Nonlinear turnover rates of soil carbon following cultivation of native grasslands and subsequent afforestation of croplands

Hernandez-Ramirez, Guillermo; Sauer, Thomas J.; Chendev, Yury G.; Gennadiev, Alexander N.

doi:https://doi.org/10.5194/soil-7-415-2021

Articles | Volume 7, issue 2

https://doi.org/10.5194/soil-7-415-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/soil-7-415-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 7, issue 2

Original research article

|

19 Jul 2021

Original research article |

| 19 Jul 2021

Nonlinear turnover rates of soil carbon following cultivation of native grasslands and subsequent afforestation of croplands

Guillermo Hernandez-Ramirez, Thomas J. Sauer, Yury G. Chendev, and Alexander N. Gennadiev

Download

Final revised paper (published on 19 Jul 2021)
Preprint (discussion started on 12 Feb 2021)

Interactive discussion

Status: closed

RC1:
'Comment on soil-2021-5', Anonymous Referee #1, 03 Apr 2021

In this study, Hernandez-Ramirez et al. use data from a previous publication (Hernandez-Ramirez et al. 2011), coupled with new measurements of C stocks and stable C isotopes to model the trajectory of carbon losses and accruals resulting from grassland conversion to cropland and subsequent afforestation via shelterbelts, respectively, using first-order kinetics. Overall, I feel the manuscript is too long, and needs significant work to shorten it and improve its readability. I realize there is quite a lot going on in terms of explaining the different sites, methods undertaken, and results, but I still feel there is a lot of opportunity to cut text without losing any information. In addition, much of what is presented here has already been presented in earlier publications - though the degree of overlap is not entirely clear - so fewer words can be spent reiterating what has already been described in those publications to further aid in shortening the manuscript. Some more emphasis on what aspects of this work are novel or noteworthy, overall and compared to the previous publications on these sites, would strengthen it as well. What follows are section-by-section suggestions and comments.

Introduction:

Define “long-term” somewhere in the introduction.

The introduction could use some background on SOM accrual rates, and the fact that accrual is assumed to be asymptotic, but that this is rarely tested/quantified. All of the popular models assume or produce this asymptotic behavior, but it’s rarely been verified in studies of accrual that are longer than xx number of years.

Some background on any previous studies which have provided data on SOM accrual for longer-term periods (or lack thereof) would be useful here.

I think the trees/shelterbelts ideas could be more clearly linked to SOM accrual and the specific knowledge gap linking the two could be better identified.

Line 76: the phrase “soil microbiology is sustained” is a little awkward and I’d suggest rewording. Maybe “soil microbial communities are sustained and diversified”. Or maybe something like “diverse microbial communities can flourish beneath mature trees”

Lines 91-98 are a bit repetitive and don’t convincingly explain why this knowledge gap needs to be filled. I think these lines could be compressed and the need for this knowledge could be more clearly/convincingly stated here. Maybe using logic similar to the above comment about background on SOM accrual.

Methods:

There is no description of soil processing (it may be in the previous publications cited, but I think it still should be stated explicitly here). Were the samples sieved, and roots and plant fragments removed prior to analysis? Was any organic material or O horizon removed?

Line 160: replace “studies” with “studied”

Lines 181-186: I like that multiple scenarios were evaluated, but some explanation of why these two scenarios were chosen would be helpful (why 95% and not some other %age?)

Line 198: Change “tall prairie” to “tallgrass prairie”

Lines 213: Replace “Normal” with “Mean” if you are presenting mean annual precipitation.

Line 214: Replace “Normal air temperature" with “Mean annual air temperature” if that is what you are presenting.

