the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Jun 2025
Forest disturbances and their impact on ground surface temperatures in permafrost-underlain forest in central Mongolia
Abstract. In the forest-steppe ecotone in central Mongolia, forest and permafrost exist close to their climatic limits and are co-located on north-facing slopes. The deciduous forest ecosystems and permafrost on these slopes are linked through interactions in the local energy and water balance. Furthermore, in this region the presence of such permafrost-forest systems provides essential services that supports local livelihoods and ecosystem function. However, forest disturbances that reduce or remove the forest canopies and lead to changes in surface cover could impact ground surface temperatures (GSTs) and potentially lead to permafrost degradation. In this study, we investigate the relationship between different forest states and GSTs at a site in the forest-steppe ecotone. We measured GSTs and surveyed vegetation density and surface cover over two years in an area that features both intact, dead and logged forest and dense stands of young regrowth. Overall, we find GSTs in summer and winter to vary substantially among the forest states, while differences in GST in spring and fall are small. Compared to the intact forest, the annual GST range is increased in the dead and logged forest while it is dampened in stands of young regrowth. Contrary to existing literature, we do not observe a general warming of the ground surface at disturbed sites, but instead find mean annual GSTs at disturbed sites to be 0.5 °C lower than at intact sites. We also find substantial floor vegetation in the dead and logged forest, which has implications for livestock grazing patterns and remote sensing of forest disturbances.
Competing interests: At least one of the (co-)authors is a member of the editorial board of The Cryosphere. The peer-review process was guided by an independent editor, and the authors also have no other competing interests to declare.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1360 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1360 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-2366', Anonymous Referee #1, 18 Jul 2025
- AC3: 'Reply on RC1', Robin B. Zweigel, 23 Oct 2025
-
RC2: 'Comment on egusphere-2025-2366', Anonymous Referee #2, 24 Sep 2025
This study presents a new dataset on GST, vegetation density, and surface conditions. These data show how forest disturbances affect the forest-steppe ecotone in central Mongolia in the context of ground surface temperature fluctuations. This is an important issue that is rarely addressed in the literature. The problem of vegetation response to permafrost degradation, and conversely, the influence of vegetation type on ground temperature, remains unclear. Although the authors present interesting conclusions regarding the values of temperature extremes depending on whether we are dealing with dead, young or logged vegetation/forest, certain questions arise.
- The authors mention permafrost in the title and refer to it throughout the article. However, key information is missing: what are the characteristics of permafrost in the study area (depth, thickness of the active layer, etc.) – is there a measuring station nearby that can be referred to? Are there any signs of permafrost degradation?
- From the included general map of the research site locations in Fig. 1, it is difficult to determine whether they are situated in areas of continuous, discontinuous, or sporadic permafrost. This is crucial for the topic under discussion, so a detailed map of the extent of permafrost (type) is necessary.
- The authors compare temperatures within plant areas classified as logged, dead and young, indicating that the area is inhabited by semi-nomadic herders who make their camps on south-facing slopes, with the closest camp found 700 metres south of the study area. Therefore, the following information is missing: what is the nature of pastoralism in the analysed area, what are the characteristics of the logged vegetation cover (what species of shrubs, grass), dead forest - what are the reasons for this? When did it die? Young forest - what is its age, etc.? This information should be presented in detail, as the article is supposed to concern the relationship between vegetation and GST.
- The authors only considered the impact of vegetation on Ground Surface Temperatures (GST). It is necessary to supplement this information with data on the abiotic environment, such as the type of mineral substrate and soil characteristics, including the amount of clay and rock in the soil (one of the key factors in heat transfer and conduction), as well as the slope of the terrain.
- As GST are mainly controlled by atmospheric factors and topographical effects It is crucial to examine and present the climatic conditions from the nearest meteorological station, not only for the analysed period of two years, but also for a longer period, showing possible changes in temperature and precipitation trends.
- A deeper discussion is needed on the relationship between exposure, altitude and location in relation to the main terrain features (close to the ridge, close to the valley floor), e.g. the lowest temperatures in winter and the highest in summer for logged sites seem obvious without the context of vegetation cover.
Citation: https://doi.org/10.5194/egusphere-2025-2366-RC2 - AC2: 'Reply on RC2', Robin B. Zweigel, 23 Oct 2025
- AC1: 'Reply on RC1', Robin B. Zweigel, 23 Oct 2025
-
EC1: 'Comment on egusphere-2025-2366', Krystyna Kozioł, 24 Nov 2025
Dear Robin B. Zweigel & Co-Authors,
in the interest of timely processing of your manuscript and the limited time available to Reviewer 1, who suggested the inclusion of quantitative analysis, I have undertaken the task of reviewing this part of the revised manuscript myself, while I would like to entrust the assessment of the manuscript revision in other matters, including the conclusions based on this analysis, to one of the Reviewers (Reviewer 2 in this case). I believe the inclusion of statistical analysis, in the form of correlation matrix, is a welcome addition and it allows to assess more comprehensively the impact of quantificable terrain characteristics on ground surface temperature parameters. While the general conclusions from obtained correlation coefficients appear justified, I would encourage the Authors to provide a more comprehensive description of the statistical method applied, without which the conclusions cannot be fully considered.
