the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Mar 2026
Knowledge gaps, research strategies and robust assessment methods for soil organic carbon storage in coffee production systems
Abstract. Regenerative agriculture is increasingly promoted in the coffee sector as a strategy to enhance soil health and contribute to climate change mitigation through soil organic carbon (SOC) storage. However, robust evidence on the magnitude and drivers of SOC change in coffee production systems remains fragmented and methodologically inconsistent. We systematically reviewed 80 peer-reviewed studies assessing SOC stocks in coffee systems across major producing regions. Our analysis shows that the literature is dominated by synchronic (space-for-time) approaches (~91 %), while true diachronic studies quantifying SOC stock change over time are rare. Only a minority of studies accounted for soil bulk density differences using equivalent soil mass (ESM) approaches, fewer than half sampled below 30 cm soil depth, and more than half did not report previous land use, an essential determinant of SOC trajectories.
Across studies, SOC stocks were generally higher in coffee systems than in annual cropping systems and lower than in natural forests, with agroforestry systems sometimes approaching forest SOC levels. However, reported management effects were highly context-dependent and often confounded by differences in soil type, climate, and land-use history. As a result, current evidence does not allow robust quantification of carbon removal potential in coffee systems, nor reliable attribution of SOC differences to specific management practices.
Based on this synthesis, we provide methodological guidance for rigorous SOC research in coffee systems, including recommendations on study design, sampling depth, bulk density correction, laboratory procedures, and reporting standards. We argue that future studies must explicitly define research objectives, baseline trajectories, and the spatial scale and context to which their findings apply and distinguish between SOC stock differences and true sequestration rates. Strengthening experimental designs, harmonizing methodologies, and improving data transparency are essential to generate credible, policy-relevant evidence on SOC storage and climate mitigation in coffee landscapes.
Competing interests: At least one of the (co-)authors is a member of the editorial board of SOIL.
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 paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(550 KB) - Metadata XML
-
Supplement
(42 KB) - BibTeX
- EndNote
Status: open (until 16 Aug 2026)
- RC1: 'Comment on egusphere-2026-1479', Anonymous Referee #1, 27 May 2026 reply
-
CC1: 'Comment on egusphere-2026-1479', Matthias Kuhnert, 29 Jun 2026
reply
Dear Editors and authors
The manuscript “Knowledge gaps, research strategies and robust assessment methods
for soil organic carbon storage in coffee production systems” is structured in three sections, with the first providing guidance for SOC measurements in a coffee crop system, section two analysing existing literature due to impacts on SOC affected by land use or management practice change in Coffee crop systems and third, a roadmap to improve research on coffee crop systems. While the content describes crucial aspects in SOC assessments and how to increase SOC stocks in crop systems, the manuscript lacks clarity and does not provide much novelty. Section 2 describes a generic guidance does SOC measurements, with little specification for coffee crop systems. Even though SOC is measured for diverse purposes (e.g. research, carbon schemes, inventories, subsidies) the methods are not presented or discussed in the context of this diversity. There are plenty of protocols available, but not many references add to the different MRV systems, standards or protocols. The big chance to work clearly out the differences and specifications for coffee but, even though the attempts are recognisable, this is not clear presented in the manuscript. The second part, an overview about existing literature, concludes that there is not enough literature. As Ledo et al. (2019) and Ledo et al. (2020) made an analysis about perennials in general (including coffee) it would have been interesting, if the findings are in line. The major conclusions about data scarcity, unharmonized data sets and lack of publications to represent the diversity of environmental conditions is already made in these two publications which are both not cited, even though they analysed the same topic (even though in a broader context). The outlook and future demand is very generic and lacks innovation. This can be assumed as general knowledge in this field.
Even though there are a lot of interesting and relevant aspects in this manuscript, thew way they are presented, the lack of precision and the too long explanation of the details make this manuscript inacceptable. Also, it does include any novelty or new aspect. To make the manuscript acceptable, the entire manuscript requires a makeover and improvement. Therefore, the rating is major revisions, but only addressing the listed issues is not enough to improve the manuscript.
