the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Age and chemistry of dissolved organic carbon reveal enhanced leaching of ancient labile carbon at the permafrost thaw zone
Heather M. Throckmorton
Jeffrey M. Heikoop
Brent D. Newman
Alexandra L. Hedgpeth
Marisa N. Repasch
Thomas P. Guilderson
Cathy J. Wilson
Download
- Final revised paper (published on 28 Feb 2022)
- Supplement to the final revised paper
- Preprint (discussion started on 20 Oct 2021)
- Supplement to the preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on bg-2021-272', Anonymous Referee #1, 05 Nov 2021
General comments:
This is a very interesting study and a nice dataset. While the ultimate conclusions of the study are not especially wide reaching, this is an interesting dataset and what appears to be a robust and interesting analysis. The manuscript is quite short, not necessarily a bad thing, but the introduction is quite long relative to the actual results/discussion. The discussion wouldn't be harmed by the addition of a bit more of a deeper discussion and exploration of wider implications.
It would have been nice to see the introduction focus more on studies that have considered headwater catchments specifically in the Arctic as there are a few in the literature now, although still lacking as noted here.
But overall, I don’t really have any substantial criticisms of the manuscript other than the relatively minor points raised below. This is a sound study with some interesting and unique data adding to the relatively sparse number of studies looking into radiocarbon export in headwater streams in the Arctic.
Specific comments:
L52, 56. Permafrost “thaws” rather than melting.
L176. The Neff and Wild studies are of larger Arctic rivers, a better comparison would be other studies that look at DO14C in headwater catchments, e.g. https://doi.org/10.1088/1748-9326/aaa1fe , or see https://doi.org/10.1029/2020GB006672 for more 14C-centric studies.
L225. It is very hard to see from Fig 4 how the correlation between DO14C and CH4 is positive, particularly for the surface samples. The surface samples look like a straight vertical line, were these log-transformed for the correlation? The positive relationship between DO14C and CH4 in the shallow samples look driven by a clustering of higher CH4 concentrations in July (circles), it would be useful to justify why you can group by timing in the correlations here but separate by timing in other places, e.g. Fig 2.
Citation: https://doi.org/10.5194/bg-2021-272-RC1 -
AC1: 'Reply on RC1', Karis McFarlane, 03 Dec 2021
Authors’ reply:
-Thank you for your review and suggestions for improvements to the manuscript. We will add the references suggested for headwater catchments and appreciate these being brought to our attention. As pointed out in the reviewer comments, this is a small study and while we will add these additional studies to our references, introduction, and discussion, we do not plan to extensively expand on the discussion in our revision.
-The specific comments include 2 clear suggestions, which we agree with and will change in the revision.
-For the final comment regarding Line 225, we see your point and agree that for the surface and shallow pore waters the correlation analysis isn’t appropriate – as we state in line 226 the relationship is really that both CH4 concentration and DOC Δ14C values decrease from July to September. In the revision, we will remove the correlation results and focus instead on the seasonal increase in CH4 concentration from July to September for the surface and shallow porewaters, providing quantitative increases for each since it is difficult to see the change in the surface waters in Figure 4 (this is because of the large range in CH4 concentrations of the pore waters).This point was a relatively minor one, leading into the more interesting patterns in the deep porewater in the following paragraph, and this change will improve that flow in the revised manuscript as well.
Citation: https://doi.org/10.5194/bg-2021-272-AC1
-
AC1: 'Reply on RC1', Karis McFarlane, 03 Dec 2021
-
RC2: 'Comment on bg-2021-272', Anonymous Referee #2, 15 Nov 2021
General comments
The study presents porewater data for DOC/TN concentrations, SUVA values and 14C dates of DOC in July and September from drainages in Alaska. Authors relate their observations with leaching of DOC during the seasonal thaw that is labile. The introduction reviews general findings from literature and mentions concepts to characterize DOM composition (aromaticity, molecular weight, aliphatic compounds-microbial processing, vegetation-derived DOM). These concepts of DOM composition need to be more clearly and directly related with the proxies that authors report (C/N ratios and SUVA values) so that the reader can follow why authors conclude about the lability of their samples.
