the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Apr 2024
Plant–soil interactions underline the development of novel ecosystems after glacier retreat
Abstract. An emblematic symptom of climate change is the retreat of glaciers worldwide. As glaciers retreat, new terrains are exposed to colonization by a variety of organisms, leading to succession in plant communities and changes in soil properties. However, little is known about how the development of novel ecosystems emerging after glacier retreat depends on plant–soil interactions. Here, we investigated how glacier retreat influences the relationships between plant communities and soil functioning. We examined the diversity and structure of plant communities (functional composition, diversity, ecological indicators) and analyzed soil properties (pH, organic carbon, total nitrogen, C/N ratio, texture, available and total elements) along a glacier foreland comprising four stages of glacier retreat. The dominance of plant functional types shifts from herbaceous to shrubs and ultimately trees. Plant diversity initially increases after glacier retreat, along with an increase in soil organic matter and nutrients. Over 120 years, soils acidify at a rate of 0.02 units per year, the C/N ratio increases while plant diversity decreases. These findings provide novel evidence on the geo-ecological processes driving the development of new ecosystems that emerge from glacier retreat. As climate is warming and glaciers are retreating at increasing rates, pioneer herbaceous communities are quickly replaced by coniferous forests. As a result, biodiversity decreases while organic matter accumulation and soil acidity become more pronounced. We highlight that local plant–soil interactions should be the target of biodiversity conservation efforts and landscape management plans aimed at mitigating the impact of glacier extinction on biodiversity and ecological systems.
- Preprint
(1036 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-991', Anonymous Referee #1, 16 Apr 2024
GENERAL COMMENTS
The paper of Charles et al. is a case study documenting soil and vegetation development along one post-glacial chronosequence from 1864 to 2017 in a site in the Swiss Alps. By measuring and estimating plant community and soil parameters they aim to infer the effects of time since glacial retreat at the scale of the little ice age on the development of novel ecosystems. They intend to link the development of plants and the development of soil along the post-glacial chronosequence across correlations.
After reading the ms. I have tried to structure my comments into five bullets. I am aware that my comments may be seen as negative, but I have tried to write them as constructively as possible, with the idea of contributing to improve the document. First, I would like to say that that working on this type of ecosystem (new proglacial margins) with the aim of understanding the future of these ecosystems seems to me to be both topical and relevant.
I feel that a strong effort has been made to provide solid physical soil data that can improve our knowledge in this field. That said, I have identified a number of issues that seem to me to be problematic for publishing this work. In particular, (1) a review of the state of the art and a contextualisation of the results that are largely inadequate compared with the existing literature on the subject, (2) what appears to be an over-generalisation/overselling of the results even though the study was conducted on a single study site, (3) major doubts about the representativeness of the results in view of the sampling design that was put in place with very few repetitions, (4) the structure of the document could be greatly improved, in particular by strictly following the initial hypotheses but also by using a precise vocabulary with clear definitions from the outset, and (5) insufficient discussion of the potential mechanisms behind the observations made. I detail these points below:
(1) Insufficient state-of-art of the existing literature. In my opinion, a major weakness in the current document is that the state of the art (and mention to it in the discussion) is largely incomplete: many references are lacking, which are directly linked with the research question and even can answer it. The review by Cauvy Fraunié et al. is nicely cited, although without much details, but many papers have been published since, some with a double focus on plants and soils after glacial retreat: they do not appear in the document. Many others have not been taken into account in this review. This is not the object of this comment to list all of these papers but not mentioning them is concerning given the purpose of the document. See, for example: Caccianiga et al. 2006; Cannone et al. 2008 ; Heckmann et al. 2016; Zimmer et al. 2018; Szymanski et al. 2019; Descombes et al. 2020 ; Llambi et al. 2021; Anthelme et al. 2022; Mainetti et al. 2021; Erschbamer et al. 2023 ; Zimmer et al. 2023. Some of these papers are clearly focused on soil development after glacial retreat, (see, e.g. the papers in the journal CATENA), the majority deal with plan colonization, others provide multivariate analyses, just as you do, including plant and soil parameters. Some of them use several sites, not just one, to test hypotheses. Finally, three papers dealing with eDNA provide a new, global insight on the biodiversity, including soil biodiversity, of proglacial areas (Cantera et al. 2024; Carteron et al. 2024; Guerrieri et al. 2024). As this document doesn’t study the biotic compoentn of the soils, which is crucial to understand soil plant interactions, this would be relevant to scrutinize these papers in order to identify current knowledge gaps.
I repeat that this list is not exhaustive but the relevance of the document strongly depends on its ability to define the current limits of knowledge in the field, which has not been properly done, in my opinion. Instead, the overuse of self-citation based on a single site (e.g., Losapio et al. 2021 cited nine times) impedes the results to be confronted to/enriched by the other works on a larger scale. For these reasons, in its current version, this does not give the document the relevance it is supposed to provide.
(2) Overgeneralisation/overselling of the results. The paper seems to have a general/global purpose whereas is it based on empirical data on one single site (chronosequence) in Switzerland. This is not mandatory to work on several sites, especially because the type of data you provide is original (but see bullet one). But the generalisation of the results that is palpable in the abstract, introduction and discussion is incorrect. For example, the plant diversity decrease that is described in the results if far from being a generality at the temporal scale of the LIA (see articles above) and seems to be dependent on the local parameters that rule the single site studied (or could be due to a potential methodological weakness, see below). Is seems to me crucial that the authors are clear about the fact that their results are applicable locally, but not generalisable at larger scales. And, of course, they should be more confronted to the existing literature (bullet one).
(3) poor sampling design. As written above, I am admirative of the work that has been done with the soil analyses, even though it is not possible to deal with plant-soil interactions. But I was surprised to read that only four plots (repetitions) have been provided in each of the four deglacierization stages of the sampling design (total: 16 plots): I have never heard of the evaluation of a post-glacial chronosequence with so few plots. This seems to go against the basic statistical rules for finding significant differences between courses and I wonder how it is possible to find such strong differences in Figures 3 and 4. Another issue with this unusually low number of repetitions is the notion of representativeness of each stage. Bayle et al (2023) emphasize the fact that local heterogeneity strongly influences plant distribution shortly after glacial retreat. I cannot imagine that with just four plots of 3*3 m it is possible to encompass the heterogeneity and, therefore, the plant diversity of a whole stage. I recommend the author to enrich their sampling, e.g., multiplying*3 the number of plots in each stage, in order to increase the representativeness of the results.
(4) Structure improvable, terms to be defined and homogenized. It is appreciable to find precise research questions and testable hypotheses at the end of the introduction. But the two hypotheses (plant-diversity SOM relationships and soil acidification) are not introduced before so it is not possible to evaluate their significance/relevance. Additionally (redundant comment) they do not lie on existing literature which has already dealt with the topics. The discussion is too descriptive and therefore, does not address very clearly the research questions. The discussion is divided into three bullets: vegetation development, soil development and plant-soil interactions. This structure does not allow the reader to come back to the two hypotheses easily. Most of all, it gives similar importance to each of the three sections. In my opinion, the second section on soil is the most interesting/relevant whereas the first section is a combination of evidence, overgeneralisation (given the fact that results are extracted form one single site) and lack of perspective with existing literature (see specific comments later). The third section looks mostly speculative and partially redundant with the second section: this may be due to the fact that there are no results that truly reveal the nature of the interactions between plants and soils, but rather many correlations, especially shown with the PCA. These are two different things. This might be acceptable to list and discuss correlations between plant and soil variables the authors must be clear that they are not highlighting any interactions between these two variables (see title).
The other issue with the structure of the documents come with the different terms used. Several terms are used to describe the same thing, other terms are not defined and other seem to be incorrectly used. I list here some of these terms with suggestions but I feel it very important that they can be improved and correctly used all along the document in order to improve its structure. (a) in the whole document, the terms deglaciation, glacial retreat and glacial extinction are, seemingly, randomly used to described the same phenomenon. these are different terms with different consequences. Please use the term deglacierization when the glacier has not disappeared and the term deglaciation when it has disappeared (see numerous references in glaciology papers). Most importantly, it is important to separate the effects of deglacierization from the effects of deglaciation on plants and soils. (b) The term “post-glacial chronosequence” is not used. This is confusing because the whole study relies upon this space-for-time substitution approach, which has been extensively described in Matthews 1992 and dozens of references since (which, in their majority are not cited in the document, see bullet one). I am convinced that using this term is a prerequisite to improve the structure of the paper and its comparability with existing works. (c) Plant functional groups” are in fact life forms (or Raunkier biological types: shrubs, trees, forbs, etc.): this is not a fully functional approach and should be discussed. If not, this brings confusion to the readers who expect other types of plants traits to be measured/estimated. (d) the two terms abundance and relative cover of plants should be homogenized (e.g., L121). “New vs novel ecosystem: homogenize. (e) the term “novel” (ecosystem) is used in the title but abandoned later (“new” ecosystems). If used, it should be properly introduced, defined and, once again, put in perspective with the studies analysing novel plant communities and novel ecosystems after glacial retreat. See e.g., Starzomski 2013 and several references listed above. (f) last but not least, the concept of soil is not properly defined and should be clarified since the beginning. I most cases, it is the combination of mineral fragmentation and organic decomposition. It is generally accepted that soils are absent after a major disturbance leading to primary succession (see, e.g., Walker and delMoral 2003, but see protosoils of Caccianiga et al. 2006 and, more generally, the fact that organic matter is present because of galcial depositories; Khedim et al. 2021). The debate is open, still, because organic matter can be delivered by the glacier itself (see, e.g., Khedim et al. 2021; Zimmer et al. 2023). The document does not give a clue about these important objects/approaches/definitions. Instead, the methods describe soil collection the same way it would be done in mature ecosystems, i.e., removing the organic litter and collect soil at 10 cm belowground. But how is it possible to collect soil 10 cm belowground just after glacial retreat when there is NO soil (0.3% of SOC in the first stage at measured in the document) and, often, only stones and sand?).
