the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Sep 2023
Environmental and vegetation control on active layer and soil temperature in an Arctic tundra ecosystem in Alaska
Abstract. Permafrost soils contain approximately twice the amount of carbon than the atmosphere, which could be released as global warming continues. Increasing global temperatures have in fact the potential to result in increased permafrost degradation, and carbon loss into the atmosphere. To properly understand the potential release of the carbon stored in permafrost soils, it is critical to understand the environmental and vegetation control on the development of active layer, the upper soil layer that thaw during the growing season in the Arctic. Arctic tundra ecosystems are dominated by mosses, which compose approximately 40 % of the vegetation, and have a critical role in regulating the heat condition into the soil. Given their importance, the role that mosses play on permafrost degradation should be investigated in more details. This study measured soil temperature together with thaw depth, a range of environmental variables, and moss thickness, to identify the most important controls on the development of the active layer across 124 plots in continuous permafrost tundra ecosystems. We found that a thicker moss layer insulated the soil and resulted in cooler temperatures deeper in the soil, despite warmer surface temperatures. A thicker moss layer was associated with a deeper depth of thaw, likely for the higher growth of mosses in the drier and warmer topographically higher elevation areas. The protective role of mosses was only relevant for the first ~3 cm of the green moss layer, suggesting that the living moss layer was more important in regulating soil temperature, possibly through a higher ability to retain water. Soil moisture was in fact an important control on surface and deeper soil temperatures, with wetter soils been associated with cooler surface temperatures because of the higher evaporative cooling, and warmer deeper temperatures likely because of the larger heat conduction to deeper soils. Overall, this study highlights the importance of a green living moss layer on soil temperature and thaw depth. Mosses are among the most vulnerable vegetation to hydrological changes, given their lack of a rooting system, and their sensitivity to climate change should be considered when predicting the response of permafrost thaw to climate change.
- Preprint
(1346 KB) - Metadata XML
-
Supplement
(41 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on bg-2023-117', Anonymous Referee #1, 13 Nov 2023
General comments
This manuscript describes a study on the controlling factors of soil temperature and thaw depth for arctic tundra in Alaska, with a focus on the role of mosses. The study collected data on thickness and composition of moss layer (green/total), soil temperature profiles, soil water content, water table, thaw depth along two transects in a well-established research site. The study found that moss layer thickness, microtopography and soil moisture influence soil temperature profiles, and that near surface temperature controls thaw depth.
This study provides some insight as to how mosses and other environmental factors influence soil microclimate and can influence permafrost thaw. However, I feel the current manuscript in general lacks clarity. It is not clear to me what exact knowledge gap this study is trying to fill and how the analysis used is addressing this. I believe the manuscript can be much improved by looking more into how these different drivers interact and the mechanisms involved. I will give further details below.
The introduction can be improved by restructuring the current text and better introducing the exact research question this study is addressing. At the moment, the introduction is a bit repetitive as certain points are made multiple times and arguments that belong together are spread across paragraphs. I also struggled to identify what knowledge gap this study is trying to fill and how the study addresses this. From the introduction it seems like the role of mosses in influencing soil temperature and thaw depth is already well established. I believe this study could still provide additional insight by looking more into specific mechanisms (non-linearity of responses), and interactions between different factors and context dependency.
The Methods section is generally clear but needs a couple of clarifications, see specific comments.
The statistical analysis and result section lacks clarity. Many predictor variables are used, but there is little reflection as to why certain variables are used. I personally also think that there are too many graphs in the results section, which adds to me losing track of what the authors really are trying to show.
The authors mention the correlation between variables and that these are not used together in multiple regression analysis. However, they do not provide clear information as to which variables were included in multiple regression and why.
In the analysis, a number of temperature parameters are used, both as predictor and response variables. However, there is little explanation as to why these different parameters are of importance or why they may have different drivers. This lack of context makes it hard to see the relevance of these different analysis and because of the many different temperature variables it becomes hard to follow and confusing.
The authors use deviation form mean elevation (dz) as a predictor variable, which indicates that there are raised plots and hollows which have distinct environmental conditions and also influence moss thickness. However, the interrelatedness does not really come forward until later.
I also wonder whether the moss dominance varies between hollows and raised plots?
Hrbáček et al. 2020 found species-specific effect moss on ground surface temperature and active layer depth due to differences in water retention capacity and structure. The authors find no difference in the influence of moss genus, which to me is an interesting point and could be further discussed. Are the dominant species in your site very similar in structure and water retention capacity or could your results be confounded by topography.
