the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Feb 2025
Treeline species Betula ermanii are more adaptable to alpine environments than non-treeline species Picea jezoensis: evidence from leaf functional traits
Abstract. Understanding functional trait differences between treeline and non-treeline species is key to exploring their adaptive strategies under environmental stress and predicting subalpine forest dynamics. On Changbai Mountain, Betula ermanii dominates over 90 % of the treeline zone, while Picea jezoensis accounts for over 70 % of the lower elevation zone. It remains unclear whether P. jezoensis, a treeline genus elsewhere, would eventually shift upward and replace B. ermanii. We thus investigated leaf functional traits, their intraspecific variation, and inter-trait relationships for both species along the elevational gradient. B. ermanii exhibited higher LDMC, N, P, and gs, but lower WUE and δ18O at higher elevations, with the greatest intraspecific variability in photosynthetic and hydraulic traits, and tighter linkages among traits. In contrast, P. jezoensis exhibited an increase in δ13C and a decrease in SLA with elevation, accompanied by the greatest intraspecific variability in photosynthetic traits and weaker correlations among traits. Overall, B. ermanii employs a resource-acquisition strategy enabling it to occupy resources and space, while P. jezoensis adopts a resource-conserving strategy by emphasizing shade and drought-tolerance, resource conservation, and long-term adaptation at lower elevation, limiting its ability of upward range expansion. These findings enhance our understanding of their adaptive strategies and responses to elevational change, informing predictions of subalpine forest dynamics.
- Preprint
(1101 KB) - Metadata XML
-
Supplement
(126 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-369', Anonymous Referee #1, 31 Mar 2025
The manuscript has an important scientific significance by giving new information about functional traits and their inter- and intraspecific variation of two tree species in an elevation gradient in Northeast China: a deciduous treeline species Betula ermanii and a conifer tree species Picea jezoensis ( non-treeline species), and a discussion on the species adaptive strategies and ability of upward range expansion. However, there are some important issues to be fixed, regarding the analyses, explanations of the obtained results and result interpretation. After these have been fixed, the manuscript is of good quality to be published in EGUsphere, according to my opinion. Below are my detailed comments, which are also marked as comments to the pdf file attached:
L23-24: What is expected from these in relation to treeline environment?
L33-34: This was against expectations, right?
L36-37: Why these traits are expected to relate adaptation to lower elevation? I would expect an opposite effect, as high elevation environments are sometimes also dry, and resource conservation strategy may be beneficial there due to harsh conditions.
L61: Do you refer here inter- or intraspecific variation in traits? Or both?
L71-73: Isn't resource conservation strategy something that would be beneficial at the treeline due to harsh environmental, not the resource acquisition strategy?
L83-85: How about the evolutionary aspect here? Physiological traits, such as photosynthesis, is an old trait in evolutionary perspective, and is therefore expected to be rather fixed, i.e. not very plastic, due to the conservative inheritance of the trait caused by its complex genetic basis.
L121-123: Why we are expecting reduced leaf trait connectivity when the construction costs need to be minimized? Could you give a short clarification for this.
L147-148: How much of leaf trait differences are explained by the fact that the other sp is conifer and other is deciduous? A short discussion on this would be good to have.
L216-218_ What were used here as fixed and random factors?
L221: Can you explain here what Q3 and Q1 are?
L251: This is an unclear statement.
L263-264: Adaptation to higher elevation (colder or drier conditions)?
L266-267: The datapoints in each elevation are not independent, as they are from the same tree. Therefore, a simple linear regression is not appropriate here. Mean values per tree should be used, but this is also a problem as there is only 6 datapoints then. Thus, some other method than simple lm should be used. Maybe using a tree as a random factor (?)
L269-270: Can you give a little bit more explanation here, e.g. was the other analysis done combining all the data together (both species)?
L280-283: I am wondering how much of this variation is due to measurement challenges, as these traits, especially photosynthetic traits, are very sensible, and therefore their measurement usually need several repetitions? A discussion about this would be good to add.
L312: Surprising that there is no relationship between A and SLA, and only a weak relationship between A and gs in P. jezwoensis and no relationship at all in B. ermanii ? There is a discussion about this in the discussion section., but it should be extended and made more clear.
L313: This is confusing as the same color combination is used to separate the species.
L317: Does this figure give significantly more information than fig 3 and 4? If so, can you explain a bit more what does each point (a, b and c) mean?
L324-325: What does the radical water use strategy mean and how was that observed? This is discussed later on (in 4.2) a bit more, but if it is mentioned here it needs some explanation.