(5) Mechanisms not studied/cited. I was surprised not to read anywhere the role played by dispersal mechanisms (Alexander et al. 2018; Zimmer et al. 2018; Erschbamer et al. 2023) and tree cover on the patterns of plant diversity found. The first part of the discussion is focused on the abilities of plant to be “pioneer” (but what does it means?) and adapted to perturbations. But dispersal capacity is the first condition to migrate upward and should be discussed here. It explains, for example, that some plants, which are not stress-tolerant, can grow first close to the glaciers and then die, not only because they are dependent on the presence of glaciers (as written in the document) but because they are just better to be the first on the site. In the older stages, I was surprised, too, not to read anything about the effects of tree cover on plant diversity: this is very specific of the site studied and it could act as a strong competitor effect and explain in part the possible reduction of plant diversity (e.g., Garbarino et al. 2010; but see also my comments above about the methods). Also, the brief mention to facilitation processes in the discussion would deserve more attention, especially the fact that there is a facilitation deficit in the pioneer stages (Erschbamer et al. 2023). I think the paper can be strongly improved by mentioning/discussing these (and others) potential mechanisms that are at work in these environments
SPECIFIC COMMENTS (chronological order)
TITLE: Not acceptable in this form since the document doesn’t deal with plant soil interactions, it deals instead with the combined characterization of soil and plant parameters. Identifying interactions would have been reachable with experimental data or with fine observational protocol comparing, e.g., soil parameters et different distances of a plant individual. Use the term “associations”? Please consider removing or not the term “novel”, which is currently not much used in the document.
ABSTRACT
-Would be interesting to cite the research questions or hypotheses, otherwise the document seems purely descriptive.
-L20: “plant functional types” is not correct: life forms?
-L24-25: this is too much generalisation: this sentence is true for your site only, not globally. It should be specified here.
INTRODUCTION
-L48 “such novel interactions”: this term is not connected with the previous sentence, why novel interactions? Lacks explanation + the definition of the term novel (see general comment and specific comment in the title).
-L49 “soil”: no, we do not expect soil to be present just after glacial retreat (see general comment)
-L52 “gain of soil habitat”: needs rephrasing
-L56 “plant species, which are adapted to the particular conditions of glacier margins”: I am not convinced here. You cite Cauvy-Fraunié & Dangles 2019 which are stream ecologists, ok this applies for aquatic invertebrates but not necessarily to plants (I have never heard of plant communities restricted to glacial habitats).
-L58 “little is known about how soil development affects plant biodiversity”: while this might be true you do not cite the existing literature that deals with it (see general comments).
-L60 “The loss of glaciers implies a change in the development of soils” this formulation is not correct: the loss of glacier implies more than a change; this is a decisive step from no soil to the presence and the development of soil (this comment applies to the whole document).
-L62-64: this sentence is pure evidence; I do not see the point in writing it here.
-L65 and whole document “deglaciation”: see general comments on terms: deglaciation describe the total disappearance of the glacier (different from deglacierization)
-L67-68: this sentence is true but there's a problem of causality (avoid the term “thus”?)
-L71-72, several papers have not been taken into account here (and whole document, see general comments).
-Research questions and hypotheses: RQ1 and RQ2: not generalisable, only applicable at the scale of your site. The two hypotheses are not introduced before, need to be introduced within the state of art (current limits of knowledge).
METHODS
-Why not mentioning that you work with post-glacial chronosequences (see general comments)?
-L94 “LIA moraines”: you need to write in full and explain what it is.
L94-95 “This transect is independent of elevation gradient, which ranges between 1890 and 1990 m a.s.l”: How can it be independent? They are crossed. Do you mean that the elevation gradient is very shallow and you expect it not to have a significant influence on the temporal gradient?
-L105 “For each stage, we randomly selected four plots of 3x3 m”: area seems good but cannot be selected without justification, please justify, using the existing literature.
L108-109 “Plots were resurveyed every two weeks for completeness until the end of August”: nice effort! Might be relevant to provide raw data as an appendix to see the usefulness of this method.
-L110 “the surface cover” do you mean “the relative cover at soil surface (of each species)”?
-L110 “assessed visually”, vocabulary: “estimated visually” (please provide details: did you use a grid? Did you spend the same time estimating each plot?)
-L113, similarly “determined” (direct measurement) is not correct, it is an estimation.
-L115 “average reaction index”: Good point: Landolt values are powerful and not sufficiently used, in my opinion”
-L119 “plant functional groups”: this is not a fully functional approach, rather a proxy, which is interesting of course but the readers expect other types of traits when referring to this term. this has been done in other studies and should be discussed here and in the discussion.
-L121 replace “abundance” by “relative cover (homogenization of terms see general comments)
-L123 “A soil sample was taken from the top 10 cm of soil”: but then what do you do when there is no soil (most probable situation)? See general comments
-L124 similarly, where is the “organic layer” just after glacial retreat? Same with “the remaining soil samples represent organic horizons” L126
RESULTS
-Use past tense in the result section.
-L206: “Indeed, the first dimension includes vegetation variables”: no causality: all these variables could be influenced by geomorphological processes as well; no? (axis 2)”
-L212 “highly correlated with a positive correlation”: rephrasing needed.
-Fig. 2 PCA: The relevance of a multivariate analysis, especially a PCA, depends on the number of variables that are included: more variables = more correlation between variables. the « game to play » is thus to explain a large % a variation with the lowest number of variables. Here, the high number of variables makes that this goal may be not reached. Why not reducing the number of variables?
L227, L352, L385, L396 and whole document “glacier retreat influences (plant diversity)”: not correct, time since glacier retreats influences plant diversity.
L237 “Statistics confirm that glacier retreat influences proportion of forbs”: how possible with so few repetitions (4)?
DISCUSSION
-Vegetation development: I find it very interesting to go into the strategies of species but we are soon frustrated because no details are given: why these species L287-288 are considered pioneer species, what is a pioneer species: improvements are clearly at hand here! Don’t forget that most of the first colonizers are here because of their high dispersal capacities instead of their pioneer, stress-tolerant traits (see general comments)
L294-295 “Losapio et al. (2021), who note that plant diversity initially increases with glacier retreat before decreasing in the late stage” be careful with the overuse of self-citations (see general comments). Many other references describe a monotonous increase in plant diversity at the same temporal scale, beginning with Matthews 1992. A poor state of art severely reduces the scope of the document currently. (see general comments).
-L298-302: this section about plant-plant facilitation should lie on solid existing literature (see general comments), especially the lack of facilitators at the beginning of the succession should be introduced/discussed.
-Section vegetation development in general: I feel that this section is much less relevant than the next section, as if primary succession was reinvented 100 years later but using one single site. I recommend a focus on the next section (soils and associations with plants instead)
-Section soil development: interesting but need more connexions with the existing literature (see general comments).
-L377_378 “the increase of organic carbon, nitrogen, and available elements positively influences the development of certain plant functional groups such as dwarf-shrubs and trees”: Correlation is not causality, you do not provide proofs that one variable has an effect on another.
-L385_386 “Glacier retreat decreases soil pH, which in turn increases soil organic carbon”: same comment, correlation is not causality, you do not provide proofs that one variable has an effect on another
-L386-387 “This model shows that when plant diversity increases, soil organic carbon tends to decrease and that high levels of soil carbon are associated with low plant diversity”. I do not agree. Once again, no causality has been demonstrated here, a great deal of caution is needed in interpreting the facts.
-L387-388 “This can be explained by the fact that, with increasing glacier retreat in stage 4, there is an accumulation of organic matter and plant diversity is the lowest” (needs rephrasing): problem of sampling (few repetitions) or tree cover (light competition)? In any case this is not generalisable (see general comments)
CONCLUSION deals with too much generalization given that the data has been collected in one single site, with very few repetitions (total: 16 plots). Sentences in this section are highly speculative and/or refer to the evidence (see below), I recommend trying not to oversell the results of the document and focus on its true contribution (which means comparing it more with the existing literature).
-L406-407 “These results suggest that novel glacier ecosystems may function as carbon sink by storing carbon in the soil” this type of sentence doesn’t bring anything new: this is evidence (after glacier retreat carbon is nearly zero), has been studied much more extensively in other studies (e.g., Khedim et al. 2021) and may be irrelevant considering the small amount of carbon stored.