In my opinion, the analysis needs to better substantiated and should better reflect the interrelatedness of these different factors as illustrated by the visual summary. A structural equation model may be a more appropriate analysis which can show these dependencies. In a structural equation model air temperature, PAR and net radiation could be used as general climate conditions that influence soil temperature. Soil temperature can further be modulated by soil moisture content and moss thickness, which in concert regulate thaw depth.The present analysis is not able to tease apart the effect of moss layer thickness from environmental conditions and the resulting effect on thaw depth.
The general readability of the manuscript can also be improved by splitting up long sentences.
Specific comments
Instead of a picture of the datalogger, a picture of the fiberglass probe would be more illustrative of the research method.
Line 110 Can the authors include specifics on the graminoid species present?
Line 120 states that moss and soil temperature were measured from 1cm below the surface? What do the authors mean with below the surface? Does it include temperatures in the moss layer as is alluded to? Or is it only soil temperatures starting from soil surface? This will matter for the analysis.
Can the authors more clearly state the frequency of measurements of the different variables.
Where measurements done in a randomized manner? Otherwise, one can expect diurnal patterns to influence the soil temperatures measured at specific locations.
I suggest using a correlation matrix to illustrate the correlation between various parameters and the significance. This will provide a much clearer overview than the current supplementary table.
Taken together I believe this manuscript needs extensive revisions to provide clarity by clearly defining research questions and how these are addressed. I also believe the authors need to rethink their analysis, so it better matches the inter-relatedness of their predictors. I therefore suggest the authors to thoroughly revise this manuscript before a potential resubmission.
Citation: https://doi.org/10.5194/bg-2023-117-RC1 -
AC2: 'Reply on RC1', Donatella Zona, 12 Dec 2023
The comment was uploaded in the form of a supplement: https://bg.copernicus.org/preprints/bg-2023-117/bg-2023-117-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Donatella Zona, 12 Dec 2023
-
RC2: 'Comment on bg-2023-117', Anonymous Referee #2, 15 Nov 2023
The paper describes a study where the relationship between moss properties and environmental variables like thaw depth, soil temperature and moisture are investigated. The basis for the analysis is a dataset consisting of two years of measurements of these variables along transects in the Alaskan tundra. The authors run their dataset through various statistical models and find that 1. Moss layers drives larger temperature gradients in the ground and limits active layer thickness, 2. Surface temperatures are cooler in moister locations, and 3. That the topmost, live part of moss is important in regulating soil temperatures. The authors relate these findings to the low thermal conductivity of dry moss, the thermal conductivity and heat capacity of water, and the latent cooling associated with evaporation. The findings are briefly discussed.
After my reading of the paper, I find it to address a relevant research question, namely which factors regulate the energy transfer between the atmosphere and the ground in moss dominated ecosystems. While the topic of the paper is within the scope of Biogeosciences, I do not find it to present the novel and substantial scientific contribution that a publication in this journal necessitates. Most notably, the papers finding 1. and 2. are previously established relationships, and finding 3. is presenter without the appropriate discussion. Overall, I find that the paper has considerable shortcomings in several aspects including scientific significance, quality and presentation. I would thus recommend the authors to rewrite and resubmit the paper in an appropriate journal.
I have summarized my comments of the paper, and suggest the authors to consider these when reworking the paper.
Title
The paper primarily investigates how soil temperatures and thaw depth are influenced by moss thickness and soil moisture, and the title should reflect that.
Abstract
The relevance, key aims, data sources and conclusions are presented in an orderly and appropriate way. As roughly ¼ of the introduction is about the risk of carbon release from permafrost soils, it would be appropriate to mention the processes through which mosses contributes to protecting this carbon stock. The vulnerability of mosses (Line 26-28) is also important context, but is not mentioned in the body of the paper (Introduction – conclusion).
Introduction
The authors provide relevant background for their study, including the protected carbon stocks found in permafrost soils, positive feedback mechanisms upon permafrost degradation, and projected climate and wetness change. The paper however fails to revisit these important contexts in the discussion.
A major issue in the introduction is that it is not made clear what research gap the authors investigate in the study. The goal is stated to be “to identify the biotic and abiotic controls regulating soil temperature and the thawing of the active layer”, while the paragraph from Line 52-62 describes that there already is an established scientific understanding of the role of mosses in this context. The introduction should clearly state what gap in current scientific understanding the paper is addressing, and why their method and study site is appropriate. If the paper mostly aims to confirm well-known relationships, there needs to be an argument for why it is still relevant. For example, that the authors have a focus on previously underrepresented ecosystems/study areas/climates, or that the study is more quantitative than previous ones.