L328-329: But there was no connection between these traits, which is surprising (fig 4). This is discussed a bit more later on in 4.2, but same as above: if it is mentioned here, it needs some explanation.
L331-335: I think this is a bit controversial, as harsh treeline environments are not resource-rich. Can you explain better this idea?
L337-339: What kind of mechanisms is this based on? Can you explain better, so that the next sentence would be more justified.
L351-352: But on the other hand, the traits showed more intraspecific variation, indicating that there is potential.
L355-356: It may limit a rapid expansion, but on the other hand, it can better ensure survival in harsh conditions. This should be also discussed.
L362-363: Can you explain what is the reasoning behind this statement?
L364-365: Is this due to a stronger adaptation to a certain type of environment (cold /dry)?
L365: Is there something missing from this sentence?
L367-371: I think that this part is somehow aiming to give an explanation for my questions /comments (regarding e.g. the lines 324-325 and 328-329), but it could be a bit more clear, expanded and combined with the statements in the corresponding lines.
L374-376: I think that this statement is not so clear, especially what comes to photosynthesis, which is an old plant trait and thus expended to be quite fixed.
What do you mean by anisotropic traits?
L377-378: Maybe some of them, but photosynthesis is an old trait and therefore expected to be more fixed.
L382-384: This makes sense, indicating that this is thus more as a species-specific difference than environmental adaptation.
L397-399: This sentence seems incomplete, or something is missing.
L417-418: What do you mean by "the central traits"?
L421-427: Could you explain these interesting mechanisms a bit more. For example, why an increase in photosynthetic rate was not observed in B. ermanii when stomatal opening increased? How the independent regulation of water and carbon is expected to work?
L433-435: This was not tested with an experimental setting. So, it is difficult to claim that the differences were driven only by the environment and not by species-specific differences. And the difference in WUE seemed to be species-specific difference anyway. This aspect should not be forgotten from the discussion.
-
AC2: 'Reply on RC1', Hong He, 10 Jun 2025
Reviewer #1:
The manuscript has an important scientific significance by giving new information about functional traits and their inter- and intraspecific variation of two tree species in an elevation gradient in Northeast China: a deciduous treeline species Betula ermanii and a conifer tree species Picea jezoensis (non-treeline species), and a discussion on the species adaptive strategies and ability of upward range expansion. However, there are some important issues to be fixed, regarding the analyses, explanations of the obtained results and result interpretation. After these have been fixed, the manuscript is of good quality to be published in EGUsphere, according to my opinion. Below are my detailed comments, which are also marked as comments to the pdf file attached:
[Response]: We sincerely thank you for the positive evaluation of our manuscript and the constructive comments provided. We have carefully addressed each of the issues raised, especially regarding the analyses, explanation of results, and their ecological interpretation. Below, we provided the point-to-point responses to all your comments.1. L23-24: What is expected from these in relation to treeline environment?
[Response]: Thank you for your insightful question. We agree that the original sentence did not clearly articulate our expectations regarding trait patternsin relation to treeline environments. In the revised manuscript, we have clarified this point by adding a sentence “Usually, treeline species would exhibit more acquisitive traits that enhance adaptability to low temperature and short growing seasons, while non-treeline species would display more conservative traits adapted to relatively stable, lower-elevation environments.” (Lines 24–27).2. L33-34: This was against expectations, right?
[Response]: Thank you for this insightful observation. We agree that the observed combination of high intraspecific variability in photosynthetic traits and weaker trait coordination in Picea jezoensis appears to contrast with the general expectation that stressful environments promote tighter trait integration and reduces trait variation. In this case, P. jezoensis is a non-treeline species experiencing conditions beyond its optimal range at higher elevations. We interpret this pattern not as maladaptive, but rather as an alternative compensatory strategy—potentially allowing individuals to explore diverse trait combinations to maintain function under novel environmental stress. This flexibility may reflect an attempt to cope with light competition, cold temperatures, and environmental heterogeneity in the upper margin of its range.3. L36-37: Why these traits are expected to relate adaptation to lower elevation? I would expect an opposite effect, as high elevation environments are sometimes also dry, and resource conservation strategy may be beneficial there due to harsh conditions.