-L407 “They also show that nitrogen is a limiting factor”: similarly, this is evidence and, more generally, this is one of the main constraints for plants in alpine regions (Körner 2003, 2021, reference that has not even been cited).
Figure 2 looks rather like a first draft directly extracted from R, which needs to be edited. Given the doubts raised about the statistical significance of the results (low repetitions, lore detailed presentation of the statistical tests is required, including tables in appendices.
Citation: https://doi.org/10.5194/egusphere-2024-991-RC1 -
AC1: 'Reply on RC1', Gianalberto Losapio, 23 May 2024
General Response: We are grateful for your work with editing our manuscript. We have followed the helpful recommendations you provided and addressed all their concerns, as explained below.
Response (1): Thank you for suggesting additional literature. We now add those references. In doing so, we introduce the results while incorporating this additional knowledge to our study case and discussing our results in light of those previous studies.
Response (2): The Reviewer is right that with “only” one site there are not many possibilities for making broader generalization. We now pay attention not to generalize too much. We also make it clear that, as always in ecology, there can be nuances and context-dependencies. We now put the results of this manuscript in the context of that suggested literature by making a narrative comparison in the discussion section. In doing so, we pay attention to stay within the limits provided by our case study.
Response (3): It is true that environmental heterogeneity is always a challenge in biogeosciences. We now provide evidence that enough variation and heterogeneity was covered by random sampling as demonstrated by the rarefaction analysis, which is now provided in the SI. Yet, we now do acknowledge this potential limitation in the methods section.
Response (4): Thanks for your suggestions, which are substantially improving the narrative and readability of our manuscript. We are following their suggestion by streamlining the different paragraph. We also pay attention to use consistent terminology.
Response (5): Soil has been defined as “the upper layer of earth in which plants grow, which consists of a mixture of sediments, organic matter, and rock (mineral) particles.
The terms relating to glaciers have been unified as follows: « glacier retreat » to designate the process where glaciers melt and retreat; “deglacierization” to refer to the process where the glacier melts without having disappeared; “glacier extinction” to refer to the state predicted with current climate change where there will no longer be a glacier.
Plant functional groups has been replaced by plant life-forms.
Abundance has been replaced by relative cover.
The term novel has been defined and properly used
We specified that litter, such as dead leaves or branches, was removed from the upper horizon.
Response (6): Thank for this suggestion. Previously, we did not refer to mechanisms as this case study was of comparative nature. Yet, it is true that our path analysis can provide a robust clue inside potential mechanisms. We now refer to and discuss potential mechanisms underlying the observed patterns.
R: Thanks for your suggestion. We changed the title to “Glacier retreat shapes plant–soil relations over spacetime”
R: research questions added in the abstract
R: changed
R: added
R: term 'novel' defined
R: rephrased
R: citation corrected
R: literature added
R: modified
R: removed
R: corrected
R: corrected
R: added and completed
R: fixed
R: added
R: Yes that was the purpose of this sentence. This has been corrected to clarify.
R: We clarified the random selection
R: specified and added
R: terminology corrected
R: in this sentence, determine has been replaced by evaluate
R: corrected
R: changed
R: rephrased
R: corrected
R: references and state of art have been improved with new references
R: the discussion has been substantially revised
Citation: https://doi.org/10.5194/egusphere-2024-991-AC1
-
AC1: 'Reply on RC1', Gianalberto Losapio, 23 May 2024
-
RC2: 'Comment on egusphere-2024-991', Anonymous Referee #2, 27 Apr 2024
Dear Charles et al.,
Your manuscript “Plant–soil interactions underline the development of novel ecosystems after glacier retreat” handles a highly relevant subject in the eye of the current global crisis of climate change and biodiversity loss. The imminent loss of (the largest part of) all glaciers in Central Europe by 2100 is a sad prospect, and the study of the ecosystems that develop at the retreating glacier front will help us protecting the newly developing communities.
The largest downside of this study is the lack of replication. The retreat of only one glacier is studied, while differences in (micro)climate, bedrock, altitude, the set of plant species (and other taxa) available for invasion, etc. could have a large influence on the outcome of the succession. Figure 2 also indicates that all plots were located rather low in the valley (or high on the ridge (?)) of the deglaciated area. This made me wonder whether the outcome could even be different if the plots had been higher (or lower) in altitude for this specific case. Is the final successional outcome always a species poor forest, or do higher ridges only reach stage 2 or 3 without continuing to stage 4? (Or does the fact that the glacier retreats on a certain spot implies that the final stable plant community in the succession is a species poor forest?)
As sampling a sequence close to a retreating glacier front is not trivial, and as the documentation of the age of the deglaciation along the deglaciated might be challenging to acquire (?), I do understand that this study was limited to one glacier only.
However, it would be a great asset for follow up studies to include more retreating glaciers, preferentially distributed over a large climatic and altitudinal range.Your study strongly focusses on species diversity and richness as the factor to optimise after deglaciation. This is a valid choice, especially in mountainous areas, where climate change pushes cold adapted species upwards until they have reached the summit and are outcompeted by more generalist species. However, with the broad range of parameters that is measured in this study, other factors could be selected to focus on. One of them (especially in the context of climate change) is carbon storage. Carbon storage is (not unexpected) the highest in Stage 4 (the species poor forest). You do mention this briefly in the conclusion (line 405-407), but it would be good to highlight this more clearly also in the abstract and maybe also in the discussion session.
Species diversity and richness are optimal (highest) in the intermediate successional stages (Stage 2 and 3). You urge several times to use this study as a base for conservation efforts for species protection and preservation. However, no suggestions for conservation methods and aims are given. I think that it would strongly strengthen your story if you could (even briefly) focus on this aspect.
Finally, your manuscript is well structured, and the storyline is easy to follow and interpret. In general, the language is fluent and well written. You can find some grammatical and spelling corrections (and suggestions) in the technical corrections section.
Specific comments
Lines 21-23:
“Plant diversity initially increases after glacier retreat, along with an increase in soil organic matter and nutrients. Over 120 years, soils acidify at a rate of 0.02 units per year, the C/N ratio increases while plant diversity decreases.”
Do I interpret it correctly that you found an initial increase of plant diversity, but an overall decrease (compared to the maximum that is reached at a certain point?) over the first 120 years? If so, when does the plant diversity reach its maximum and starts to decrease again after the initial increase? I would rephrase this sentence to make it easier to interpret.
Line 26-28:
Do I understand it correctly that, after glacier retreat, the stable state of the plant community is a species poor forest (which is, however, probably still more species rich than a glacier ecosystem but containing less rare / threatened species?). You mention that targeted biodiversity conservation efforts should be made to move the ecosystems back to their initial, more species rich states. Could you elaborate on the types of conservation efforts would be effective in this specific case (e.g. increased disturbance by X / mowing / …)?
Lines 78-79:
“We also expect that an increase in plant diversity will lead to an increase in soil organic matter,”
Why is it specifically the increase in plant diversity that would lead to an increase in soil organic matter? Would this also not happen with the colonization of a species poor plant community?
Line 106:
“we randomly selected four plots of 3x3 m”
How was the random selection performed?
All the plots seem to be quite high on the ridge (or low in the valley?). Do you think that the results have been different if the plots were located at lower (or higher?) altitudes?Lines 125-126
“Litter and the first few centimetres of soil composed of slightly degraded organic debris were discarded, the remaining soil samples represent organic horizons.”
Does this procedure only apply to the two plots in stage 4 that had a very thick organic layer?
Otherwise, I´m a bit confused. In line 124, there is stated that the organic layer was removed.Line 141
At what temperatures where the soil samples oven-dried before C and N analyses?
Line 295
What is the starting stage from which the plant diversity increases after glacial retreat? I did assume that there were close to no plants growing on a glacier, but maybe I am wrong? Except from mosses, are there vascular plants that occur on glaciers?
This question occurred to me several times while reading your paper. Maybe mention this briefly when describing the successional sequence? You could call it ´stage 0´?Line 386
“This model shows that when plant diversity increases, soil organic carbon tends to decrease […]”
I find this a strange statement. This is true if one goes from stage 4 to stage 3 or 2 (going back in the successional sequence), but not if one goes from stage 1 to 2 or 3, which is the natural course of events. Going back from stage 4 to 3 or 2 requires a disturbance, which doesn´t even necessarily reduce the soil organic carbon.
I do agree with the second part of the sentence (high levels of soil organic carbon are associated with low plant diversity).Technical corrections
Line 36:
Global warming is currently increasing by …
--> Global (average) temperatures are currently increasing byLine 37:
Thus by 2030 and 2050, temperatures are expected to increase further by 1.5 °C (IPCC, 2022).
Should there not be two temperatures, indicating the expected temperature increase by 2030 and by 2050?
Line 95:
“m a.s.l.”--> “m.a.s.l.”
Line 102:
When listing the different stages of ecosystem development, they are called “pioneer stage”, “early stage”, “intermediate stage” and “late successional stage”.
However, just after that, they are called “stage 1-4”.
I would already in the list add which developmental stage corresponds to which stage number:- Pioneer stage: 1989–2022 AD (Stage 1)
- Early stage: 1925–1989 AD (Stage 2)
- Intermediate stage: 1900–1925 AD (Stage 3)
- Late successional stage: 1864–1900 AD (Stage 4)
Line 103:
The four stages are all about 250 m wide.
