the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Jan 2025
Fertilization turns a rubber plantation from sink to methane source
Abstract. The rapid expansion of rubber cultivation, driven by the demand for natural rubber in the tire industry constitutes a significant land-use change in Southeast Asia. This significant land-use change has reduced soil methane (CH4) uptake, thereby weakening atmospheric CH4 removal over extensive areas. While fertilization is a widespread practice in rubber plantations, its role in further weakening the soil CH4 sink remained poorly understood. Over 1.5 years, we measured soil CH4 fluxes biweekly in an experimental rubber plantation with four distinct fertilization treatments to evaluate their impact on the soil CH4 budget. Our findings revealed that fertilization not only reduced soil CH4 consumption, but also increased soil CH4 production. The difference in soil CH4 uptake between unfertilized plots (-2.9 kg CH4 ha-1 yr-1) and those with rational fertilization (-2.1 kg CH4 ha-1 yr-1) was moderate. Recommended fertilization rates reduced soil CH4 uptake by 60 % (-1.1 kg CH4 ha-1 yr-1), and heavy fertilization transformed the soil into a net source of CH4 (+0.3 kg CH4 ha-1 yr-1). The suppression of soil CH4 oxidation was likely driven by increased mineral nitrogen in the soil solution and soil acidification, while elevated dissolved organic carbon likely stimulated CH4 production in the topsoil. Most rubber tree trunks emitted CH4, likely of internal origin. Trunk CH4 fluxes ranged from -0.10 to 0.51 nmol s-1 per tree, with no significant fertilization effect. At the national level, adopting rational fertilization practices in Thailand could enhance the net soil CH4 sink by 5.9 Gg CH4 yr-1. However, this mitigation strategy would have a limited impact on the overall greenhouse gas budget of the agricultural sector in Southeast Asia, unless it is extended to other tree plantations and cropping systems.
- Preprint
(1528 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (extended)
-
RC1: 'Comment on egusphere-2025-2', Yit Arn Teh, 03 Mar 2025
reply
GENERAL FEEDBACK
Apologies to the authors and editor for my delayed review; I was off work on sick leave for part of February which put me behind on my external responsibilities such as reviewing manuscripts.
This is an interesting study tackling an important emerging issue; namely, the potential impacts of rubber plantation expansion on biogeochemical cycling of key compounds, such as methane (CH4). The manuscript was generally well-written and the experimental design was sound. The strength of this paper is in its mechanistic focus, which will enable the biogeochemical and modelling community to develop better predictive frameworks for understanding how rubber production systems contribute to land-atmosphere interactions. In particular, the paper provides strong evidence for the impacts of varying N inputs on soil CH4 flux, characterises temporal/seasonal trends in CH4 flux, and explores the relationship between CH4 flux and key environmental variables (e.g. air-filled pore space).
However, what I am less certain of is if the data collected here is sufficient to construct a whole ecosystem CH4 budget (see points 6 and 17). There are two reasons for this criticism: first, while soil CH4 flux is collected at regular intervals over 1.5 years, stem CH4 flux is only measured at 3 timepoints. This is not sufficient to infer CH4 emissions from tree stems for the entire observation period. Second, given that there are an increasing number of studies that use eddy covariance or automated chamber measurements to collect quasi-continuous CH4 measurements, it is my view that manual chamber measurements are insufficient at this point in time to construct an accurate ecosystem CH4 budget from. The main reason for this is that CH4 flux has been proved to be highly heterogeneous in time, meaning that low frequency, manual chamber measurements have been shown to cause large underestimates in CH4 flux. If this paper were published 25 years ago, where eddy covariance and automated chamber measurements were uncommon, then I think it would be reasonable to construct a soil CH4 budget from these data. However, I am concerned that in 2025 that a manual chamber paper may not contain accurate enough data to construct a credible budget from, given what we now know about the temporal heterogeneity of CH4 flux and the increasing use of quasi-continuous measurement systems. That said, I still believe that the authors can still discuss what these observations could mean for rubber plantation CH4 budgets. One way to do this is to calculate the budget as a post-hoc exercise and clearly flag in the paper that these calculations represent a “first approximation” or “preliminary” budget (see point 17). This means that the authors could still present their budget numbers but could also transparently acknowledge the limitations of the methods they have used. This is a subtle but important difference, because the authors are not claiming a high level of accuracy for a “preliminary” budget, but are providing indicative numbers in order to advance knowledge.
