the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Jun 2026
Enrichment of lignin reveals consistent anaerobic degradation and persistent vegetation signatures in the organic matter of diverse lowland tropical peatland profiles
Abstract. Tropical peatlands store a significant amount of carbon but are also one of the most vulnerable carbon stocks due to anthropogenic pressures and climate change. The stability and accumulation of the organic carbon stored in tropical peat systems, and its sensitivity to changing temperature and/or hydrology, is intrinsically linked to the organic matter (OM) character. However, we currently lack a detailed understanding of the OM characteristics in tropical peatlands, hindering the accurate prediction of tropical peatland stability in the 21st century.
In this study, we characterise the macromolecular composition of peatland vegetation, leaf litter, and across peat depth profiles using a range of tropical (n = 7) and some temperate (n = 2) peatland ecosystems that serve as comparison. This characterisation is achieved primarily via Pyrolysis Gas Chromatography Mass Spectrometry (Py-GC-MS), complemented by Fourier-Transform Infrared Spectroscopy (FTIR). We find that silicate mineral interference in hydrologically active sites makes FTIR challenging to apply in these tropical systems. Our results also demonstrate that all sites exhibit distinct pools of putatively labile and recalcitrant (plant) OM, with both shared and distinct downcore degradation features. Most sites exhibit a downcore relative enrichment in aromatic pyrolysates, such as from lignin, vs polysaccharide pyrolysates. This relative enrichment follows a logarithmic decline, especially in the anoxic horizons. Regardless of the decomposition of the peat, however, a pyrolytic fingerprint of the original vegetation persists. This unique fingerprint is likely a driver behind the microbial community’s speciality to degrade the OM in its specific peatland, an effect known as the home advantage theory. The predicable preferential loss of polysaccharides at depth and consistent aromaticity of the leaf litter in the tropical sites can aid peatland accumulation modelling and enable more accurate predictions of peatland dynamics under future climate change.
- Preprint
(4286 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 16 Aug 2026)
- RC1: 'Comment on egusphere-2026-3069', Anonymous Referee #1, 25 Jun 2026 reply
-
RC2: 'Comment on egusphere-2026-3069', Anonymous Referee #2, 06 Aug 2026
reply
The pygcms data for peat and plants of several tropical peatlands are very interesting and certainly a valuable addition to the existing literature. However, and unfortunately, the MS is not yet mature, in many aspects.
1) Figures and text are far too much, it can be divided into two manuscripts, or reduced to about 50%. Interpretations are often poorly connected to the data. On the one hand details that are relevant to support main interpretations are not shown (or not explored), for example the discrepancies between FTIR and PyGCMS. On the other hand, many details are highlighted, but these are often not well demonstrated by own data, and therefore for me not convincing/consistent, and likely misleading to a more neutral reader. These details cause a loss of focus. I recommend to make a choice, focus on comparing FTIR and PyGCMS of the peat samples, or focus on comparing peat and plant composition (PyGCMS), this could be two manuscripts. It is very courageous to include so many plant samples, but if you do, it needs a more critical data analysis.
I understand the addition of the FTIR data to enable comparison with other studies on tropical peatlands using aromaticity as a proxy, but would it be an idea to exclude the FTIR data to improve the focus? If you do include it, could you try and find support for the discrepancies between FTIR and py data (and thus interpret the data in more detail)? The interpretation of silica is now very quickly made, but can you check this with the TOC data, for example? The depth trends are the same for the peatland with large difference between the two methods, perhaps it helps to improve the presentation of the figure with the depth records, to visualize the similarities and differences better. And consult the literature on each of the peatlands to find out more about the composition of the inorganic part.
Section 3.3.3 can be omitted and/or combined with other sections. l770-805 can be omitted, there is no connection to the data. l806-829 content on future work can be omitted too, and the remaining can be combined with conclusions and/or discussion.