Is 250 m long a better description? I interpret 250 m wide the with perpendicular on the length of the glacier, while long corresponds (in my perspective) on the length of the stage parallel to the length of the glacier.
Line 106:
“Hereafter, these plots are named “a” to “d”.”
Here, small letters are used to refer to the replicates, while in Figure one, capitals are used. I would choose one of both and use it consistently.
Line 125:
“centimeters”--> centimetres
Line 128:
“For each plot, at least 200 g of soil were taken back to the lab.”
--> For each plot, at least 200 g of soil was taken back to the lab.Line 133:
“We conducted soil analyses including measurement […]”
--> We conducted soil analyses, including measurements […]
Table 1:
“Plant reaction value – Community weight man […]”
--> Plant reaction value – Community weight mean […]
Table 1:
“SOC – Soil organic carbon content (SOC)”
--> SOC – Soil organic carbon content
Table 1:
I would add units to the description where applicable.
Lines 159-160:
“As exploratory analysis, Pearson correlation values were calculated among all dependent variables, making it possible to obtain 160 a correlation matrix and observe linear relationships between pairs of variables (results reported in SI).”
I don´t find this back in the Supplementary Material.
Line 188:
“[…] predictors are qualitatively similar […]”
--> […] predictors were qualitatively similar […]
Figure 2 (Line 220):
The contribution of an observation to a component is visualized by the darkness of the arrow in Figure 2. However, I do not see how the legend to the right (contrib: 1 / 2 / 3 / 4) explains this? I would remove this legend or make it more intuitive.
Lines 233-235:
“For each stage, the median reaction value drops from 3.2 to 2.9, 2.4 and finally 1.9 from stages 1 to 4, respectively. Our stages data therefore goes from lightly acidic to neutral, to acidic and finally to very acidic.”
à Did you mean lightly alkaline?
Figure 3.c:
Legend: “subshrubs”--> dwarf shrubs
Figure 4:
Text:
- Also here I would add the comment that “boxplots have been jittered for facilitating data visualization; the actual years since deglaciation is the written one.”
- (a) Soil organic carbon content (SOC)
- (b) Soil total nitrogen content (Ntot)
- (c) Organic carbon / total nitrogen ratio (C/N ratio)
- Y scale à The y-axis is log distributed …
Line 259:
“[…] around c 2.4% […]”
--> […] around ca. 2.4% […]
Lines 266-270:
Here, the C/N ratio is expressed in %. The C/N ratio, however, does not have a unit and is usually expressed as a ratio of the number of units of carbon to the number of units of nitrogen. I would change the text in this way (e.g. Stage 1 has a C/N ratio of 2:1).
Line 271:
It is not necessary to indicate the age where stage 1 corresponds to, as this is not done elsewhere or for other stages.
Line 331:
“Late development stage shows particularly low pH […]”
--> The late developmental stage shows particularly low pH […]
Lines 354-355:
“The increase in plant cover and diversity characterizing by […]”
--> The increase in plant cover and diversity characterizing by […]
The increased transfer of organic matter to the soil is not a consequence of increased plant diversity, but purely of increased plant cover.
Line 362:
“[…] total nitrogen are often below 0.1 % […]”
--> […] total nitrogen is often below 0.1 % […]
(Or better: The soil total nitrogen content is often below …)
Lines 364-365:
“The presence of microorganisms involved in the nitrogen cycle including microorganisms associated to nitrogen-fixing plants can explain […]”
--> The presence of microorganisms involved in the nitrogen cycle, including microorganisms associated to nitrogen-fixing plants, can explain […]
Line 365:
“Early and intermediate development stages […]”
--> Early and intermediate developmental stages […]
Or if you mean it specifically for your study (and not for early and intermediate developmental stages in general):
--> The early and intermediate developmental stages […]
Lines 368-370:
“In stages 2 and 3, SOC and total nitrogen content are similar corresponding to similar vegetation indices in these stages. On the contrary, plants in Ericaceaea family Rhododendron ferrugineum, Vaccinium myrtillus, Calluna vulgaris) dominate late development stages.”
I (think I) understood this sentence after reading it a few times (that the SOC and Ntot are similar in stage 2 and 3 and that also their vegetation composition was very similar), but I would rewrite the sentence to make it easier to understand.
Line 371:
“[…] matter that can seen in the increasing […]”
--> […] matter that can be seen in the increasing […]
Lines 376-378:
“While low concentrations of available elements in the early stages limit plant growth and establishment (Chapin et al., 1994), the increase of organic carbon, nitrogen, and available elements positively influences the development of certain plant functional groups such as dwarf-shrubs and trees.”
--> While low concentrations of available elements in the early stages limit plant growth and establishment (Chapin et al., 1994), the increase of organic carbon, nitrogen, and available elements in the later stages positively influences the development of certain plant functional groups such as dwarf-shrubs and trees.
Line 381-383:
“Furthermore, the presence of bacteria and fungi associated to pioneer plants is decisive for the availability of certain elements, such as phosphorus and nitrogen, for shrubs and trees (Egli et al., 2012).”
“Furthermore, the presence of bacteria and fungi associated to pioneer plants is decisive for the availability of certain elements, such as phosphorus and nitrogen, which are essential for species that are associated to later stages (such as shrubs and trees) to establish (Egli et al., 2012).”
Line 384:
“The path analysis model provides an […]”
--> The path analysis model provided an […]
For the whole paragraph: In past tense as the model is built on own results?
Line 399:
“Pioneer plant species enables diversity […]”
--> Pioneer plant species enable diversity […]
Line 400:
“[…] soil conditions, species become increasingly dominant […]”
--> […] soil conditions, certain (or competitive?) species become increasingly dominant […]
Line 405:
“[…] although both soil carbon and nutrient increase […]”
--> […] although both soil carbon and nutrients increase […]
Line 409:
“[…] mitigation actions shall consider […]”
--> […] mitigation actions should consider […]
Supplementary material, Figure A1(b)
- The axis labels are not readable.
- It is not clear where Stage 1, 2, 3 and 4 are located in the soil texture triangle classification.
Supplementary material, Figure A1 - Legend:
“Additional soil and plants variables […]”
--> Additional soil and plant variables […]
Supplementary material, Figure A1 - Legend:
“[…] including plants available elements concentrations […]”
--> […] including plant available elements concentrations […]
Supplementary material, Figure A1 - Legend:
“Total element proportions (c)”
--> Total element proportions (c)Also, the header of Figure A1.c refers to “Major elements”, while the legend refers to “Total elements”. Please make this conform.
Citation: https://doi.org/10.5194/egusphere-2024-991-RC2 -
AC2: 'Reply on RC2', Gianalberto Losapio, 23 May 2024
Response: We are grateful for your work with editing our manuscript. We have followed the helpful recommendations you provided and addressed all their concerns, as explained below.
Response: The Reviewer is right that with “only” one site there are not many possibilities for making broader generalization. We now pay attention not to generalize too much. We also make it clear that, as always in ecology, there can be nuances and context-dependencies. We now put the results of this manuscript in the context of that suggested literature by making a narrative comparison in the discussion section. In doing so, we pay attention to stay within the limits provided by our case study.
Response: Thank you for your suggestion. This point has been removed in the manuscript as suggested by other referees who refer to the article by Khedim et al. 2021 and consider that this is evidence and may be irrelevant considering the small amount of carbon stored.
Response: those points have been added in the discussion (at the end of 4.1)
R: Yes, the sentence has been reformulated and clarifications added
R: some suggestions on conservation efforts have been added such as removal of encroaching woody vegetation
R: That’s right, an increase in plant diversity doesn’t mean an increase in soil organic matter. This was not worded correctly and has been corrected consequently. Here we expected that an increase in plant colonization will lead to an increase in soil organic matter. We suppose that the increase in plant colonization will be accompanied by an increase in plant diversity.
R: we specified that the foreland was examined, avoiding the lateral moraines (which are highly unstable and pose a risk to researchers)
R: Yes, the explanation has been clarified. We removed the litter horizon. The remaining soil samples represent the organo-mineral horizons. But as it was not possible to remove the organic horizon completely for two plot in stage 4, litter and the first centimetres of soil were discarded.
R: precision added
R: in some glaciers such as debris-covered glaciers, there are vascular plants occuring on the glacier debris (https://link.springer.com/article/10.1007/s11829-016-9444-x). We have no evidence that the same occurs here though. We improve the description of the vegetation at the study site.
R: we see the misunderstanding; that sentence has been rewritten
R: changed
R: corrected, after checking the source, temperatures are expected to rise by 1.5°C between 2030 and 2052
R: changed
R: Yes, fixed
R: Yes, fixed
R: that's right, everything has been homogenized in capital letters
R: changed
R: changed
R: changed
R: changed
R: changed
R: added
R: added
R: changed
R: this legend has been removed
R: We understand that this is not clear, it's from neutral/lightly acidic, to acidic and finally to very acidic. This sentence has therefore been reworded
R: corrected
R: added and modified
R: modified
R: changed
R: removed
R: modified
R: Thank you for your comment, this has been corrected
R: modified
R: commas added
R: modified by the second suggestion
R: modified as: “In stages 2 and 3, vegetation composition is similar, as are SOC and total nitrogen. These stages are different from the late developmental stage where plants from the Ericaceaea family (Rhododendron ferrugineum, Vaccinium myrtillus, and Calluna vulgaris) dominate stage 4.”