Specific comments
Abstract
Line 23 – Specify which “biogeochemical indicators” you mean to make the statement less vague.
Line 24-25 – Based on the data and its interpretation, it is unclear how authors conclude that the old DOC is highly labile.
Introduction
Line 27 – Consider adding a more recent reference: https://bg.copernicus.org/articles/11/6573/2014/
Line 41 – Consider replacing "exceeds" for "may exceed" to not overgeneralize as not all data in the cited references support DOC export being greater than NEE. For example, the study of Billet 2004 shows DOC export is greater but the data in the study of Christensen 2007 is less clear and for the study of Roulet 2007, DOC export exceeds some years but NEE is on average greater.
Lines 44-51 – The main message in this paragraph is that old DOC is labile thus relating permafrost thaw with potentially large C loss downstream. It may be also worth adding other studies that indicate that little old C via DOC seems to be mobilized or mineralized in thawing ecosystems - see references below:
https://onlinelibrary.wiley.com/doi/epdf/10.1111/gcb.15756
https://www.nature.com/articles/s41467-020-15511-6
Line 52 – Replace "melting" for "thawing", also in line 56
Line 59 – Do you mean "surface soils" here?
Line 62 – It seems the statement of DOC export increasing with streamflow is based on the spring thaw. The relation of DOC export being water rather than carbon limited is thus based on the fact that large DOC pools accumulate throughout several months in winter and then are flushed during snowmelt. As currently phrased it seems that this relation would also occur during other hydrological events such as after precipitation. Consider rephrasing as "This seems to be the current case for the Arctic as DOC export by streams and rivers largely occurs during the increase in streamflow during snowmelt, implying that DOC transport and production is water, not carbon, limited (REFS)" or alike.
Line 69 – Add “age” after “provide insights into the”
Line 74-79 – Consider adding before this last sentence that age typically increases with depth and that the flowpath associated to seasonal thaw may thus be reflected in the 14C-DOC downstream.
Line 89 – Rephrase sentence for clarity. As stated, it reads as the biodegradability decline from January to December rather than seasonally.
Line 100 – Rephrase. The part of "do not provide information about the locations within their watersheds..." is unclear
Line 99-115: Authors state their expected findings in this paragraph. The introduction has reviewed broad aspects of DOC cycling but it is hard to relate with these expected findings at the end of the paragraph. Specifically, 1) why do authors expect DOC to become more enriched in aromatics over the course of the summer? If this is related with the degree of microbial processing, it would be helpful to spell out more clearly this relation in the introduction. 2) Also, the first time that the concept of thaw lake appears is here and it seems reasonable to introduce it before to better understand the stated expectation.
Methods
Line 119 - Please describe what you mean by “drainages” or where were samples collected from. In soils/sediments from channels and streams or in soils adjacent to water channels? Just
Line 121 – This is unclear - what are internal and external drainages?
Line 145 – The statement of calculation of SUVA follows after a statement about absorption coefficients. For clarity, please state how you calculated SUVA and whether you used "absorption coefficients" (from your Equation 1) or "spectral absorbance".
Line 157 – Add "Dissolved oxygen" for DO
Line 163 – Add "significance level"
Results and discussion
Figure 2 caption – Specify units of C/N ratio: either mass (g/g) or mols (mol/mol)
Line 186-187 – What does it mean that DOC increased from July to September "in samples from the thaw table depth"? Do you mean the "deeper samples"? Rephrase for clarity.
Line 191 – Based on the data presented so far (14C-DOC, DOC concentration, DOC:DON, DON concentration), it is unclear why "undecomposed" fits in the statement. Also, it is unclear why "vegetation-derived C" fits in the statement. It is not simply "organic carbon"? Please rephrase or provide information that allows understanding how you link the measurements with those adjectives/sources.
