwe, and Marijn Van de Broek

Viewed

Total article views: 257 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
193	54	10	257	21	3	1

HTML: 193
PDF: 54
XML: 10
Total: 257
Supplement: 21
BibTeX: 3
EndNote: 1

Views and downloads (calculated since 23 Sep 2024)

Month	HTML	PDF	XML	Total
Sep 2024	66	9	4	79
Oct 2024	69	21	3	93
Nov 2024	35	13	2	50
Dec 2024	23	11	1	35

Cumulative views and downloads (calculated since 23 Sep 2024)

Month	HTML	PDF	XML	Total
Sep 2024	66	9	4	79
Oct 2024	69	21	3	93
Nov 2024	35	13	2	50
Dec 2024	23	11	1	35

Viewed (geographical distribution)

Total article views: 244 (including HTML, PDF, and XML) Thereof 244 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 18 Dec 2024

Short summary

We studied how different organic and inorganic nutrient inputs affect soil organic carbon (SOC) down to 70 cm in Kenya. After 19 years, all organic treatments increased SOC stocks as compared to the control, but mineral nitrogen had no significant effect. Manure was the organic treatment that significantly increased SOC the deepest as its effect could be observed down to 60 cm. Manure was the best strategy to limit SOC loss in croplands and maintain soil quality after deforestation.


Total:	0
HTML:	0
PDF:	0
XML:	0