R: corrected
R: added
R: added
R: done
R: corrected
R: “certain” added
R: corrected
R: corrected
R: SI revised and updated
R: corrected
R : corrected, and made uniform with “major elements”
Citation: https://doi.org/10.5194/egusphere-2024-991-AC2
Status: closed
-
RC1: 'Comment on egusphere-2024-991', Anonymous Referee #1, 16 Apr 2024
GENERAL COMMENTS
The paper of Charles et al. is a case study documenting soil and vegetation development along one post-glacial chronosequence from 1864 to 2017 in a site in the Swiss Alps. By measuring and estimating plant community and soil parameters they aim to infer the effects of time since glacial retreat at the scale of the little ice age on the development of novel ecosystems. They intend to link the development of plants and the development of soil along the post-glacial chronosequence across correlations.
After reading the ms. I have tried to structure my comments into five bullets. I am aware that my comments may be seen as negative, but I have tried to write them as constructively as possible, with the idea of contributing to improve the document. First, I would like to say that that working on this type of ecosystem (new proglacial margins) with the aim of understanding the future of these ecosystems seems to me to be both topical and relevant.
I feel that a strong effort has been made to provide solid physical soil data that can improve our knowledge in this field. That said, I have identified a number of issues that seem to me to be problematic for publishing this work. In particular, (1) a review of the state of the art and a contextualisation of the results that are largely inadequate compared with the existing literature on the subject, (2) what appears to be an over-generalisation/overselling of the results even though the study was conducted on a single study site, (3) major doubts about the representativeness of the results in view of the sampling design that was put in place with very few repetitions, (4) the structure of the document could be greatly improved, in particular by strictly following the initial hypotheses but also by using a precise vocabulary with clear definitions from the outset, and (5) insufficient discussion of the potential mechanisms behind the observations made. I detail these points below:
(1) Insufficient state-of-art of the existing literature. In my opinion, a major weakness in the current document is that the state of the art (and mention to it in the discussion) is largely incomplete: many references are lacking, which are directly linked with the research question and even can answer it. The review by Cauvy Fraunié et al. is nicely cited, although without much details, but many papers have been published since, some with a double focus on plants and soils after glacial retreat: they do not appear in the document. Many others have not been taken into account in this review. This is not the object of this comment to list all of these papers but not mentioning them is concerning given the purpose of the document. See, for example: Caccianiga et al. 2006; Cannone et al. 2008 ; Heckmann et al. 2016; Zimmer et al. 2018; Szymanski et al. 2019; Descombes et al. 2020 ; Llambi et al. 2021; Anthelme et al. 2022; Mainetti et al. 2021; Erschbamer et al. 2023 ; Zimmer et al. 2023. Some of these papers are clearly focused on soil development after glacial retreat, (see, e.g. the papers in the journal CATENA), the majority deal with plan colonization, others provide multivariate analyses, just as you do, including plant and soil parameters. Some of them use several sites, not just one, to test hypotheses. Finally, three papers dealing with eDNA provide a new, global insight on the biodiversity, including soil biodiversity, of proglacial areas (Cantera et al. 2024; Carteron et al. 2024; Guerrieri et al. 2024). As this document doesn’t study the biotic compoentn of the soils, which is crucial to understand soil plant interactions, this would be relevant to scrutinize these papers in order to identify current knowledge gaps.
I repeat that this list is not exhaustive but the relevance of the document strongly depends on its ability to define the current limits of knowledge in the field, which has not been properly done, in my opinion. Instead, the overuse of self-citation based on a single site (e.g., Losapio et al. 2021 cited nine times) impedes the results to be confronted to/enriched by the other works on a larger scale. For these reasons, in its current version, this does not give the document the relevance it is supposed to provide.
(2) Overgeneralisation/overselling of the results. The paper seems to have a general/global purpose whereas is it based on empirical data on one single site (chronosequence) in Switzerland. This is not mandatory to work on several sites, especially because the type of data you provide is original (but see bullet one). But the generalisation of the results that is palpable in the abstract, introduction and discussion is incorrect. For example, the plant diversity decrease that is described in the results if far from being a generality at the temporal scale of the LIA (see articles above) and seems to be dependent on the local parameters that rule the single site studied (or could be due to a potential methodological weakness, see below). Is seems to me crucial that the authors are clear about the fact that their results are applicable locally, but not generalisable at larger scales. And, of course, they should be more confronted to the existing literature (bullet one).
(3) poor sampling design. As written above, I am admirative of the work that has been done with the soil analyses, even though it is not possible to deal with plant-soil interactions. But I was surprised to read that only four plots (repetitions) have been provided in each of the four deglacierization stages of the sampling design (total: 16 plots): I have never heard of the evaluation of a post-glacial chronosequence with so few plots. This seems to go against the basic statistical rules for finding significant differences between courses and I wonder how it is possible to find such strong differences in Figures 3 and 4. Another issue with this unusually low number of repetitions is the notion of representativeness of each stage. Bayle et al (2023) emphasize the fact that local heterogeneity strongly influences plant distribution shortly after glacial retreat. I cannot imagine that with just four plots of 3*3 m it is possible to encompass the heterogeneity and, therefore, the plant diversity of a whole stage. I recommend the author to enrich their sampling, e.g., multiplying*3 the number of plots in each stage, in order to increase the representativeness of the results.
(4) Structure improvable, terms to be defined and homogenized. It is appreciable to find precise research questions and testable hypotheses at the end of the introduction. But the two hypotheses (plant-diversity SOM relationships and soil acidification) are not introduced before so it is not possible to evaluate their significance/relevance. Additionally (redundant comment) they do not lie on existing literature which has already dealt with the topics. The discussion is too descriptive and therefore, does not address very clearly the research questions. The discussion is divided into three bullets: vegetation development, soil development and plant-soil interactions. This structure does not allow the reader to come back to the two hypotheses easily. Most of all, it gives similar importance to each of the three sections. In my opinion, the second section on soil is the most interesting/relevant whereas the first section is a combination of evidence, overgeneralisation (given the fact that results are extracted form one single site) and lack of perspective with existing literature (see specific comments later). The third section looks mostly speculative and partially redundant with the second section: this may be due to the fact that there are no results that truly reveal the nature of the interactions between plants and soils, but rather many correlations, especially shown with the PCA. These are two different things. This might be acceptable to list and discuss correlations between plant and soil variables the authors must be clear that they are not highlighting any interactions between these two variables (see title).
The other issue with the structure of the documents come with the different terms used. Several terms are used to describe the same thing, other terms are not defined and other seem to be incorrectly used. I list here some of these terms with suggestions but I feel it very important that they can be improved and correctly used all along the document in order to improve its structure. (a) in the whole document, the terms deglaciation, glacial retreat and glacial extinction are, seemingly, randomly used to described the same phenomenon. these are different terms with different consequences. Please use the term deglacierization when the glacier has not disappeared and the term deglaciation when it has disappeared (see numerous references in glaciology papers). Most importantly, it is important to separate the effects of deglacierization from the effects of deglaciation on plants and soils. (b) The term “post-glacial chronosequence” is not used. This is confusing because the whole study relies upon this space-for-time substitution approach, which has been extensively described in Matthews 1992 and dozens of references since (which, in their majority are not cited in the document, see bullet one). I am convinced that using this term is a prerequisite to improve the structure of the paper and its comparability with existing works. (c) Plant functional groups” are in fact life forms (or Raunkier biological types: shrubs, trees, forbs, etc.): this is not a fully functional approach and should be discussed. If not, this brings confusion to the readers who expect other types of plants traits to be measured/estimated. (d) the two terms abundance and relative cover of plants should be homogenized (e.g., L121). “New vs novel ecosystem: homogenize. (e) the term “novel” (ecosystem) is used in the title but abandoned later (“new” ecosystems). If used, it should be properly introduced, defined and, once again, put in perspective with the studies analysing novel plant communities and novel ecosystems after glacial retreat. See e.g., Starzomski 2013 and several references listed above. (f) last but not least, the concept of soil is not properly defined and should be clarified since the beginning. I most cases, it is the combination of mineral fragmentation and organic decomposition. It is generally accepted that soils are absent after a major disturbance leading to primary succession (see, e.g., Walker and delMoral 2003, but see protosoils of Caccianiga et al. 2006 and, more generally, the fact that organic matter is present because of galcial depositories; Khedim et al. 2021). The debate is open, still, because organic matter can be delivered by the glacier itself (see, e.g., Khedim et al. 2021; Zimmer et al. 2023). The document does not give a clue about these important objects/approaches/definitions. Instead, the methods describe soil collection the same way it would be done in mature ecosystems, i.e., removing the organic litter and collect soil at 10 cm belowground. But how is it possible to collect soil 10 cm belowground just after glacial retreat when there is NO soil (0.3% of SOC in the first stage at measured in the document) and, often, only stones and sand?).