Figure S1 and S2 – Please correct the units on the x axis of Figures S1 and S2
Line 215-217 – How do authors conclude the last part of the sentence "that has not previously undergone microbial processing and may be biolabile." The paper does not provide any experimental evidence of DOC lability. If authors want to relate their SUVA values with the degree of lability, they should describe and put in context the relation of SUVA vs lability observed in other studies. Clarifying how authors relate SUVA values with lability would help to clarify this.
Line 223 - Replace "were" for "was"
Line 226 – Explaining the potential reasons of these results and their relations is missing. In its current state, the results and correlations are presented without any further explanation. Why do CH4 concentrations are greater in July than in Sept? Why do 14C-DOC is higher ("younger DOC") in July than in Sept? Suggestion: Is it possible that both CH4 and 14C-DOC are related to vegetation seasonal patterns with more active vegetation in July releasing younger substrates (higher 14C-DOC) that are preferentially used resulting in CH4 production and higher soil concentrations? This influence of vegetation could be stronger in shallow depths which could explain why these relations are only observed in samples of the top 10 cm and not deeper down (30-40cm).
Figure 4 – Correct units of x axis in panel 4a. It probably should be mM or mmol/L?
Line 247-254 – It is unclear how this paragraph adds to the story. I guess the point of this paragraph is to discuss about the dominance of hydrogenotrophic vs acetoclastic at the site but it seems all the information is based on previous studies and poorly related with the findings presented in the paper. Consider reworking this paragraph and merge it with the previous one where the alpha fractionation data is presented.
Line 271 – Decreasing DOC "concentrations" or "14C-DOC"? Please clarify
Line 273-275 – Apart from the age, can authors explain how DTBLs differ in other ecological characteristics among young, medium, old and ancient? Is there a vegetation succession during the maturing of these ecosystems? Is it commonly expected to have shallower thaw depths in younger DTBL? Such additions may help to better interpret the ecological relevance of the differences in DOC concentrations and SUVA presented in Figure 6.
Line 275 – Does Figure 6a present only "deep pore water samples"? If so, please include in the caption of Figure 6 as it currently only mentions "DOC concentration across depths", which is unclear.
Line 280 – Authors refer to "the degree of organic matter decomposition". This goes back to a previous comment - authors should more clearly state whether high or low SUVA values are associated with higher or lower lability based on previous findings. This would avoid the confusion of thinking that high SUVA results in lower lability but rather here it seems that authors relate the SUVA values with the quantity of vegetation derived C, which seems to decline with DTBL aging. Please clarify how you relate SUVA values with lability and DTBL aging in your study.
Line 296 – Replace “older DOC and younger DOC” for “old and young DOC”?
Citation: https://doi.org/10.5194/bg-2021-272-RC2 -
AC2: 'Reply on RC2', Karis McFarlane, 06 Dec 2021
Authors’ reply - Thank you for your review and suggestions for improvements to the manuscript. We agree with your comments and will make the minor edits suggested in the revision, including the additional referencs you’ve provided. You also make several important points that we will address in the revision as we agree they will improve the manuscript.
1. Clarification of how we relate our measurements to lability of DOC. Thank you for identifying this issue. We will include the revision a more developed explanation of how characteristics (SUVA and DOC:DON ratio) of the dissolved organic matter chemistry can relate to decomposability and the extent of microbial processing of organic matter. This should address several comments (in the abstract, introduction, and results/discussion) related to how we interpret our data to suggest a highly labile DOC source and differences with age across the drained thaw lake basins. For the sake of clarity in this response, we interpret higher DOC:DON and SUVAs to reflect fresh vegetation inputs, which should have: 1) higher C contents relative to N (these ratios converge with microbial processing); 2) higher SUVA 440, indicating more high molecular weight compounds (which break down into smaller molecular weight compounds with microbial processing); and 3) higher SUVA 254 and 350, indicating greater aromaticity and lignin content – compounds that are present in plant C inputs and break down with microbial processing.