References:
Ledo et al. (2019). A global, empirical, harmonised dataset of soil organic carbon changes under perennial crops; Scientific Data
Ledo et a. (2020). Changes in soil organic carbon under perennial crops, GCB
Review comments
Box 1: The box is a good idea. The description of SOC stock incorrect or at least misleading. The description is for measured SOC stock. SOC stock in itself can be expressed for a layer or for the entire soil core. However, the stock is there independent of the measurement approach. Yes, depending on the measurement approach the result differs, but this formulation is miss-leading. Please correct this and describe only the carbon stock.
Box1: SOC accrual: the changes are not related to the BAU, but to a baseline. This baseline can be defined by the BAU or due to other criteria (e.g. actual management, conventional management, traditional management). Please change BAU to baseline (BAU can be added as example). Also, arguably an increase in SOC is climate change mitigation. A reduction of a decrease can be also accounted as climate change mitigation. SOC accrual does not necessarily affect an increase of SOC compared to the initial SOC stock, but this is a difference.
Box 1: SOC equilibrium: The last sentence is very specific. This can be more general: If there is a change in conditions… What defined environmental disturbances? Agricultural use is already a disturbance. SO, this is not about disturbances, but about changes.
Box1: SOC saturation: Please remove or refine this description. This is a very complex topic with different view Breuer et al. (2025, Nature communication) contradicts the here presented description (the clay mineral based theory and links the process to MAOC) and some studies question the theory about having saturation. Here. it could be described as a theoretical effect, but should not be concluded as existing, but presented as a controversial or still to discover condition. There is no clear conclusion yet in the community.
Box 1: Climate change mitigation: Please remove the text after “…IPCC, 2021).” There is no need to describe the complex dynamic and the actual description is incomplete. Starting a discussion about this must include permanence, leakage, upstream emissions, side effects, etc. This would be simply to long. Picking out one aspect is not sufficient. Cutting the text short would be the better option.
Section 2: The authors jump directly into the topic without highlighting the obvious: The analysis of SOC drivers require data. These data need to be harmonised and complete. … The actual text is too long and not precise. Please refine this section.
Table 1: I would not use “highest interpretability” as lab conditions might be better. Also, try to avoid superlatives, as you do not relate the criteria to anything (higher/greater than what?). Just use the terms high and great.
Table 1: The main objective of the analysis is not clear. Are these option describe baseline settings for an assessment in the field or do they describe options to analyse change in literature?
Sections 2.2 and 2.3: The text jumps between generic analysis and description and coffee specific examples. Please structure this clearer and focus only on coffee, only generic or start generic and then introduce examples for coffee.
Sections 2.2 and 2.3: These sections are too long, as they do not present anything novel. It is appreciated that this topic is summarized and structures, but please shorten these sections. Further, it might be best to focus on the coffee production in these sections. Presumably, the authors did this already, but it is partly mixed with generic text. Make the focus on coffee clear in the first sentence of the sections or the headings. The authors shouldn’t forget that the main impact on the study design are the capacities, which includes cost- and labour-restrictions. This is mentioned in the text, but often this is a major factor.
Section 2.2 and 2.3: this is more or less a baseline discussion. Please use the terminology and refer to more protocols which define the baselines. The chosen option of the baseline depends often on the objective of the task (e.g. research field or commercial farm) which should be elaborated clearer.
Section 2.4 summarises challenges of picking the sampling size for SOC measurements in agroforestry systems very well. However, the question is about how realistic is it to see this applied and what is the purpose of describing the ideal method here? The reason why this sampling strategy is partly to the knowledge, bu mainly to the capacities.
Section 2.6: This is not a very good section and has no clear objective. It also provides more a personal opinion linked to some publications. First, it should be separated between the duration for a project and the frequency. Second, the duration is suggested by the IPCC, third, there are several MRV systems available, that define different protocols for the frequency, fourth, (this was partly indicated) it is the question how it is quantified (measurement, modelling, remote sensing,….). Further, the example about tillage seems to be picked arbitrary. The remeasurement must be at the same time within the management dynamic, if possible. This is not only relevant for tillage, but also for pruning, harvest, fertilizer application, etc. This section lacks precision and clarity and needs to be re-written and completed.