My other key criticism is that the authors need to write more clearly in the introduction about how they investigated the effects of environmental drivers on methanogenesis and methanotrophy (see points 7 and 9). The authors need to revise what they have written or describe the methods they used to infer changes in gross CH4 production or oxidation, given that they did not measure gross methanogenesis or methanotrphy directly.
Specific comments on different sections of the text are provided below.
SPECIFIC COMMENTS
- Lines 42-43 “...on the CH4 budget of rubber plantation remain poorly understood...” – Please provide references to evidence this claim and to serve as points of comparison with this study.
- Line 58 “...CH4 microbial oxidation...” – Word order; “microbial CH4 oxidation” is more grammatically correct.
- Line 80 “...competes with CH4 for the active site of methane monooxygenase...”: Please provide a reference to evidence this claim. This is a well-known phenomena so there are a large range of papers to chose from.
- Line 83 “...releasing NH4+ into the soil solution...”: Please provide a reference to evidence this claim.
- Liens 89-90 “...organic substrates derived from primary production, which can be stimulated by fertilizer inputs...”: Please provide a slightly more detailed explanation of how fertilizer inputs can stimulate production of organic inputs to soil.
- Line 107: Two concerns; first, given that an increasing number of studies are using eddy covariance, automated chambers, and/or modelling to construct trace gas budgets, use of manual flux measurements may not be considered sufficient by some readers. It may be safer to frame the research as process-based CH4 flux study (rather than using the word “budget”) or an experiment to establish a preliminary baseline for this type of production system. Second, given that stem fluxes were sampled at only a limited number of timepoints, my recommendation – if you do intend to frame this as CH4 budget study – to frame this as a soil CH4 budget study as I do not think you have enough data for the stem fluxes to do this part of the budget justice.
- Line 109 “...methanotrophic and methanogenic activities...”: Please expand this text to state how you quantified net CH4 flux, methanotrophy and methanogenesis. Quantification of gross methanotrophy or methanogenesis is non-trivial, so experts in the field would be interested to know how you determined the rates of these processes.
- Line 122 “...planting density of 500 trees ha−1...”: Please indicate if this is a normal planting density for the region, so that non-experts are able to ascertain if this plantation is representative of standard smallholder practices.
- Line 132 2.2 Methane Flux measurement: The introduction indicates that methanotrophic and methanogenic activity were investigated, but this section does not describe gross CH4 flux measurements or incubation experiments which could be used to directly ascertain methanotrophic and methanogenic activity at different times of year. If direct measurement of methanotrophy or methanogenesis were conducted, then my recommendation is to revise the introduction to better represent the actual research/measurements performed.
- Lines 142-143 “Trunk CH4 fluxes (FT-CH4) were measured in August 2023, October 2023, and February 2024 on 8 to 13 trees per treatment.”: Since trunk CH4 fluxes were only quantified on 3 occasions, these measurements will provide indicative values but may not enable the researchers to extrapolate more broadly with respect to the annual CH4 budget or overall system behaviour.
- Line 170 “Soil CH4 mole fractions ([CH4]S) were measured at two soil depths (10 and 40 cm) near 24 soil collars (six per fertilization treatments, though not evenly distributed across the four blocks).”: Soil gas concentrations appear to have been quantified at the same time as trunk CH4 flux, with 3 sampling campaigns performed over the duration of the study. For clarity, I recommend that the authors state at the start of this section stating that soil CH4 concentrations were only collected 3 times over the course of the study. Again, similar to the point made above (point 8), these measurements will only be indicative.
- Line 190 “Soil mineral nitrogen (NO3− and NH4 +) and phosphate (PO43-) availability was assessed using ion exchange resin bags.”: For clarity, add a sentence stating that inorganic N and phosphate were quantified at 4 intervals over the duration of the study. In addition, it’s worth stating that some of these measurements coincided with fertiliser application, enabling the investigators to quantify the effects of fertiliser input on inorganic N and P pools. I am aware that you state exactly when these measurements were conducted (lines 197-198), but these revisions at the start of this section would help the reader to understand the overall sampling strategy.