2. Clarification of the trends in CH4 concentration for the surface and shallow samples. R1 also pointed out an issue with this sentence (L225-226). In the revision, we will remove the correlation results – this is not a meaningful correlation between 14C and CH4 concentration, but a shift in surface and porewater chemistry from June to September. We had attempted to clarify that it is the seasonal pattern that was meaningful, but agree that it was confusing to include the correlation results. We will only provide the ANOVA results, which show the decline in CH4 concentration in surface and shallow porewaters from July to September. CH4 concentrations were presented and discussed in the Throckmorton et al., 2015 paper, and we will refer to that paper here as well for more discussion on these patterns. This point was intended to be a relatively minor one to place the new data in the context of the other geochemical data available for these samples and to lead into the following paragraph focused on the deep porewater. We think making this change to this sentence will help clarify our point concisely. You pose interesting questions, however! As discussed by Throckmorton et al., there does seem to be a shift not only in the methanogenesis pathway (more acetoclastic in July and more hydrogenotrophic in September) but also in the depth where most of the CH4 is being produced (in shallow porewater in July to deep porewater in September). Throckmorton et al. also suggested this might be driven by a shift in C source related to vegetation inputs. We have looked to see if we observed correlations between CH4 concentrations and the DOM chemistry variables we measured. If the increase in CH4 was driven by plant inputs, we would expect to see a correlation with the indicators of fresh plant organic matter sources (higher CH4 with higher DOC:DON or SUVAs) – but we did not observe any correlation between CH4 concentration and OM chemistry. An alternative to the plant C inputs hypothesis, is that the CH4 values may be higher in July than in September because of a buildup of acetate over the winter when cold temperatures should inhibit acetoclastic methanogenesis, but we did not measure acetate at our sites and this a speculation. We do not think our data are strong enough to support one of these explanations over the other. This is a point that we could develop in the paragraph following the discussion of the CH4 fractionation factor, though we had not intended to do so as we have not added anything new to what was discussed previously in the Throckmorton et al paper.
3. More information about DTLB ecology to aid in interpretation of results. This is a great suggestion and will be worked into the introduction and this discussion paragraph (starting at line 273). In the introduction, this will include a bit more description of how DOC chemistry and age might change with different watershed drainage characteristics as well. For example, in DTLBs in our study region, organic layer thickness and degree of organic matter decomposition increase with increasing DTLB age. Young and medium aged basins have relatively productive plant communities with grasses in young basins yielding to sedges in medium basins. Advancing in age to old basins, distinct ponds form and mosses are found in addition to Carex. In ancient basins, the distinctive polygonal ground has formed with pronounced microtopographic rims and lows and overall vegetation growth is lower than in the younger basins. In addition, we will add to the discussion that our finding of lower SUVA350 and SUVA440 in DOC from old and ancient basins, is consistent with succession, soil, and landform development patterns as we would expect to see a decline in these indicators of vegetation-derived, unprocessed organic matter as the basins age and the vegetation communities become less productive. We agree these changes will improve the manuscript.
We are confused by one of the specific comments for the methods section: “Line 119 - Please describe what you mean by “drainages” or where were samples collected from. In soils/sediments from channels and streams or in soils adjacent to water channels? Just” We will use “watershed drainages” rather than “drainages” here in the revised manuscript. The rest of the paragraph explains from where and how the samples were collected and we are unsure what additional information you are hoping for in this first sentence. We are happy to consider additional clarifications if you can better describe what you find missing in the study area and sampling description in this paragraph.
Citation: https://doi.org/10.5194/bg-2021-272-AC2
-
AC2: 'Reply on RC2', Karis McFarlane, 06 Dec 2021