Section 2.7: These are correct and crucial points highlighted here. However, bulk density is also partly derived by pedo-transfer function (this is not a very good option, but it is done). Further, there is no objective of this section. For ESM sampling the bulk density is required. Here it is mentioned, it is difficult to measure, and the result are very uncertain. What is the purpose of adding the two potentially new methods?
Section 2.8: What is the purpose of this section? Is this advice, is this a description about the options? Arguably, ESM is the better option. However, fixed soil depth is still a well-established method and is applied in several studies. Both, fixed soil depth and ESM have their uncertainty and introducing errors due to the method. In the end the only value that is correct is the total mass of carbon in a total soil core. Further, ESM is not practical for approaches outside research and sometimes in research. A more neutral presentation of the used methods can be added in the soil depth section. Just mention the two options and highlighting that literature found that fixed soil depth measurements over-estimate the SOC stock changes.
In summary: Based on the objectives section 2 is supposed to be a definition for best practice for a SOC assessment in coffee crop systems. In the actual form it is more an overview about generic option to assess SOC, which is provided in several protocols, MRV systems and standards, which are hardly cited. There is no need to list these methods, as there is hardly anything specific for coffee systems. This section is too long and it lacks precision for a suitable guideline. Writing style is partly too casual and reflects partly personal opinion. Assessments or “rigorous SOC research in coffee systems” can be applied for different purposes and the objective of the task will affect the method of choice, which is not addressed. This section must be shorter, if kept in, more precise and more focus and highlight what is a generic method and which additions are addressing coffee specific tasks. Further, the purpose of an assessment should be identified (e.g. for research, carbon schemes, subsidies, inventories) and the specification of the objective discussed. If the guidance is only for research, this can be shorter, as there exist enough measuring protocols, which can be referred to. Coffee specific adjustments can be highlighted. In this section only limited literature is used, as most likely these publication summarise the different points already good enough.
Section 3
Table 2: Is it management effects or management changes? Please list the management practices that are considered.
Table 2: there are indicators missing in the list: study duration, size of target are (or plot size)
Section 3.2: please indicate how many sites in how many publications. Does n indicates the number of sites or the number of publications?
In line 435: is this a control or a baseline?
Section 4 is very generic about SOC assessments and the understanding of impacts of practice changes on SOC in crop systems. This is mainly already well established knowledge and partly not capturing the entire content (e.g. only a limited selection of promising future technologies for measurements).
Conclusion
Lines 586-587: This is not worked out using literature or studies, but this is mentioned in the text.
Lines 588-589: NO, first the available methods must be applied under agreed standards to achieve harmonisation of existing data sets.
Line 589-590: This is no true. The suggested option are not cost-effective solution but costly options which are widely impractical for generic implementation.
Equal soil mass measurements
The authors highlight the ESM method as the only correct measurement option for the SOC stock change. Applying this method, corrects for changes in bulk density, which, de facto, takes the bulk density out of the equation. Comparing an initial SOC stock sample with a by the bulk density corrected re-measured soil sample will show only an SOC change (not an SOC stock change) except the SOC content that is additional or missing, in the added or not considered soil (e.g. if 27 cm in the 3cm missing to the 30 cm or for 34 c, the SOC content in these additional 4 cm ), which is only a minor amount. It is correct to have a critical view on the fixed depth measurement option, but the ESM is not the solution either.
General comments
Lines 81-83: This is not coffee specific and please do not the formulation context. The listed parameters and variables affect the growth of all plants, including coffee and the growth is affected by diverse environmental factors.
Lines 89: How often is coffee grown in an agroforestry system?
Line 152: double brackets
Lines 162-164: This doesn’t matter as the target scale for conclusion is also different. The more relevant problem is the data availability, as it is difficult to get data from commercial farms. Furter, the authors are correct in the point that either more data are required or the data are associated with a high uncertainty, which is difficult to quantify without additional measurements.
Line 189: here is repeats the content of lines 118 and following. Maybe the text of line 118 can be moved here.