- Line 250 “...nitrogen concentrations...”: The fact that there is no significant impact of fertilizer input rate on soil N concentration implies that the applied N is lost from soil due to plant uptake, N gas flux, and leachate loss.
- Line 264 “3.2 Soil Methane Flux”: Given that soil CH4 flux was measured at different distances from the trees (see lines 135-136), was there differences in soil CH4 flux depending on proximity to trees or not? If not, please add a sentence to this section indicating that there did not appear to be differences in soil CH4 at different distances from trees. This is important given that some field studies have identified rhizosphere effects.
- Figure 2: It would be useful to see these data shown as CH4 mole fractions (ppbv) in addition to the ΔCH4. Please show these data, either by revising Figure 2 or placing the additional data in an appendix.
- Line 38 “3.6 Trunk methane flux”: One option may be to group the stem flux data with the soil flux and soil concentration data, so that all the CH4 data are grouped together.
- Line 353 “3.7 Methane budget”: Given that there are an increasing number of studies that use eddy covariance or automated chamber measurements to quantify CH4 budgets, it may be more appropriate to calculate and discuss the annual CH4 budget as a post-hoc exercise, rather than reporting on the annual budget in the Results; i.e. I recommend that this paragraph is moved to the Discussion and the authors discuss the annual budget in more speculative terms, rather than placing the annual budget calculations in the Results. The reason for this is that manual chamber measurements may not provide high enough frequency measurements to accurately construct an annual budget for CH4, given the high temporal heterogeneity than CH4 is known to exhibit. By presenting these data as a post-hoc calculation in the Discussion, the authors can openly acknowledge potential imperfections in the dataset (e.g. lack of high frequency measurements) while simultaneously providing readers with a first approximation of the what the annual budget is likely to be.
- Line 373 “These broad ranges likely reflect differences in edaphic factors across sites...”: Can you briefly summarise for the reader what you think the key edaphic differences are which could contribute to variation among sites?
- Line 376 “Seasonal variation in FS-CH4 were closely linked to changes in AFP...”: My recommendation would be to revise this paragraph so you state from the get go that the findings challenge or are different from prior research, showing that AFP was not as important for modulating CH4 uptake as fertiliser input rates. I would also recommend expanding the discussion to explore why AFP may not be a good predictor in this context; for example, does texture (i.e. sandy loam) or other soil physical properties mean that air/O2 diffuses more readily into the pore spaces? Many methanotrophs are microaerophilic (i.e. can function at 2% O2 or less), so unless the soil is really pushed close to anaerobiosis then it is possible that they may continue to function effectively even if AFP is low.
- Lines 388-391: Variation in the effect of N input on CH4 uptake is likely contingent on background N availability in soil (lines 396-397), with more N-rich soils exhibiting uptake inhibition whereas more N-poor soils are likely to show enhanced CH4 uptake (at least until other resources or environmental conditions constrain methanotrophy). Given that this is something which we have known since the early 2000’s (see for example Bodelier et al. 2000 Nature 403), my recommendation would be revise this paragraph to take this knowledge into account as the manuscript would then better acknowledge the theoretical framework that is in place.
- Lines 405-415: To some extent, some of this information was already raised in the introductory text; there could be value in shortening this paragraph (given that the mechanisms etc. have already been discussed in the intro), to make the text more compact.
- Lines 416-424: Please include the meta-analysis published by Banger et al. (2012) Global Change Bio as this publication pulls together data from a wide cross-section of papers on fertilizer impacts in rice systems (and is there pertinent within the context of understanding fertilizer impacts on methanogenesis. I believe that they even included data from rice systems that were intermittently drained, which provides insight into the role of moisture dynamics in altering CH4 flux.
- Lines 425-434: Joe von Fischer from Colorado State University has also published extensively on microsite-driven production of CH4 using 13C pool dilution, so would recommend reading and referencing his papers here, as there are some interesting insights from his work about C flow via anaerobic pathways (and their implications for CH4 production & emission in otherwise “aerobic” soil).