Line 236: “...assuming that net contributions to soil C storage were negligible…” gives me pause. To clarify, you are assuming that there is no total increase in C from crop-derived C inputs, and that allows for these quantification approaches to be used. But there will certainly be contributions of crop-derived C to the C pools, even if they are not increasing in size (see xxx). Can you clarify whether you are assuming no contributions at all (i.e. no change in ð13C from that of the native systems during cropping) or you are allowing for that change in ð13C but assuming no net increase in total C? This relates to the following comment:

Lines 250-258: does “ð13C native soil” refer only to the soil under native vegetation, or just whatever vegetation was present before afforestation? I’m finding this wording confusing because the cropped soils are also being used to represent the “ð13C native soil” here (lines 251-252). I think the wording here needs to be clarified or better explained, since this is key to the interpretation of the results. Based on line 395 and the caption of Fig. 5, it seems as though the authors are assuming “the rest of the soil C was attributed to remaining prairie C”, but this can not be the case if crop inputs contributed to the C there, which they surely will have.

Results:

In general I found this section to be long and difficult to follow at times. I recommend the authors condense much of the writing and work to make it easier to digest.

Line 288: Remove “(A)."

Line 290: Change "(Fig. 2A, Fig. 2B, and Fig. 2C)" to (Fig. 2A-C), here and elsewhere (e.g., line 300).

Line 296: Can you give the time spans here for reference?

Line 307: change “which can question” to “calling into question the” or “challenging the"

Lines 311-315: Sentence beginning on line 311 is not a full sentence. Maybe “while” need to be removed? Or this needs to be combined with the following sentence. In general these two sentences are a bit awkwardly placed and might make more sense if placed elsewhere.

Line 323-324: I think it would be good to clarify that this is only in the soil underneath the trees, rather than xx% C accrual for the whole field.

Line 318-328: If this data (Table 1) is the subject of another paper, it might be better to summarize it only very briefly and simply refer to the table.

Line 332: What did “normalization” entail?

Line 340: Maybe give unit or context after “i.e., from 1-0.711"

Line 350-353: This sentence is confusing to me. It is not clear what the 81% and the 94.5% represent - are they different timepoints or maybe different sites? After multiple readings I think I understand this to mean that after replenishing 81% of the C lost during cropping, the shelter belts had 94.5% of the C of the native grassland. I think you can word this more clearly.

Line 384: What do you mean by “clearly discernable” ?

Line 421: “stoichiometry of absolute capacities” is an odd phrase and I am not sure I understand what it means. “Proportionality of the initial prairie-C” is also a little odd, so I would suggest rewriting this sentence to be a little easier to understand.

Ind 456: remove “study"

Table 1. Perhaps add columns that shows the amount of time for each land use, or say in the caption?

Many of the figure captions are quite long. I think they can be shortened significantly without losing any key information.

Fig. 3. The Y axis wording is a little odd, maybe “Proportion of original C” or similar?

Discussion:

Overall, this section feels a little disjointed and lacks a good flow of ideas from paragraph to paragraph. Restructuring it so that it has a clearer arc and/or follows a similar flow of ideas as the introduction would help make the ideas easier to digest and follow.

Lines 492-497: This is a pretty clear explanation and I think it would be useful to have this much earlier, in the methods. I found it really hard to figure out whether this was the case from reading the methods section.

Lines 516-517: Rather than saying “trees generated the conservation of the entire prairie- legacy” I would say something like “afforestation did not lead to loss of prairie-C” or “prairie-C was conserved during afforestation”.

Line 527: I think it would make more sense to use “climatic” instead of “hydrothermal”, and do this throughout the discussion. Could soil texture or aggregation play a role here?

Lines 530-536: What about physicohemical protection mechanisms? SOM fractionation might give additional insights, and I’d suggest that would be another important avenue for future research.

Line 540-541: Change “shown to continually being lost” to “was shown to be lost continually”

Line 541-542: This sentence seems to contradict the prior and following sentences. This makes it sound like not much of the prairie C was lost (only 1.7% of the original prairie C over 55 years is a very small amount). How does this compare to the rate of loss of native prairie C in the cropland? Is it really something to note, or is it a relatively normal turnover rate for old C?

Lines 570-574: This is a nice explanation and something to this effect should be included in the introduction (either move this text there, or introduce the idea there and reiterate it here).