(5) Mechanisms not studied/cited. I was surprised not to read anywhere the role played by dispersal mechanisms (Alexander et al. 2018; Zimmer et al. 2018; Erschbamer et al. 2023) and tree cover on the patterns of plant diversity found. The first part of the discussion is focused on the abilities of plant to be “pioneer” (but what does it means?) and adapted to perturbations. But dispersal capacity is the first condition to migrate upward and should be discussed here. It explains, for example, that some plants, which are not stress-tolerant, can grow first close to the glaciers and then die, not only because they are dependent on the presence of glaciers (as written in the document) but because they are just better to be the first on the site. In the older stages, I was surprised, too, not to read anything about the effects of tree cover on plant diversity: this is very specific of the site studied and it could act as a strong competitor effect and explain in part the possible reduction of plant diversity (e.g., Garbarino et al. 2010; but see also my comments above about the methods). Also, the brief mention to facilitation processes in the discussion would deserve more attention, especially the fact that there is a facilitation deficit in the pioneer stages (Erschbamer et al. 2023). I think the paper can be strongly improved by mentioning/discussing these (and others) potential mechanisms that are at work in these environments
SPECIFIC COMMENTS (chronological order)
TITLE: Not acceptable in this form since the document doesn’t deal with plant soil interactions, it deals instead with the combined characterization of soil and plant parameters. Identifying interactions would have been reachable with experimental data or with fine observational protocol comparing, e.g., soil parameters et different distances of a plant individual. Use the term “associations”? Please consider removing or not the term “novel”, which is currently not much used in the document.
ABSTRACT
-Would be interesting to cite the research questions or hypotheses, otherwise the document seems purely descriptive.
-L20: “plant functional types” is not correct: life forms?
-L24-25: this is too much generalisation: this sentence is true for your site only, not globally. It should be specified here.
INTRODUCTION
-L48 “such novel interactions”: this term is not connected with the previous sentence, why novel interactions? Lacks explanation + the definition of the term novel (see general comment and specific comment in the title).
-L49 “soil”: no, we do not expect soil to be present just after glacial retreat (see general comment)
-L52 “gain of soil habitat”: needs rephrasing
-L56 “plant species, which are adapted to the particular conditions of glacier margins”: I am not convinced here. You cite Cauvy-Fraunié & Dangles 2019 which are stream ecologists, ok this applies for aquatic invertebrates but not necessarily to plants (I have never heard of plant communities restricted to glacial habitats).
-L58 “little is known about how soil development affects plant biodiversity”: while this might be true you do not cite the existing literature that deals with it (see general comments).
-L60 “The loss of glaciers implies a change in the development of soils” this formulation is not correct: the loss of glacier implies more than a change; this is a decisive step from no soil to the presence and the development of soil (this comment applies to the whole document).
-L62-64: this sentence is pure evidence; I do not see the point in writing it here.
-L65 and whole document “deglaciation”: see general comments on terms: deglaciation describe the total disappearance of the glacier (different from deglacierization)
-L67-68: this sentence is true but there's a problem of causality (avoid the term “thus”?)
-L71-72, several papers have not been taken into account here (and whole document, see general comments).
-Research questions and hypotheses: RQ1 and RQ2: not generalisable, only applicable at the scale of your site. The two hypotheses are not introduced before, need to be introduced within the state of art (current limits of knowledge).
METHODS
-Why not mentioning that you work with post-glacial chronosequences (see general comments)?
-L94 “LIA moraines”: you need to write in full and explain what it is.
L94-95 “This transect is independent of elevation gradient, which ranges between 1890 and 1990 m a.s.l”: How can it be independent? They are crossed. Do you mean that the elevation gradient is very shallow and you expect it not to have a significant influence on the temporal gradient?
-L105 “For each stage, we randomly selected four plots of 3x3 m”: area seems good but cannot be selected without justification, please justify, using the existing literature.
L108-109 “Plots were resurveyed every two weeks for completeness until the end of August”: nice effort! Might be relevant to provide raw data as an appendix to see the usefulness of this method.
-L110 “the surface cover” do you mean “the relative cover at soil surface (of each species)”?
-L110 “assessed visually”, vocabulary: “estimated visually” (please provide details: did you use a grid? Did you spend the same time estimating each plot?)
-L113, similarly “determined” (direct measurement) is not correct, it is an estimation.
-L115 “average reaction index”: Good point: Landolt values are powerful and not sufficiently used, in my opinion”
-L119 “plant functional groups”: this is not a fully functional approach, rather a proxy, which is interesting of course but the readers expect other types of traits when referring to this term. this has been done in other studies and should be discussed here and in the discussion.
-L121 replace “abundance” by “relative cover (homogenization of terms see general comments)
-L123 “A soil sample was taken from the top 10 cm of soil”: but then what do you do when there is no soil (most probable situation)? See general comments
-L124 similarly, where is the “organic layer” just after glacial retreat? Same with “the remaining soil samples represent organic horizons” L126
RESULTS
-Use past tense in the result section.
-L206: “Indeed, the first dimension includes vegetation variables”: no causality: all these variables could be influenced by geomorphological processes as well; no? (axis 2)”
-L212 “highly correlated with a positive correlation”: rephrasing needed.
-Fig. 2 PCA: The relevance of a multivariate analysis, especially a PCA, depends on the number of variables that are included: more variables = more correlation between variables. the « game to play » is thus to explain a large % a variation with the lowest number of variables. Here, the high number of variables makes that this goal may be not reached. Why not reducing the number of variables?
L227, L352, L385, L396 and whole document “glacier retreat influences (plant diversity)”: not correct, time since glacier retreats influences plant diversity.
L237 “Statistics confirm that glacier retreat influences proportion of forbs”: how possible with so few repetitions (4)?
DISCUSSION
-Vegetation development: I find it very interesting to go into the strategies of species but we are soon frustrated because no details are given: why these species L287-288 are considered pioneer species, what is a pioneer species: improvements are clearly at hand here! Don’t forget that most of the first colonizers are here because of their high dispersal capacities instead of their pioneer, stress-tolerant traits (see general comments)
L294-295 “Losapio et al. (2021), who note that plant diversity initially increases with glacier retreat before decreasing in the late stage” be careful with the overuse of self-citations (see general comments). Many other references describe a monotonous increase in plant diversity at the same temporal scale, beginning with Matthews 1992. A poor state of art severely reduces the scope of the document currently. (see general comments).
-L298-302: this section about plant-plant facilitation should lie on solid existing literature (see general comments), especially the lack of facilitators at the beginning of the succession should be introduced/discussed.
-Section vegetation development in general: I feel that this section is much less relevant than the next section, as if primary succession was reinvented 100 years later but using one single site. I recommend a focus on the next section (soils and associations with plants instead)
-Section soil development: interesting but need more connexions with the existing literature (see general comments).
-L377_378 “the increase of organic carbon, nitrogen, and available elements positively influences the development of certain plant functional groups such as dwarf-shrubs and trees”: Correlation is not causality, you do not provide proofs that one variable has an effect on another.
-L385_386 “Glacier retreat decreases soil pH, which in turn increases soil organic carbon”: same comment, correlation is not causality, you do not provide proofs that one variable has an effect on another
-L386-387 “This model shows that when plant diversity increases, soil organic carbon tends to decrease and that high levels of soil carbon are associated with low plant diversity”. I do not agree. Once again, no causality has been demonstrated here, a great deal of caution is needed in interpreting the facts.
-L387-388 “This can be explained by the fact that, with increasing glacier retreat in stage 4, there is an accumulation of organic matter and plant diversity is the lowest” (needs rephrasing): problem of sampling (few repetitions) or tree cover (light competition)? In any case this is not generalisable (see general comments)
CONCLUSION deals with too much generalization given that the data has been collected in one single site, with very few repetitions (total: 16 plots). Sentences in this section are highly speculative and/or refer to the evidence (see below), I recommend trying not to oversell the results of the document and focus on its true contribution (which means comparing it more with the existing literature).
-L406-407 “These results suggest that novel glacier ecosystems may function as carbon sink by storing carbon in the soil” this type of sentence doesn’t bring anything new: this is evidence (after glacier retreat carbon is nearly zero), has been studied much more extensively in other studies (e.g., Khedim et al. 2021) and may be irrelevant considering the small amount of carbon stored.
-L407 “They also show that nitrogen is a limiting factor”: similarly, this is evidence and, more generally, this is one of the main constraints for plants in alpine regions (Körner 2003, 2021, reference that has not even been cited).
Figure 2 looks rather like a first draft directly extracted from R, which needs to be edited. Given the doubts raised about the statistical significance of the results (low repetitions, lore detailed presentation of the statistical tests is required, including tables in appendices.
Citation: https://doi.org/10.5194/egusphere-2024-991-RC1 -
AC1: 'Reply on RC1', Gianalberto Losapio, 23 May 2024
General Response: We are grateful for your work with editing our manuscript. We have followed the helpful recommendations you provided and addressed all their concerns, as explained below.
Response (1): Thank you for suggesting additional literature. We now add those references. In doing so, we introduce the results while incorporating this additional knowledge to our study case and discussing our results in light of those previous studies.