- Which type of correlation coefficient was applied (Pearson, Spearman rank)? Is it the correlation coefficient (r or rho) reported, or the coefficient of determination (r2)? My interpretation in the further comments is that the numbers provided are Pearson correlation coefficients (r) and therefore they reflect linear relationships between the selected variables.
- Were all variables subject to correlation analysis normally distributed? If not, were they subjected to any transformations, e.g. logarithmic transformation?
- What was the statistical significance of the correlation coefficients provided? How does it correspond to the value r=0.7? With as small a sample as n=10 (if I understand correctly), the significance criteria should be explained in detail.
- While it is understandable that the choice of a boundary value for strong correlation coefficients is somewhat arbitrary, how was the value r = 0.7 chosen? Why not describe if the general conclusions would hold with the inclusion of values, e.g. 0.7 > r > 0.6, or with the exclusion of values r < 0.75 or r < 0.8?
- Finally, while not strictly necessary to apply another method, it would be beneficial to justify the choice of correlation matrix as the statistical technique applied. Especially the assumption of linear relationships between the input data should be considered, since non-linear relationships would not be reflected well with the correlation coefficient (at least, if Pearson correlation was applied).
- It would also be beneficial to include both a brief statistical method and statistical result description in the main manuscript, and only expand on them in the supplement, so that all the necessary information for the conclusions drawn may be found in main text.
Citation: https://doi.org/10.5194/egusphere-2025-2366-EC1
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-2366', Anonymous Referee #1, 18 Jul 2025
Ground surface temperature (GST) is a crucial factor for permafrost research and predictive mapping. However, the current study requires substantial improvements in the following aspects:
Mechanistic Analysis
- It is necessary to establish quantitative relationships between various influencing factors and GST, and to calculate the contribution rate of each factor.
- Simply determining the dominant influencing factors through qualitative discussion is scientifically inadequate.
Permafrost Context
- GST studies alone cannot confirm whether the study area is located in a permafrost zone.
- We recommend: Incorporating existing permafrost data in the study area description. Providing a map clearly indicating the study area's position within the permafrost region
Line 40: Such disruptions? The above text does not mention such disruptions.
Lines 45-49: Many studies have also examined the changes in winter in the permafrost region.
For Introduction: It is essential to highlight the significance of ground surface temperatures (GSTs) research in the introduction, particularly in the context of permafrost. Studying GST can reveal its connection with permafrost, such as whether it can reflect the dynamic changes in permafrost conditions.
Line 101: What is A1?
Figures 1 and 2: The general title for all figures should be provided first, followed by individual captions for each subplot.
2.1 Site description: It is recommended to specify the permafrost classification characteristics and thermal regime parameters in the study area. For instance: Permafrost type: continuous/discontinuous/sporadic/isolated patches. Mean annual ground temperature (MAGT) range, active layer thickness, permafrost thickness.
Line 134: What are the start and end months of the hydrological year?
Line 185: December represents the early snow accumulation phase, while March marks the end of winter when snow begins to melt. An additional snow measurement should be conducted in February during the stable snow cover period.
Figure 3: The authors have two years of data, so seasonal and annual averages should be presented with error bars. These error bars could help reveal certain patterns in the data. The authors have not provided complete figure titles for any of the figures, only including subfigure captions.
Line 272: Table C2?
Lines 285-286: What’s the reason? Such a result seems counterintuitive.
Lines 291-296: This content has already been mentioned in the Introduction and Study Area sections, and its repetition here is redundant.
Lines 367-368: What is the reason? Why has summer disappeared?
Line 403: The interference may not necessarily lead to an increase in the MAGST, but it could still cause a significant rise in summer GST—it's just that the MAGST is offset by the decrease in winter GST.
Line 417: Why is that?
Lines 415-429: Different vegetation cover can lead to changes in wind speed, and does wind speed affect snow cover? I believe livestock trampling might have some impact, but it’s definitely not the main factor.
Line 433-434, Table C2: How do you determine whether livestock trampling is sparse or dense? By the number of hoofprints? But could there be cases where heavy trampling is later covered by snow?