See also below more suggestions to reduce the ms in length
2) Can you reconsider and homogenise the term aromatics? Its use is very confusing, and often not correct. Aromatics is used for i) everything else then carbohydrates, but aliphatics and for example aromatics are very different, ii) as overarching term for MAH and PAH pyrolysates, and iii) as overarching term for MAH, PAH, phenol, and lignin pyrolysates (but these have a different environmental meaning). Avoid mixing the meaning of this general term ‘aromatics’. Best avoid using the term aromaticity for interpreting the own data. If referring to FTIR data (or literature that uses FTIR data), best use ‘aromatic band’ consistently (and provide the band e.g. 1506 cm-1). Furthermore, be aware when reading the literature, how ‘lignin’ or ‘aromatic’ is measured. Could you improve the terminology, and specify and correct where needed. Also, I would not mix ‘vegetation type’ and ‘chemical groups’, a woody peat is of course more ‘aromatic’ than a moss peat. Thus if you compare northern with tropical peat, and one is moss and the other wood, why not name it moss and wood. Decomposition of wood or moss, and thus the resulting more stable fraction, is not following the same pathway.
3. Pyrolysis data
- The hemicellulose assignment is not correct, pyrolysis of cellulose results in many of these products .
Table S1- identification
- Why not order to RT within each group?
- what is m-guaiacol? Can you be consistent in the naming of products?
- G012 not correct. the m/z suggests methylguaiacol, but not sure without seeing the spectrum
- P16 name is not correct. But it is a phenol indeed
- P19 this product is not common in NOM. Without more information on the spectrum I canmot judge it, but I suppose this is syringol based on RT and MW.
- G012, G014 and G017. If these are guaiacols, can you use the nomenclature of guaiacols?
- Catechols why naming some pyrocatechols and some not?
- B009 not correct
- H019 this is not a carbohydrate but a terpene
Table S6
- I did not and cannot check the quantificion of the data, but I checked p-isopropenylphenol, and it seems unlikely that this products, specific for sphagnum, is abundant in the plants that were analysed, for example in trunk and root (row 1283-1287) or rhizome, leaves, wood (rows19589-19591).
- on the other hand, you mention 4-vinylphenol to be a biomarker for some peatland plants in the MS, but as you can see in your data it was present in all plant tissues and plant species. It may raise doubts on the correctness of the data. Thus, I would recommend to check the quantifications, and the interpretations related to it.
Table S3.
- You include the phenols into the lignin pool, this is not correct, and biasing the statistics. P lignin (or H-lignin) is p-hydroxyphenyl not phenols
- what are the 3 last columns? Lignin and polysaccharides differ from the columns with the same name. and what do you consider as cellulose?
4. external standards -Section 3.3.1
L557-565 I do not understand this. Could you include the pygcms results of these standards in supplement? If you have a cellulose standard, then you should have observed that this did not result in levoglucosan alone, but included many products that are now labeled hemicellulose. It is not clear how the peat signal is compared with that of these standards, is it expressed as proportion of the sam total of products? Furthermore, the idea of including external standards is nice, but you should be very cautious and demonstrate that you understand your analytical method, your samples, and the effects of environmental processes. But even if the peat carbohydrates were exclusively composed of cellulose and hemicellulose and would not share products, the composition of (pyrolysis products of) hemicellulose differ among plants. Also, the ‘lignin cellulose hemicellulose space’, should be defined better, it must be clear which products are in there. Also, the lignin standard is a wood extract. If you have the lignin composition of your plant tissues and species available, does it make any sense to use this standard? And if it does, why would it help in interpreting the peat lignin composition if you have the lignin composition of the input materials? The alkali used to prepare the standard also changes the lignin structure. And also for the aliphatics, if you have the results of all these plant species, why would a standard of Agave be used to represent cutan and suberan sources in the peat?