Lines 587, 591, 592: I have only ever seen “MRT” refer to the C itself, never an MRT of soil C accretion or depletion. What does this mean? I think it would make more sense to refer to the rate of accretion/depletion, or average time to xx% of total accretion or something like that.

Paragraph beginning on line 604: I agree with the suggestion of studying different physical pools of C in terms of stability, and I like the ideas introduced here. However, they seem to come out of nowhere given that they have not been introduced earlier in the manuscript. I might suggest including them earlier on in the discussion (e.g. when discussing potential differences in native prairie C turnover between sites) and then coming back to it here to expand on it.

Summary:

The authors state that afforestation is a suitable means to address climate change, but it would be helpful to put their results in context of the amount of area being afforested in these cases. While there was significant C accrual in the soils under the trees, the area being planted to trees seems relatively small, so what is the overall gain from this practice? Is it really very significant given that most of the cropland is not being planted to trees? I think that noting these relative area sizes and total C losses and gains in the croplands vs afforested areas is very important to put the results into larger context.

Citation: https://doi.org/10.5194/soil-2021-5-RC1
- AC1: 'Reply on RC1', Guillermo Hernandez-Ramirez, 08 May 2021
  
  The comment was uploaded in the form of a supplement: https://soil.copernicus.org/preprints/soil-2021-5/soil-2021-5-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/soil-2021-5-AC1
RC2:
'Comment on soil-2021-5', Anonymous Referee #2, 10 Apr 2021

General comments

In this study, Nonlinear turnover rates of soil carbon following cultivation of native grasslands and subsequent afforestation of croplands, Hernandez-Ramirez et al., use existing soil C data produced in the previous studies (Hernandez-Ramirez et al., 2011) in combination with new measurements of stable C isotopes to evaluate long-term C turnover rates in relation to land use changes, from grassland to cropland and subsequently from cropland to forest.

The paper reads well and I find this an interesting paper worth to be considered for publication in the SOIL journal.

The study is well introduced and the authors build on existing literature to highlight the effect of land use change on SOM. But I did not find information on previous related studies especially on long-term carbon turnover or approaches that have been used.

The authors did a good job on the materials and methods section. The study sites, the model, and the land use investigated are well described. However, the reasons behind the choice of the model and uncertainty related to the model are not clearly provided. Very little information is provided on basic soil physical-chemical properties that are known to influence C turnover such as texture, pH, oxides among others. These factors drive the C stabilization mechanisms which are briefly mentioned in the introduction and discussion. Having such information in the manuscript may highlight future studies as you mentioned in the discussion section.

The results section is clearly described. It has a lot of information that supports most of the statements made in the manuscript. The figures and tables are informative. Figure 4, the Y-axis is not scaled across panels. Is there a reason for this?

The discussion and conclusion are short but very informative. The authors did a good job here too. As for the materials, and methods, the authors mentioned the role of mineral and physical protection of soil C. Having information on the basic soil properties related to these mechanisms may strengthen some statements of the conclusions.

In the following sections, I will provide comments and suggestions for each section of the manuscript as indicated by lines.

Specific comments

Lines 83-84: You may consider revising this sentence. Does “biological-mediated decomposition” different from “decomposition”?

Line 86: Do “decomposition” and “mineralization” have different meanings in this context? One of them would suffice.

Lines 92, 94, 96, and 102 Authors mentioned “long-term” but it would be clear to the reader if this is clarified in terms of numbers.

Line: 112, Be precise by using numbers to reflect the chronosequence

Line 123: How deep was the core? Was it a one meter core or did the authors focus on 0-30cm? They also refer to previous studies for more details but it may help the reader if this information is added here.

Lines 143-145. It is good that excluding the contribution from erosion is supported with data on topography. But it would have been much better to provide this information here. The authors may consider saying “At our study sites, the slope gradient ranges from X to Y and the topography is classified as flat. Given the flat topography, we also assumed negligible C removals or additions due to erosion or deposition.”