Line 290：The authors should construct a relational diagram between all factors and GST (Ground Surface Temperature), calculate the contribution rate of each factor to GST, and then discuss which one is the dominant influencing factor. Otherwise, the discussion can only superficially address whether the factors affect GST and which one plays a major role. Due to the presence of anthropogenic factors such as livestock trampling, it is impossible to analyze GST differences caused solely by variations in vegetation cover.
Line 464: I believe such a conclusion is unreasonable. Livestock trampling compacts snow cover, reducing its thickness while increasing its density. However, this factor has not been quantified, nor has the influence of snow layers—another critical aspect—been quantitatively assessed.
Line 475: The understory vegetation in logged or dead forest plots is dense, resulting in a Plant Area Index (PAI) similar to that of intact forests. However, their effects on ground surface temperature (GST) still differ. Relying solely on PAI may fail to distinguish the varying impacts of different vegetation covers on GST.
Citation: https://doi.org/10.5194/egusphere-2025-2366-RC1 - AC3: 'Reply on RC1', Robin B. Zweigel, 23 Oct 2025
-
RC2: 'Comment on egusphere-2025-2366', Anonymous Referee #2, 24 Sep 2025
This study presents a new dataset on GST, vegetation density, and surface conditions. These data show how forest disturbances affect the forest-steppe ecotone in central Mongolia in the context of ground surface temperature fluctuations. This is an important issue that is rarely addressed in the literature. The problem of vegetation response to permafrost degradation, and conversely, the influence of vegetation type on ground temperature, remains unclear. Although the authors present interesting conclusions regarding the values of temperature extremes depending on whether we are dealing with dead, young or logged vegetation/forest, certain questions arise.
- The authors mention permafrost in the title and refer to it throughout the article. However, key information is missing: what are the characteristics of permafrost in the study area (depth, thickness of the active layer, etc.) – is there a measuring station nearby that can be referred to? Are there any signs of permafrost degradation?
- From the included general map of the research site locations in Fig. 1, it is difficult to determine whether they are situated in areas of continuous, discontinuous, or sporadic permafrost. This is crucial for the topic under discussion, so a detailed map of the extent of permafrost (type) is necessary.
- The authors compare temperatures within plant areas classified as logged, dead and young, indicating that the area is inhabited by semi-nomadic herders who make their camps on south-facing slopes, with the closest camp found 700 metres south of the study area. Therefore, the following information is missing: what is the nature of pastoralism in the analysed area, what are the characteristics of the logged vegetation cover (what species of shrubs, grass), dead forest - what are the reasons for this? When did it die? Young forest - what is its age, etc.? This information should be presented in detail, as the article is supposed to concern the relationship between vegetation and GST.
- The authors only considered the impact of vegetation on Ground Surface Temperatures (GST). It is necessary to supplement this information with data on the abiotic environment, such as the type of mineral substrate and soil characteristics, including the amount of clay and rock in the soil (one of the key factors in heat transfer and conduction), as well as the slope of the terrain.
- As GST are mainly controlled by atmospheric factors and topographical effects It is crucial to examine and present the climatic conditions from the nearest meteorological station, not only for the analysed period of two years, but also for a longer period, showing possible changes in temperature and precipitation trends.
- A deeper discussion is needed on the relationship between exposure, altitude and location in relation to the main terrain features (close to the ridge, close to the valley floor), e.g. the lowest temperatures in winter and the highest in summer for logged sites seem obvious without the context of vegetation cover.
Citation: https://doi.org/10.5194/egusphere-2025-2366-RC2 - AC2: 'Reply on RC2', Robin B. Zweigel, 23 Oct 2025
- AC1: 'Reply on RC1', Robin B. Zweigel, 23 Oct 2025
-
EC1: 'Comment on egusphere-2025-2366', Krystyna Kozioł, 24 Nov 2025
Dear Robin B. Zweigel & Co-Authors,
in the interest of timely processing of your manuscript and the limited time available to Reviewer 1, who suggested the inclusion of quantitative analysis, I have undertaken the task of reviewing this part of the revised manuscript myself, while I would like to entrust the assessment of the manuscript revision in other matters, including the conclusions based on this analysis, to one of the Reviewers (Reviewer 2 in this case). I believe the inclusion of statistical analysis, in the form of correlation matrix, is a welcome addition and it allows to assess more comprehensively the impact of quantificable terrain characteristics on ground surface temperature parameters. While the general conclusions from obtained correlation coefficients appear justified, I would encourage the Authors to provide a more comprehensive description of the statistical method applied, without which the conclusions cannot be fully considered.
- Which type of correlation coefficient was applied (Pearson, Spearman rank)? Is it the correlation coefficient (r or rho) reported, or the coefficient of determination (r2)? My interpretation in the further comments is that the numbers provided are Pearson correlation coefficients (r) and therefore they reflect linear relationships between the selected variables.