Response (2): The Reviewer is right that with “only” one site there are not many possibilities for making broader generalization. We now pay attention not to generalize too much. We also make it clear that, as always in ecology, there can be nuances and context-dependencies. We now put the results of this manuscript in the context of that suggested literature by making a narrative comparison in the discussion section. In doing so, we pay attention to stay within the limits provided by our case study.
Response (3): It is true that environmental heterogeneity is always a challenge in biogeosciences. We now provide evidence that enough variation and heterogeneity was covered by random sampling as demonstrated by the rarefaction analysis, which is now provided in the SI. Yet, we now do acknowledge this potential limitation in the methods section.
Response (4): Thanks for your suggestions, which are substantially improving the narrative and readability of our manuscript. We are following their suggestion by streamlining the different paragraph. We also pay attention to use consistent terminology.
Response (5): Soil has been defined as “the upper layer of earth in which plants grow, which consists of a mixture of sediments, organic matter, and rock (mineral) particles.
The terms relating to glaciers have been unified as follows: « glacier retreat » to designate the process where glaciers melt and retreat; “deglacierization” to refer to the process where the glacier melts without having disappeared; “glacier extinction” to refer to the state predicted with current climate change where there will no longer be a glacier.
Plant functional groups has been replaced by plant life-forms.
Abundance has been replaced by relative cover.
The term novel has been defined and properly used
We specified that litter, such as dead leaves or branches, was removed from the upper horizon.
Response (6): Thank for this suggestion. Previously, we did not refer to mechanisms as this case study was of comparative nature. Yet, it is true that our path analysis can provide a robust clue inside potential mechanisms. We now refer to and discuss potential mechanisms underlying the observed patterns.
R: Thanks for your suggestion. We changed the title to “Glacier retreat shapes plant–soil relations over spacetime”
R: research questions added in the abstract
R: changed
R: added
R: term 'novel' defined
R: rephrased
R: citation corrected
R: literature added
R: modified
R: removed
R: corrected
R: corrected
R: added and completed
R: fixed
R: added
R: Yes that was the purpose of this sentence. This has been corrected to clarify.
R: We clarified the random selection
R: specified and added
R: terminology corrected
R: in this sentence, determine has been replaced by evaluate
R: corrected
R: changed
R: rephrased
R: corrected
R: references and state of art have been improved with new references
R: the discussion has been substantially revised
Citation: https://doi.org/10.5194/egusphere-2024-991-AC1
-
AC1: 'Reply on RC1', Gianalberto Losapio, 23 May 2024
-
RC2: 'Comment on egusphere-2024-991', Anonymous Referee #2, 27 Apr 2024
Dear Charles et al.,
Your manuscript “Plant–soil interactions underline the development of novel ecosystems after glacier retreat” handles a highly relevant subject in the eye of the current global crisis of climate change and biodiversity loss. The imminent loss of (the largest part of) all glaciers in Central Europe by 2100 is a sad prospect, and the study of the ecosystems that develop at the retreating glacier front will help us protecting the newly developing communities.
The largest downside of this study is the lack of replication. The retreat of only one glacier is studied, while differences in (micro)climate, bedrock, altitude, the set of plant species (and other taxa) available for invasion, etc. could have a large influence on the outcome of the succession. Figure 2 also indicates that all plots were located rather low in the valley (or high on the ridge (?)) of the deglaciated area. This made me wonder whether the outcome could even be different if the plots had been higher (or lower) in altitude for this specific case. Is the final successional outcome always a species poor forest, or do higher ridges only reach stage 2 or 3 without continuing to stage 4? (Or does the fact that the glacier retreats on a certain spot implies that the final stable plant community in the succession is a species poor forest?)
As sampling a sequence close to a retreating glacier front is not trivial, and as the documentation of the age of the deglaciation along the deglaciated might be challenging to acquire (?), I do understand that this study was limited to one glacier only.
However, it would be a great asset for follow up studies to include more retreating glaciers, preferentially distributed over a large climatic and altitudinal range.Your study strongly focusses on species diversity and richness as the factor to optimise after deglaciation. This is a valid choice, especially in mountainous areas, where climate change pushes cold adapted species upwards until they have reached the summit and are outcompeted by more generalist species. However, with the broad range of parameters that is measured in this study, other factors could be selected to focus on. One of them (especially in the context of climate change) is carbon storage. Carbon storage is (not unexpected) the highest in Stage 4 (the species poor forest). You do mention this briefly in the conclusion (line 405-407), but it would be good to highlight this more clearly also in the abstract and maybe also in the discussion session.
Species diversity and richness are optimal (highest) in the intermediate successional stages (Stage 2 and 3). You urge several times to use this study as a base for conservation efforts for species protection and preservation. However, no suggestions for conservation methods and aims are given. I think that it would strongly strengthen your story if you could (even briefly) focus on this aspect.
Finally, your manuscript is well structured, and the storyline is easy to follow and interpret. In general, the language is fluent and well written. You can find some grammatical and spelling corrections (and suggestions) in the technical corrections section.
Specific comments
Lines 21-23:
“Plant diversity initially increases after glacier retreat, along with an increase in soil organic matter and nutrients. Over 120 years, soils acidify at a rate of 0.02 units per year, the C/N ratio increases while plant diversity decreases.”
Do I interpret it correctly that you found an initial increase of plant diversity, but an overall decrease (compared to the maximum that is reached at a certain point?) over the first 120 years? If so, when does the plant diversity reach its maximum and starts to decrease again after the initial increase? I would rephrase this sentence to make it easier to interpret.
Line 26-28:
Do I understand it correctly that, after glacier retreat, the stable state of the plant community is a species poor forest (which is, however, probably still more species rich than a glacier ecosystem but containing less rare / threatened species?). You mention that targeted biodiversity conservation efforts should be made to move the ecosystems back to their initial, more species rich states. Could you elaborate on the types of conservation efforts would be effective in this specific case (e.g. increased disturbance by X / mowing / …)?
Lines 78-79:
“We also expect that an increase in plant diversity will lead to an increase in soil organic matter,”
Why is it specifically the increase in plant diversity that would lead to an increase in soil organic matter? Would this also not happen with the colonization of a species poor plant community?
Line 106:
“we randomly selected four plots of 3x3 m”
How was the random selection performed?
All the plots seem to be quite high on the ridge (or low in the valley?). Do you think that the results have been different if the plots were located at lower (or higher?) altitudes?Lines 125-126
“Litter and the first few centimetres of soil composed of slightly degraded organic debris were discarded, the remaining soil samples represent organic horizons.”
Does this procedure only apply to the two plots in stage 4 that had a very thick organic layer?
Otherwise, I´m a bit confused. In line 124, there is stated that the organic layer was removed.Line 141
At what temperatures where the soil samples oven-dried before C and N analyses?
Line 295
What is the starting stage from which the plant diversity increases after glacial retreat? I did assume that there were close to no plants growing on a glacier, but maybe I am wrong? Except from mosses, are there vascular plants that occur on glaciers?
This question occurred to me several times while reading your paper. Maybe mention this briefly when describing the successional sequence? You could call it ´stage 0´?Line 386
“This model shows that when plant diversity increases, soil organic carbon tends to decrease […]”
I find this a strange statement. This is true if one goes from stage 4 to stage 3 or 2 (going back in the successional sequence), but not if one goes from stage 1 to 2 or 3, which is the natural course of events. Going back from stage 4 to 3 or 2 requires a disturbance, which doesn´t even necessarily reduce the soil organic carbon.
I do agree with the second part of the sentence (high levels of soil organic carbon are associated with low plant diversity).Technical corrections
Line 36:
Global warming is currently increasing by …
--> Global (average) temperatures are currently increasing byLine 37:
Thus by 2030 and 2050, temperatures are expected to increase further by 1.5 °C (IPCC, 2022).
Should there not be two temperatures, indicating the expected temperature increase by 2030 and by 2050?
Line 95:
“m a.s.l.”--> “m.a.s.l.”
Line 102:
When listing the different stages of ecosystem development, they are called “pioneer stage”, “early stage”, “intermediate stage” and “late successional stage”.
However, just after that, they are called “stage 1-4”.
I would already in the list add which developmental stage corresponds to which stage number:- Pioneer stage: 1989–2022 AD (Stage 1)
- Early stage: 1925–1989 AD (Stage 2)
- Intermediate stage: 1900–1925 AD (Stage 3)
- Late successional stage: 1864–1900 AD (Stage 4)
Line 103:
The four stages are all about 250 m wide.
Is 250 m long a better description? I interpret 250 m wide the with perpendicular on the length of the glacier, while long corresponds (in my perspective) on the length of the stage parallel to the length of the glacier.
Line 106:
“Hereafter, these plots are named “a” to “d”.”
Here, small letters are used to refer to the replicates, while in Figure one, capitals are used. I would choose one of both and use it consistently.
Line 125:
“centimeters”--> centimetres
Line 128:
“For each plot, at least 200 g of soil were taken back to the lab.”