It is also confusing to use different groupings of the standards, it would help in clarity to be consistent with these groups in the MS. Using levoglucosan to reflect cellulose seems reasonable, but it represents the major part of the ‘polysaccharides’. In figure 5 you use the labels hemicellulose and cellulose, and also polysaccharides, which is confusing and unclear. Be consistent in labeling pyrolysis products and/or its interpretation, thus levoglucosan instead of cellulose (etc), (the other labels are also not microbial material, cutan, etc. bult aliphatics and N-compounds). But if you chose to present interpreted terms (hemicellulose), then make sure that it is clarified which products include this label. Same for aromatics, phenols, lignin, P3 (not in the caption, what this is?)
5) PCA.
The figures are difficult to read. Can you label the products that do not behave similar to their assigned group in PCA figures, and/or provide the loading in table format as supplement? In the text is often mentioned ‘some specific phenols’ or ‘some specific carbohydrates’, but why not name them? The thin sample samples and lables are often not visible behind the bold arrows.
6, More technical suggestions (not complete)
Abstract
L25 Perhaps mention the range of max depths? This is a strength of the data.
L30 could you specify which groups, in addition to lignin, are in the aromatic pyrolysates?
This is to avoid confusion, lignin pyrolysis products are indeed aromatic according to the chemical nomenclature, but lignin and aromatic pyrolysates often have distinct environmental meaning. If it is everything else than polysaccharides, perhaps limit to mentioning a loss of the sugars?
L31 enrichment and decline is a bit confusing for me.
L34 I do not see a relation between the loss of sugars and assigned aromaticity of leaf litter.
Introduction
L67 can you clarify what is meant with reactive?
L65-80 I feel this part can be condensed considerably, since you focus on tropical peat perhaps there is no need to compare northern with tropical peat so detailed? In my opinion it can be reduced to two sentences.
L92-94 palm hardwood etc can be ombrotropic as well. I recommend using the same classification for the tropical and temperate sites.
L96-97 so you assume that no changes in plant species composition occurs when redox conditions change? I would use the word explore also here, instead of identify. I understand that differences between the peatlands are likely much larger, but it can be mentioned.
Methods
201-202 could you clarify the composition of the plant samples in more detail? To me it reads as if leaves and roots are mixed. I propose to extend column J in Table S2, instead of ‘tissues included’, provide a column for each sample to make clear these are not mixed.
L222 that depends on the pH and is not valid for peatlands in general, in minerotrophic peat there can be lateral input. Is there information on pH available from previous studies perhaps?
l230-235 can you add these data to the Supplement? It would help to understand what has been done
L241-242 if two different columns were used, could you indicate in supplement which samples were with Restek and which with the other one? To avoid that this separates the samples as a different column may contribute to the signal.
L248 ion fragment instead of ion chromatogram
L255. Could you explain how products were selected? Is it based on dominance, how? Do you have a homogeneous dataset, i.e. all selected products were quantified for all samples? Or did you quantify each sample based on the abundance in that sample? What is meant with peak area of all? Please make sure tha no misunderstandings can occur in how data were obtained.
L265 what is meant with cellulose? Pyrolysis of cellulose results in many products that cannot on beforehand linked to cellulose since they are now unique for cellulose.
L266-267 I would not name this grouping functionally. It is more very strictly (but out of context) chemical structure. Please define the full names of abbriviations P G S etc. it is confusing to use aromatics for benzenes and also for overarching aromatics.
R&D
L270 bands instead of carbon?
L273-274 how do you know it reflects vegetation and not oxidation?
Fig 2 seems that some records continue to deeper sections, why are these not included?
L274 I suppose everglades is USWCA1. Can you provide the abbreviations that are used in the caption also in the text?
L274-275 why is here the low sugar content interpreted as decomposition and above as plant characteristic? Perhaps it would help to be consistent in mentioning the vegetation type.
Section 3.1 it is very difficult to follow the reasoning if site abbreviations and names are not used consistently
L297-300 It may also related to mineral interference in pyrolysis-GCMS? Catalyst reactions can increase aliphatic and aromatic products (not silica, but metals, carbonates, and clay minerals)
309-322 I think this is not really relevant. And can be excluded completely. Perhaps parts of it can be moved to the discussion below, if relevant there.