- Were all variables subject to correlation analysis normally distributed? If not, were they subjected to any transformations, e.g. logarithmic transformation?
- What was the statistical significance of the correlation coefficients provided? How does it correspond to the value r=0.7? With as small a sample as n=10 (if I understand correctly), the significance criteria should be explained in detail.
- While it is understandable that the choice of a boundary value for strong correlation coefficients is somewhat arbitrary, how was the value r = 0.7 chosen? Why not describe if the general conclusions would hold with the inclusion of values, e.g. 0.7 > r > 0.6, or with the exclusion of values r < 0.75 or r < 0.8?
- Finally, while not strictly necessary to apply another method, it would be beneficial to justify the choice of correlation matrix as the statistical technique applied. Especially the assumption of linear relationships between the input data should be considered, since non-linear relationships would not be reflected well with the correlation coefficient (at least, if Pearson correlation was applied).
- It would also be beneficial to include both a brief statistical method and statistical result description in the main manuscript, and only expand on them in the supplement, so that all the necessary information for the conclusions drawn may be found in main text.
Citation: https://doi.org/10.5194/egusphere-2025-2366-EC1
Peer review completion
Journal article(s) based on this preprint
Data sets
Research data for egusphere-2025-2366 Robin B. Zweigel https://doi.org/10.5281/zenodo.15470689
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 791 | 149 | 42 | 982 | 29 | 42 |
- HTML: 791
- PDF: 149
- XML: 42
- Total: 982
- BibTeX: 29
- EndNote: 42
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Dashtseren Avirmed
Khurelbaatar Temuujin
Clare Webster
Hanna Lee
Sebastian Westermann
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1360 KB) - Metadata XML
Ground surface temperature (GST) is a crucial factor for permafrost research and predictive mapping. However, the current study requires substantial improvements in the following aspects:
Mechanistic Analysis
Permafrost Context
Line 40: Such disruptions? The above text does not mention such disruptions.
Lines 45-49: Many studies have also examined the changes in winter in the permafrost region.
For Introduction: It is essential to highlight the significance of ground surface temperatures (GSTs) research in the introduction, particularly in the context of permafrost. Studying GST can reveal its connection with permafrost, such as whether it can reflect the dynamic changes in permafrost conditions.
Line 101: What is A1?
Figures 1 and 2: The general title for all figures should be provided first, followed by individual captions for each subplot.
2.1 Site description: It is recommended to specify the permafrost classification characteristics and thermal regime parameters in the study area. For instance: Permafrost type: continuous/discontinuous/sporadic/isolated patches. Mean annual ground temperature (MAGT) range, active layer thickness, permafrost thickness.
Line 134: What are the start and end months of the hydrological year?
Line 185: December represents the early snow accumulation phase, while March marks the end of winter when snow begins to melt. An additional snow measurement should be conducted in February during the stable snow cover period.
Figure 3: The authors have two years of data, so seasonal and annual averages should be presented with error bars. These error bars could help reveal certain patterns in the data. The authors have not provided complete figure titles for any of the figures, only including subfigure captions.
Line 272: Table C2?
Lines 285-286: What’s the reason? Such a result seems counterintuitive.
Lines 291-296: This content has already been mentioned in the Introduction and Study Area sections, and its repetition here is redundant.
Lines 367-368: What is the reason? Why has summer disappeared?
Line 403: The interference may not necessarily lead to an increase in the MAGST, but it could still cause a significant rise in summer GST—it's just that the MAGST is offset by the decrease in winter GST.
Line 417: Why is that?
Lines 415-429: Different vegetation cover can lead to changes in wind speed, and does wind speed affect snow cover? I believe livestock trampling might have some impact, but it’s definitely not the main factor.
Line 433-434, Table C2: How do you determine whether livestock trampling is sparse or dense? By the number of hoofprints? But could there be cases where heavy trampling is later covered by snow?
Line 290：The authors should construct a relational diagram between all factors and GST (Ground Surface Temperature), calculate the contribution rate of each factor to GST, and then discuss which one is the dominant influencing factor. Otherwise, the discussion can only superficially address whether the factors affect GST and which one plays a major role. Due to the presence of anthropogenic factors such as livestock trampling, it is impossible to analyze GST differences caused solely by variations in vegetation cover.
Line 464: I believe such a conclusion is unreasonable. Livestock trampling compacts snow cover, reducing its thickness while increasing its density. However, this factor has not been quantified, nor has the influence of snow layers—another critical aspect—been quantitatively assessed.
Line 475: The understory vegetation in logged or dead forest plots is dense, resulting in a Plant Area Index (PAI) similar to that of intact forests. However, their effects on ground surface temperature (GST) still differ. Relying solely on PAI may fail to distinguish the varying impacts of different vegetation covers on GST.