Methods
The study site description does not provide the reader with a proper overview of the site that is investigated. Here, I feel there should be a description of the general area in terms of climate, topography and dominant ecosystems. The local site description needs also to address the representativity of the site – i.e. can findings from this site be used elsewhere, both regionally and globally? The study site description also markets other data and experiments (eddy covariance, Biocomplexity etc.) without any clear explanation of why these are relevant to the paper. Credit of previous research efforts should be limited to the acknowledgements, whereas the study site description simply cites the studies that provide the data or statements used in the current study.
I also find the sampling description to be ambiguous. Why do the authors choose to sample along these transects? Do they follow some environmental or topographical gradients of interest? It is also completely unclear what the measurement period is, i.e. do you sample weekly year-round or only in a limited period in summer? And are the temperature and moisture measurements instantaneous or daily values? Without such metadata and general information, it is not possible for the reader to assess whether the acquired dataset is suitable to answer the questions at hand.
For the environmental variables, all those that are used in the study need to be listed, potentially in a table also indicating their units, annual range and type of sensor. From line 136-137 it is unclear if these are alle variables used in the study, or if there are additional ones not listed.
This section (or the results) also lacks a short description of the data that is actually sampled. What are the annual ranges and averages? Was some sort of filtering applied? Are there clear clusters or thresholds for some of the variables?
Statistical analysis
I was surprised that the paper does not make an argument for the choice of statistical methods. While I am not an expert in statistical modelling, I would expect at least a simple statement to why the methods used in this study are suitable for the data and research question. This might be especially relevant as moisture has natural limits at 0 and 100%, and temperature has a hiatus around 0°C during thawing/freezing and thus does not behave linearly. I also found that stepwise regression is a controversial and partially discouraged method (e.g. Flom & Cassel (2007)), and I would expect the authors to comment on the applicability of this method.
It was also puzzling that in line 155 the depth -15 cm is stated to be “most consistently represented”, while in 120-121 says temperature is recorded every centimetre until 20 cm for each plot. If there are issues implementing the sampling routine outlined, this should be mentioned and explained. In general, the state and nature of the dataset and choice of methods is not described in a manner fostering replicability.
Results
This section does not present the findings in a clear and straight forward manner. Figures are presented before they appear in the text, there is a mixture of variable names and symbols in both text and graphs, and the metrics such as “Akaike information criterion” are not explained. It is also not clear which results are based on some sort of average and which are time series (e.g. Figure 4 where soil water contents are regressed against maximum temperatures).
Discussion
While the topics brought forward here are of relevance for the study, this section fails to convey how this study takes research forward. That mosses insulate the soil from the atmosphere and that evaporation cools the surface are established concepts within the field. I do find it highly interesting that the topmost live layer of moss has such a strong impact on soil temperatures, and would have expected a more thorough discussion of possible processes.
A major issue with the paper is that the discussion does not revisit the important context provided in the introduction; the carbon stocks in permafrost soils, positive climate feedback mechanisms, changes in precipitation and the vulnerability of moss ecosystems. Several of these topics are strongly linked to the authors findings, and a proper discussion of them would greatly improve the relevance of the paper. The paper also needs to address the quality and robustness of the sampling routine, data and methods used, and potentially outline suggested improvements.
Conclusion
This section is concise and to the point. The statements that a wider range of soil moisture (line 307) and moss thickness (line 308) would be required to understand this topic comes as a surprise as this topic is not mentioned in the discussion. The conclusion should briefly present the aims and how they were/where not achieved, rather than presenting new topics.
References
Flom, P. L., & Cassell, D. L. (2007, November). Stopping stepwise: Why stepwise and similar selection methods are bad, and what you should use. In NorthEast SAS Users Group Inc 20th Annual Conference (Vol. 11).
Citation: https://doi.org/10.5194/bg-2023-117-RC2 -
AC1: 'Reply on RC2', Donatella Zona, 12 Dec 2023
The comment was uploaded in the form of a supplement: https://bg.copernicus.org/preprints/bg-2023-117/bg-2023-117-AC1-supplement.pdf
-
AC3: 'Reply on RC2', Donatella Zona, 12 Dec 2023
The comment was uploaded in the form of a supplement: https://bg.copernicus.org/preprints/bg-2023-117/bg-2023-117-AC3-supplement.pdf
-
AC1: 'Reply on RC2', Donatella Zona, 12 Dec 2023
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 273 | 79 | 22 | 374 | 29 | 15 | 14 |
- HTML: 273
- PDF: 79
- XML: 22
- Total: 374
- Supplement: 29
- BibTeX: 15
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1