L324-331 Caution if minerals other than silica are involved. 324-326 can be excluded I think. The semi quantitative aspect of the data is well-known and no need to prove that again. I would suggest to compare with TOC and CN. The matrix effect can be relevant and is not limited to the free aliphatic extracted fraction. The last sentence is also out of context.
Section 3.2
L338 needs more explanation. Refer to fig A1?
L339 phenols in your list are not p-hydroxyphenol lignin, except for 4-vinylphenol
L347 23 alkenes?
L346 proteins from fungal/microbial biomass instead of i.e. proteins. Plants also contain proteins
L349-350 phenols do not represent lignin
L349-351 I find it difficult to read this sentence. Could you simplify or clarify.
L354 so you quantified the products found in the peat also in the plant pyrograms? I suppose the plant were very very different? it is a good approach, but it is not clearly explained
L357 their assigned macromolecular origin. But I do not see that in fig A2, aromatics from wood and sugars from peat with other vegetation type? So you likely mean the chemical structure instead of origin, it also demonstrates the phenols being very variable.
L 365 it is not macromolecular assignment but chemical structure assignment.
The title of 3.2 is not clear for me, it seems that you mean vegetation type. Instead of shifts in plant species distribution. Please clarify, because this is something substantially different. If you mena plant species composition, then you should take into account changes with time/depth, and how you decipher this from decomposition. Or mention that you assume a homogenous peat (in terms of plant composition) with depth
L357 Be cconsistent in the grouping, in the figure caption it is different.
Fig 4a. is the symbol of the plant sample referring to the site it was sampled? Would it make sense to aseparate them based on family and or tissue? The symbols of samples and their name are not readable or difficult to read
L370 could you visualize this?
L371. Was the trunk sample without bark? If so that is worth mentinoning as it would explain this observation
L374 can you calculate whether the 29 and 31 indeed had a higher contribution to the alkane group in these samples, and to which extend? If not sufficient, exclude this explanation on residual plant waxes? You have the data, so no need for speculation and unclarity.
I cannot read the samples of plant and litter
L377-379 I see two phenols with low negative loading, I would not name that a variety, and your data seem to indicate that this is due to polysaccharides and cellulose in particular
379 combine with lines 371-372
PC3 aliphatics and two phenols negative, and some sugars and aromatics positive along with one or two phenols.
L383-385 again you come up with explanations that are not visible in the data. if you see a separation in P and S lignin, then label the compounds so that this can be verified by the reader (given your compound list, I suppose that the phenols with negative loading on PC3 is 4-vinylphenol?) Ah now I see that the aromatics may be syringols, could you use a more contrasting colour, I hardly see a difference. What about the G lignin negative on PC4, is this not worth to mention, seems that PC3-4 separates the different lignin moeites.
L384-385 it is not clear for me, the figures must be improved. What do you mean with plant type, species, or family, tissue? Main controls on plant composition, this sentence seems self evident?
Perhaps consider moving the whole plant PCA to the supplement, and summarize in a table the main findings of PC1-PC3, regarding tissue and species. If the individual products behave according to chemical grouping, why not provide a Table with the average abundance of groups for familily and tissue. or somehow visualize the differences between plant samples better
L389-391 again a statement from the literature that is not connected to the data. you have the data, so if diketodipyrole is quickly degraded it should be higher in plant than peat. From your N compounds the others are not commonly associated with plant proteins in SOM.
L392 are catechols derived from G and S lignin, is that from the reference you provide? and.. catechols were likely present in your sphagnum samples as well. Did you have them in you alkali lignin standard?
386-391 I see the reverse for alkanes etc: positive. Can you also provide an overall interpretation for this combination of products, N, aliphatics and aromatics? It may also point towards algae?