The authors assume that the C input and output are balanced. Are there references to support this statement? It might be from the previous studies in the region or similar conditions. Usually, long-term C turnover is accurately or well described with multiple-pools models. What makes Eq. [1] a suitable approach for this study? The authors may consider adding that information in the method section.

Lines 152-156: Excellent idea, to use the cross-validation method to assess the model performance. In this section, the authors may also provide information on the total number of observations in the dataset used in this process. Given that sample size can greatly affect a model.

Line 236: The authors assume negligible contribution to C storage by annual cropland. How small is the C input from these sources? Are they really negligible? Any data from previous studies that support this statement? Unless crop residues (straws) are taken from the field, they should contribute to some extent to the topsoil C.

Line 314: As a followup to the previous comment, it looks like there are contributions from the recently added plant residues. How did the authors separate the C isotopic signature of the original land use (native grassland) from the recent crop residues?

Line 323-324: Same as the previous comments related to the “contributions of crop residues recently added”.

Line 456: I guess “study studies” should be “study sites”.

Lines 509-511. This is a very interesting finding. Why would old prairie-C decompose faster compared to fresh C input from roots and litter? The remaining prairie-C should otherwise be considered as stable C (less labile)? This comment is also linked to the statement in lines 470-471. I did not see data on the C turnover rate of the forest input.

Lines 521-522: Sites located in Norfolk and Mead have comparable temperature and precipitation. Did the authors look at the mineralogical characteristics of these sites? The difference may also be related to the C stabilization mechanisms.

Citation: https://doi.org/10.5194/soil-2021-5-RC2
- AC2: 'Reply on RC2', Guillermo Hernandez-Ramirez, 08 May 2021
  
  The comment was uploaded in the form of a supplement: https://soil.copernicus.org/preprints/soil-2021-5/soil-2021-5-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/soil-2021-5-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (28 May 2021) by Carolina Boix-Fayos

Dear authors,
My apologies for the delay of this reply. I have now reviewed the responses provided to all the questions raised by the referees and I have to congratulate both parties for the excellent work done in comments and responses. Both have fostered a rich scientific discussion that have certainly improved the manuscript. Please take care of including all the changes that you mentioned in your answers within the manuscript.
Just a minor question: the last question raised by the second referee on whether you have made mineralogical analysis of the samples that could made a difference in the OC storage is only indirectly answered. I understand from your answer that your idea is that microclimate and tree factors have more effect on soil carbon than soil properties. This also agrees with your suggestion that in this case soil texture does not seem to play a role on OC storage. But, I understand also from this, that the direct answer to the specific question of the second referee is that you did not have mineralogical analysis of the soil samples, did you?
Moreover I think your results are remarkable and that some of them called the attention of both referees, for instance the fact that crop residues are not increasing SOM on the long term. Those are important results that can be probably extrapolated to different situations and can be used to contrast opposite findings of other scientists. Those are also important results to question the positions of some politicians that justify very intensive agricultural practices on fragile areas on the basis of hight rates of organic carbon sequestration by agricultural soils.
Please go on with the corrections of the paper, and congratulations for the good work.

Carolina Boix-Fayos

Hide

AR by Guillermo Hernandez-Ramirez on behalf of the Authors (04 Jun 2021) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (16 Jun 2021) by Carolina Boix-Fayos

ED: Publish subject to technical corrections (18 Jun 2021) by Jeanette Whitaker (Executive editor)

AR by Guillermo Hernandez-Ramirez on behalf of the Authors (20 Jun 2021) Author's response Manuscript

Short summary

We evaluated how sequestration of soil carbon changes over the long term after converting native grasslands into croplands and also from annual cropping into trees. Soil carbon was reduced by cropping but increased with tree planting. This decrease in carbon storage with annual cropping happened over centuries, while trees increase soil carbon over just a few decades. Growing trees in long-term croplands emerged as a climate-change-mitigating action, effective even within a person’s lifetime.