--> For each plot, at least 200 g of soil was taken back to the lab.Line 133:
“We conducted soil analyses including measurement […]”
--> We conducted soil analyses, including measurements […]
Table 1:
“Plant reaction value – Community weight man […]”
--> Plant reaction value – Community weight mean […]
Table 1:
“SOC – Soil organic carbon content (SOC)”
--> SOC – Soil organic carbon content
Table 1:
I would add units to the description where applicable.
Lines 159-160:
“As exploratory analysis, Pearson correlation values were calculated among all dependent variables, making it possible to obtain 160 a correlation matrix and observe linear relationships between pairs of variables (results reported in SI).”
I don´t find this back in the Supplementary Material.
Line 188:
“[…] predictors are qualitatively similar […]”
--> […] predictors were qualitatively similar […]
Figure 2 (Line 220):
The contribution of an observation to a component is visualized by the darkness of the arrow in Figure 2. However, I do not see how the legend to the right (contrib: 1 / 2 / 3 / 4) explains this? I would remove this legend or make it more intuitive.
Lines 233-235:
“For each stage, the median reaction value drops from 3.2 to 2.9, 2.4 and finally 1.9 from stages 1 to 4, respectively. Our stages data therefore goes from lightly acidic to neutral, to acidic and finally to very acidic.”
à Did you mean lightly alkaline?
Figure 3.c:
Legend: “subshrubs”--> dwarf shrubs
Figure 4:
Text:
- Also here I would add the comment that “boxplots have been jittered for facilitating data visualization; the actual years since deglaciation is the written one.”
- (a) Soil organic carbon content (SOC)
- (b) Soil total nitrogen content (Ntot)
- (c) Organic carbon / total nitrogen ratio (C/N ratio)
- Y scale à The y-axis is log distributed …
Line 259:
“[…] around c 2.4% […]”
--> […] around ca. 2.4% […]
Lines 266-270:
Here, the C/N ratio is expressed in %. The C/N ratio, however, does not have a unit and is usually expressed as a ratio of the number of units of carbon to the number of units of nitrogen. I would change the text in this way (e.g. Stage 1 has a C/N ratio of 2:1).
Line 271:
It is not necessary to indicate the age where stage 1 corresponds to, as this is not done elsewhere or for other stages.
Line 331:
“Late development stage shows particularly low pH […]”
--> The late developmental stage shows particularly low pH […]
Lines 354-355:
“The increase in plant cover and diversity characterizing by […]”
--> The increase in plant cover and diversity characterizing by […]
The increased transfer of organic matter to the soil is not a consequence of increased plant diversity, but purely of increased plant cover.
Line 362:
“[…] total nitrogen are often below 0.1 % […]”
--> […] total nitrogen is often below 0.1 % […]
(Or better: The soil total nitrogen content is often below …)
Lines 364-365:
“The presence of microorganisms involved in the nitrogen cycle including microorganisms associated to nitrogen-fixing plants can explain […]”
--> The presence of microorganisms involved in the nitrogen cycle, including microorganisms associated to nitrogen-fixing plants, can explain […]
Line 365:
“Early and intermediate development stages […]”
--> Early and intermediate developmental stages […]
Or if you mean it specifically for your study (and not for early and intermediate developmental stages in general):
--> The early and intermediate developmental stages […]
Lines 368-370:
“In stages 2 and 3, SOC and total nitrogen content are similar corresponding to similar vegetation indices in these stages. On the contrary, plants in Ericaceaea family Rhododendron ferrugineum, Vaccinium myrtillus, Calluna vulgaris) dominate late development stages.”
I (think I) understood this sentence after reading it a few times (that the SOC and Ntot are similar in stage 2 and 3 and that also their vegetation composition was very similar), but I would rewrite the sentence to make it easier to understand.
Line 371:
“[…] matter that can seen in the increasing […]”
--> […] matter that can be seen in the increasing […]
Lines 376-378:
“While low concentrations of available elements in the early stages limit plant growth and establishment (Chapin et al., 1994), the increase of organic carbon, nitrogen, and available elements positively influences the development of certain plant functional groups such as dwarf-shrubs and trees.”
--> While low concentrations of available elements in the early stages limit plant growth and establishment (Chapin et al., 1994), the increase of organic carbon, nitrogen, and available elements in the later stages positively influences the development of certain plant functional groups such as dwarf-shrubs and trees.
Line 381-383:
“Furthermore, the presence of bacteria and fungi associated to pioneer plants is decisive for the availability of certain elements, such as phosphorus and nitrogen, for shrubs and trees (Egli et al., 2012).”
“Furthermore, the presence of bacteria and fungi associated to pioneer plants is decisive for the availability of certain elements, such as phosphorus and nitrogen, which are essential for species that are associated to later stages (such as shrubs and trees) to establish (Egli et al., 2012).”
Line 384:
“The path analysis model provides an […]”
--> The path analysis model provided an […]
For the whole paragraph: In past tense as the model is built on own results?
Line 399:
“Pioneer plant species enables diversity […]”
--> Pioneer plant species enable diversity […]
Line 400:
“[…] soil conditions, species become increasingly dominant […]”
--> […] soil conditions, certain (or competitive?) species become increasingly dominant […]
Line 405:
“[…] although both soil carbon and nutrient increase […]”
--> […] although both soil carbon and nutrients increase […]
Line 409:
“[…] mitigation actions shall consider […]”
--> […] mitigation actions should consider […]
Supplementary material, Figure A1(b)
- The axis labels are not readable.
- It is not clear where Stage 1, 2, 3 and 4 are located in the soil texture triangle classification.
Supplementary material, Figure A1 - Legend:
“Additional soil and plants variables […]”
--> Additional soil and plant variables […]
Supplementary material, Figure A1 - Legend:
“[…] including plants available elements concentrations […]”
--> […] including plant available elements concentrations […]
Supplementary material, Figure A1 - Legend:
“Total element proportions (c)”
--> Total element proportions (c)Also, the header of Figure A1.c refers to “Major elements”, while the legend refers to “Total elements”. Please make this conform.
Citation: https://doi.org/10.5194/egusphere-2024-991-RC2 -
AC2: 'Reply on RC2', Gianalberto Losapio, 23 May 2024
Response: We are grateful for your work with editing our manuscript. We have followed the helpful recommendations you provided and addressed all their concerns, as explained below.
Response: The Reviewer is right that with “only” one site there are not many possibilities for making broader generalization. We now pay attention not to generalize too much. We also make it clear that, as always in ecology, there can be nuances and context-dependencies. We now put the results of this manuscript in the context of that suggested literature by making a narrative comparison in the discussion section. In doing so, we pay attention to stay within the limits provided by our case study.
Response: Thank you for your suggestion. This point has been removed in the manuscript as suggested by other referees who refer to the article by Khedim et al. 2021 and consider that this is evidence and may be irrelevant considering the small amount of carbon stored.
Response: those points have been added in the discussion (at the end of 4.1)
R: Yes, the sentence has been reformulated and clarifications added
R: some suggestions on conservation efforts have been added such as removal of encroaching woody vegetation
R: That’s right, an increase in plant diversity doesn’t mean an increase in soil organic matter. This was not worded correctly and has been corrected consequently. Here we expected that an increase in plant colonization will lead to an increase in soil organic matter. We suppose that the increase in plant colonization will be accompanied by an increase in plant diversity.
R: we specified that the foreland was examined, avoiding the lateral moraines (which are highly unstable and pose a risk to researchers)
R: Yes, the explanation has been clarified. We removed the litter horizon. The remaining soil samples represent the organo-mineral horizons. But as it was not possible to remove the organic horizon completely for two plot in stage 4, litter and the first centimetres of soil were discarded.
R: precision added
R: in some glaciers such as debris-covered glaciers, there are vascular plants occuring on the glacier debris (https://link.springer.com/article/10.1007/s11829-016-9444-x). We have no evidence that the same occurs here though. We improve the description of the vegetation at the study site.
R: we see the misunderstanding; that sentence has been rewritten
R: changed
R: corrected, after checking the source, temperatures are expected to rise by 1.5°C between 2030 and 2052
R: changed
R: Yes, fixed
R: Yes, fixed
R: that's right, everything has been homogenized in capital letters
R: changed
R: changed
R: changed
R: changed
R: changed
R: added
R: added
R: changed
R: this legend has been removed
R: We understand that this is not clear, it's from neutral/lightly acidic, to acidic and finally to very acidic. This sentence has therefore been reworded
R: corrected
R: added and modified
R: modified
R: changed
R: removed
R: modified
R: Thank you for your comment, this has been corrected
R: modified
R: commas added
R: modified by the second suggestion
R: modified as: “In stages 2 and 3, vegetation composition is similar, as are SOC and total nitrogen. These stages are different from the late developmental stage where plants from the Ericaceaea family (Rhododendron ferrugineum, Vaccinium myrtillus, and Calluna vulgaris) dominate stage 4.”
R: corrected
R: added
R: added
R: done
R: corrected
R: “certain” added
R: corrected
R: corrected
R: SI revised and updated
R: corrected
R : corrected, and made uniform with “major elements”
Citation: https://doi.org/10.5194/egusphere-2024-991-AC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 356 | 118 | 29 | 503 | 21 | 22 |
- HTML: 356
- PDF: 118
- XML: 29
- Total: 503
- BibTeX: 21
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1