L394 not relevant if they are common here, but where they plot in PCA, and how they differ from the furans etc., and this is not visible from the figures.
L398 do you mean aromatics or aromatics?
L401 more complex than pyrogenic OM and polyphenols? It is not clear what you want to say here, can you indicate what type of molecules
L401-402 I see negative on PC3 particularly aromatics and N
L403 I see that the catechols are particular high on PC4
L408-409 p-coumaryl = 4-vinylphenol
L410-411 yes seems logical , but that can be said on beforehand with the ID of your selected phenols.
Tor Royal etc is not indicated in the figure or its caption, please provide the abbreviations also in the text.
L416 aliphatic are positive on PC1
L417 I see UMP all with negative scores on PC2 and positive ones on PC1. I see USWCA positive on PC1 and PC2.
L417-418 very difficult to extract from the figure. And particularly a loss of sugars, instead of enrichment of lignin and aliph?
L418 it is not peat decomposition at depth but more decomposed peat at depth
L420 what is common, the loss of sugars with depth? Please be clear
420 thus assuming that the vegetation was similar with time
L425 not clear what you mean. Why stronger preferential decomposition (of what) (with depth?) at tropical sites.
Section 3.2.2
L430 is there a difference in peat composition and OM characteristics?
Fig 6 If you could plot alkanes and alkanes below each other it would be easier to read this figure. What is the meaning of fractional abundances? Percentage of aliphatics? Alkanes as percentage of alkanes?
Section 3.2.3 can be reduced to a few sentences in my opinion. There is too much detail that is not connected to the data, and or the data are not relevant in the context of the ms (e.g. fig 6b)
The ternary diagrams can perhaps be moved to supplement as well? To me it is not helping to get the message.
I stopped with reading in detail here. I recommend to first improve the focus.
Citation: https://doi.org/10.5194/egusphere-2026-3069-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 54 | 27 | 7 | 88 | 6 | 5 |
- HTML: 54
- PDF: 27
- XML: 7
- Total: 88
- BibTeX: 6
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript characterizes the macromolecular composition of organic matter (OM) in seven lowland tropical and two temperate peatlands, using pyrolysis–gas chromatography–mass spectrometry (Py-GC-MS) as the primary method. Through analysis of vegetation, leaf litter, and downcore peat profiles, the authors show a consistent enrichment of aromatic (lignin-derived) pyrolysates relative to polysaccharides with depth, following a logarithmic decline that is most pronounced in anoxic horizons of the tropical sites. They also find that a pyrolytic fingerprint of the original vegetation persists in highly degraded deep peat, and discuss what this means for peatland carbon accumulation modelling and the home-field advantage (HFA) concept. The geographic and ecological breadth of the dataset is a real strength, and Py-GC-MS provides molecular-level detail that FTIR alone cannot. That said, a number of issues need to be addressed before the manuscript is ready for publication.
1. The role of FTIR should not be dismissed
In Section 3.1, the authors show that FTIR is affected by silicate mineral interference at the 1030 cm⁻¹ band in several hydrologically active tropical sites (Pantano, Inírida, Mpologoma), producing apparent carbohydrate enrichments that the Py-GC-MS data do not support. This point is well taken. But the manuscript goes on to frame FTIR as more or less inapplicable for tropical peatland systems, which overstates things. The downcore trends revealed by the Py-GC-MS-derived PPA ratio (preferential loss of polysaccharides, relative enrichment of aromatics with depth, logarithmic decline in OM lability) are in fact very similar to what Hodgkins et al. (2018, Nature Communications) found using FTIR-based ratios on a broadly comparable set of tropical and temperate peatlands. That the two methods converge on similar conclusions, despite FTIR’s known limitations in mineral-rich settings, says something worth noting: FTIR is still a useful screening tool. It is fast, cheap, and works well where mineral input is low. A more balanced treatment would acknowledge the two methods as complementary rather than presenting FTIR as a method that ‘could be inappropriate.’