Lines 199-200: Here it should be distinguished between the kind of project and involvement of the farmers. It might be a poor quality field is provided for the observation and the sampling which would affect the results. If it is an experimental farm or the farmer is committed to the project, fields that more represent the average conditions might be picked.
Lines 209-211: Where are the numbers coming from. Either provide references or remove the numbers. Considering the limited number of studies for perennials it seems unrealistic to speculate about SOC variability, especially, as there are no studies that can prove this. If the authors use numbers, please use arbitrary numbers and make clear that these are arbitrary numbers used as example.
Lines 220: double brackets.
Line 280: write VM0042
Line 281: 50 cm for agroforestry or in general?
Line 283-286: Should all studies that do not follow this guidance be discarded?
Line 293-295: Recommendations for SOC measurements in croplands or for agroforestry?
Lines 478-480: This contradicts the demand of measuring beyond 30 cm as requested in section 2.
Line 480-481: Is the bulk density measure on different occasions?
Line 491: please check the brackets.
Lines 513-515: Do the analysed studies do not specify the objective? The problem is mainly that each study has an individual objective, which makes the data harmonisation challenging. Further, applied methods are mainly constrained by economic limitations, which makes a suggestion for improve measurement concepts redundant.
Line 517-519: It should be mentioned that usually the studies that do not follow this concept are not designed, but using existing information. The key problem is that there are not enough funded studies to analyse SOC changes affected by changes in land use or management practices. This should be mentioned, as the knowledge about how to applied a good measurement concept is not the limiting factor.
Suggestions for improvement
Leave the section 2 out but summarise this in a short description about challenges in SOC measurements (refer more to other studies) and focus on the specific challenges for coffee crop systems. Analyse the practice changes but also list potential options and which of them are considered and which are not. There are several practices presented all over the manuscript but more as examples rather than an about complete list of options, combined with information about if this is already published, is there a specific measurement required, etc. Overall, focus more on coffee specific aspects and use other studies (e.g. Ledo et al) to highlight the differences between this generic study and the collated specific studies.
Citation: https://doi.org/10.5194/egusphere-2026-1479-CC1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 597 | 246 | 44 | 887 | 44 | 38 | 53 |
- HTML: 597
- PDF: 246
- XML: 44
- Total: 887
- Supplement: 44
- BibTeX: 38
- EndNote: 53
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Dear Editors and Authors,
This manuscript focuses on soil organic carbon (SOC) storage in coffee production systems and aims to identify knowledge gaps, evaluate methodological limitations, establish best practices, and propose a research roadmap for robust SOC assessment in global coffee‑growing regions. This work provides a systematic review of 80 studies, a comprehensive methodological assessment, and actionable guidance for SOC monitoring in coffee systems, representing an important contribution to the field. The core contribution of the manuscript is not that coffee systems are uniquely complex, but that SOC methods in coffee research are highly fragmented, inconsistent, and non‑comparable, creating an urgent need for standardized protocols. This is a valuable and timely contribution that can strongly support future research, monitoring, and policy applications.
However, the manuscript currently overemphasizes the particular complexity of coffee systems, which weakens its logical rigor and scientific credibility. Major revisions should be done before acceptance.
1. The paper contains too much generic background and common-sense discussion, which makes the text redundant and verbose. Please cut down repetitive general descriptions and condense the whole manuscript appropriately. 2. It is hard to see the special features of coffee production systems compared with other perennial crops such as cocoa and orchards. The authors need to clearly summarize the unique traits of coffee in root distribution, litter input, smallholder management, terrain and microclimate, and explain why coffee needs an independent SOC assessment system. 3. Most methodological suggestions are universal guidelines, not tailored to coffee’s actual planting conditions and field heterogeneity. More targeted operational recommendations for coffee farms should be added. 4. The manuscript only gives qualitative comparisons of SOC stocks, without quantitative summary of SOC ranges under different coffee management modes. I suggest that relevant quantitative synthesis needs to be supplemented. 5. Many cited studies confuse SOC stock difference with real carbon sequestration (net sequestration). The authors should clarify this conceptual confusion throughout the text. 6. Some paragraphs are too wordy; please simplify sentence structure and unify table, figure and reference formats.