- Lines 435-446: Another, alternative perspective is that it is not the absolute amount of organic matter in the soil that predicts CH4 production but the amount of C flowing through methanogenic pathways (i.e. the “methanogenic fraction”, sensu Von Fisher & Hedin 2007). This is a subtle but significant difference, as isotope tracer and pool dilution studies have found it difficult to correlate the absolute amount of organic matter produced by plants (i.e. NPP or inputs of exudates to soil) with methanogenesis. See for instance Von Fisher and Hedin 2007 Global Biogeochemical Cycles and Yang et al. 2017 Global Biogeochemical Cycles 31.
- Lines 490 and lines 494-496: Is there published yield response data – i.e. data on the impact of different N input rates on rubber yield for Thailand? Or, are there data on the benefits of alternative N management practices (e.g. use of organic inputs like compost, manure, biochar, etc.). If so, this information could be useful to insert here because to persuade policymakers, local authorities and growers, one needs to provide a credible route towards reducing GHG emissions while also not threatening yields. The statement that “Applying rational fertilization practices to other tree plantations...” implies that growers are currently overfertilizing (which is believable) but data need to be provided showing that lower N inputs could provide similar yield outcomes.
Citation: https://doi.org/10.5194/egusphere-2025-2-RC1 -
RC2: 'Comment on egusphere-2025-2', Anonymous Referee #2, 17 Mar 2025
reply
General feedback
This study provides a comprehensive investigation of how fertilization rates affect methane emissions from both soil and trunks in rubber plantations. While the dataset is extensive and covers a wide range of indicators, I believe the discussion section could be further refined to enhance the manuscript's clarity and impact.
Specific comments
1. Provide experiment design details
- When did fertilization begin on this rubber plantation?
- What was the start date of the treatment?
- How was the fertilizer applied? Was it evenly distributed across the plantation?
2. Position of soil CH4 flux measurement
In L135-136, soil CH4 fluxes were measured at three distances from the tree rows (0.7, 2.0, and 3.3m).
- Could the authors elaborate on why these particular distances were selected?
- It would be helpful to discuss whether soil CH4 fluxes vary significantly at different distances.
3. Methane oxidation and production
Since this study only measured the net soil CH4 flux and gradients in soil CH4 mole fractions but did not directly measure the oxidation and production fluxes of soil methane, the discussion of fertilization's specific impacts on these processes should be approached with caution.
The discussion that net CH4 production mainly occurred in the top soil layer (L425-434) cannot be definitively supported by the current dataset.
4. Contribution of termites and soil invertebrates
L444: ”Termite colonies or Scarabaeidae larvae might also contribute to localised hotspots of CH4 production” and L432: ”Variations of O2 demand could arise from soil invertebrates,
such as leaf-cutting ants and earthworms, that bury plant debris or organic matter”.
If termites and soil invertebrates are considered important factors influencing CH4 production, please provide more information about their presence and activity in this rubber plantation.
5. Soil pH impact on methane processes
L410-411” Additionally, the decrease in soil pH observed from T1 to T4 with increased nitrogen addition is another factor known to inhibit soil CH4 oxidation”.
The significant decrease in soil pH from T1 to T4 is an important finding that warrants more in-depth discussion.
Given that soil pH has a profound impact on microbial processes, including methane oxidation and production, please elaborate on the potential mechanisms linking pH changes to methane dynamics.
6. Scale extrapolation
The extrapolation of results to a much larger scale in section 4.5 may be too speculative given the significant variability in soil CH4 emission fluxes observed across rubber plantations (as mentioned in L371-372).
Additionally, rubber plantations on peatlands may exhibit different responses to fertilization (as mentioned in L422-424), which should be acknowledged when discussing broader implications.
7. Yield data
It is recommended to supplement the yield data of different treatments. This is of great significance for understanding the treatments described in this article.
Citation: https://doi.org/10.5194/egusphere-2025-2-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 167 | 36 | 12 | 215 | 7 | 5 |
- HTML: 167
- PDF: 36
- XML: 12
- Total: 215
- BibTeX: 7
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 66 | 31 |
| Japan | 2 | 27 | 12 |
| France | 3 | 20 | 9 |
| China | 4 | 15 | 7 |
| Germany | 5 | 14 | 6 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 66