2. Section 3.2.3: The argument for aliphatics as an "early degradation" signal needs clarification
Section 3.2.3 argues that aliphatic compounds are enriched in peat relative to vegetation and that this enrichment “signals early degradation.” As written, the section is ambiguous and could easily be misread. The authors should clarify their reasoning. Specific concerns:
The text moves between several mechanisms (condensation of biological alkyl compounds during humification, selective preservation of aliphatic-rich macromolecules such as cutin and cutan, preferential degradation of polysaccharides and lignin relative to aliphatics) without specifying which diagnostic features are actually being used to infer early degradation. Is it the enrichment of aliphatics relative to the parent vegetation? A shift in chain-length distribution? The n-alkane/n-alk-1-ene ratio? It is not clear from the text.
Plant pyrolysis products themselves contain abundant aliphatics (leaf waxes, cutin, suberan, etc.). The authors note that “both peat and plants have a similar pattern among the HMW components” (lines 524–526) and that the C14/C15 doublets “probably derived from cutin, which is common in plants.” So a key question remains: how much of the aliphatic enrichment in surface peat is simply inherited from the parent vegetation (aliphatic-rich roots and leaves), and how much reflects genuine diagenetic change?
The authors should spell out the diagnostic criteria used to identify early degradation, distinguish between inherited and diagenetically enriched aliphatic pools, and revise the section title and text so that aliphatic enrichment alone is not presented as sufficient evidence of early degradation.
3. Vegetation signal preservation at depth needs refinement
The authors state repeatedly that “a pyrolytic fingerprint of the original vegetation persists” in deep peat (abstract; lines 41, 760–762, 1270–1272). This is an important and well-supported finding, but the current formulation is too general. Two points need refining:
The HCA (Section 3.2.2) and PCA (Section 3.2.1) results show that the vegetation fingerprint is not retained uniformly. Some signals, such as 4-vinylphenol (P3) as a sedge biomarker, S-lignin as a hardwood indicator, and the polysaccharide signature of Sphagnum, are detectable in shallow peat but fade with depth as degradation progresses. What persists most robustly seems to be tied to the relative stability of certain pyrolysis products, particularly lignin-derived guaiacols and syringols, and to a lesser extent high-molecular-weight n-alkanes from leaf waxes. Their resistance to microbial degradation is what allows a chemotaxonomic signature to survive even in highly decomposed peat. The authors should state explicitly which classes of pyrolysis products carry the persistent fingerprint and why (their chemical recalcitrance), rather than implying that the entire vegetation signal is preserved.
4. Section 3.3.3: The HFA discussion is overextended and should be substantially condensed
Section 3.3.3 has no microbiological data of any kind: no 16S surveys, no metagenomics, no enzyme assays, no decomposition experiments. The section is therefore speculative and should be cut back substantially. The manuscript
The home-field advantage hypothesis, as originally formulated and as the authors themselves cite (Dehaen et al., 2025; Hoyos-Santillan et al., 2017), concerns the differential decomposition of autochthonous versus allochthonous organic matter. That is, microbial communities are adapted to decompose litter from their local vegetation more efficiently than litter imported from elsewhere. The authors extend this concept to argue that microbial communities at different depths within a single peatland are specialized to degrade the OM at that depth (lines 1252–1254). These are not the same claim.
Deep peat in tropical systems can be centuries to millennia old (the authors’ own age models span “decades to millennia”; line 1143). The vegetation, climate, and hydrology under which that deep peat formed may have been very different from conditions at the surface today. Attributing the persistence of a vegetation fingerprint at depth to a specialized, depth-adapted microbial community overlooks the possibility that the deep microbial community has itself shifted over time in response to changing conditions. Without compositional or functional data on the microbial community at different depths, the authors cannot distinguish between “the microbial community is adapted to the deep OM” and “the deep OM is simply more recalcitrant, and the microbial community is whatever happens to be there.”