[Response]: Thank you for raising this important point. We agree that resource-conserving strategies, including traits related to drought tolerance and water use efficiency, can be beneficial in high-elevation environments where water limitations exist. However, in our study, on Changbai Mountain, the treeline zone is characterized by plentiful precipitation, low temperatures, strong radiation, and short growing seasons, rather than drought stress. In contrast, the lower elevation zones where P. jezoensis dominates are characterized by denser canopies, lower light availability, and more stable microclimates.
The resource-conserving traits of P. jezoensis, such as lower SLA, higher δ¹³C, and relatively limited trait integration, align more closely with adaptation to shade, slower growth, and long-term structural investment—traits advantageous in stable, low-light environments. These traits may restrict its ability to cope with the high-light, low-temperature conditions near the treeline, limiting its potential for upward expansion.4. L61: Do you refer here inter- or intraspecific variation in traits? Or both?
[Response]: Thank you for your helpful comment. In this sentence, we refer to both interspecific and intraspecific variation in functional traits, as both levels of variation are shaped by environmental gradients such as elevation. To clarify this, we have revised the sentence in the manuscript to explicitly state that environmental variation can influence trait means between species (interspecific variation) as well as trait variability within species (intraspecific variation) (Lines 67–68).5. L71-73: Isn't resource conservation strategy something that would be beneficial at the treeline due to harsh environmental, not the resource acquisition strategy?
[Response]: Thank you for this important comment. You are quite right that, in general, resource-conservation strategies are often thought to be more beneficial under harsh conditions such as those founded at treelines. However, recent studies have shown that under some high-elevation conditions—especially where light is not limiting and the growing season is short—species that adopt a rapid resource acquisition strategy can gain a competitive advantage by maximizing resource use during limited favorable periods (Reich, 2014; Liao et al., 2021). In our study, B. ermanii exhibits such a strategy, with higher leaf N and P content and stomatal conductance at higher elevations, allowing for rapid carbon assimilation. This trait profile may support its dominance and potential upward expansion under warming scenarios.6. L83-85: How about the evolutionary aspect here? Physiological traits, such as photosynthesis, is an old trait in evolutionary perspective, and is therefore expected to be rather fixed, i.e. not very plastic, due to the conservative inheritance of the trait caused by its complex genetic basis.
[Response]: Thank you for raising this important evolutionary perspective. We agree that physiological traits such as photosynthesis are highly conserved over evolutionary timescales due to their fundamental role and complex genetic control. However, in this context, we were referring to ecological plasticity—i.e., the short-term, environmentally induced variation within species—rather than evolutionary lability. To clarify this distinction, we have revised the sentence to indicate that physiological traits may exhibit higher phenotypic plasticity at ecological timescales, even if they are evolutionarily conserved. The revised text now was added as "Although physiological traits such as photosynthesis are evolutionarily conserved due to complex genetic constraints, they can still exhibit high phenotypic plasticity at ecological timescales because of their reversible and low-cost adjustments (Grime et al., 2002)." (Lines 90–93)7. L121-123: Why we are expecting reduced leaf trait connectivity when the construction costs need to be minimized? Could you give a short clarification for this.
[Response]: Thank you for this thoughtful question. The expectation of reduced trait connectivity under harsh conditions is based on the idea that strong coordination among traits requires more developmental and physiological integration, which can be costly to maintain. In highly stressful environments such as treelines, plants may prioritize flexibility and survival by decoupling trait relationships, allowing independent adjustment of traits to minimize energy and resource expenditure. We have clarified this rationale in the revised text (Lines 135–138).8. L147-148: How much of leaf trait differences are explained by the fact that the other sp is conifer and other is deciduous? A short discussion on this would be good to have.
[Response]: Thank you for this important comment. We agree that the distinction between coniferous and deciduous leaf types may partly contribute to the observed differences in leaf functional traits. However, our study focused on intraspecific trait variation and trait coordination along elevational gradients within each species, rather than direct trait comparisons between functional types. In the revised manuscript, we have now included a short discussion acknowledging that some trait differences—such as SLA, LDMC, and photosynthetic rates—may inherently reflect differences between evergreen conifers and deciduous broadleaved trees. Nonetheless, we argue that the contrasting patterns in intraspecific variability and trait integration are primarily driven by environmental adaptation along the elevational gradient, as similar trends have also been observed within comparable functional groups in other studies (He et al., 2020; Rao et al., 2022). We have added this clarification in the Discussion (Lines 126–135).9. L216-218： What were used here as fixed and random factors?
[Response]: Thank you for your question. In our linear mixed models (LMMs), elevation, species, and their interaction (elevation × species) were used as fixed effects, and the individual tree was used as a random effect. We have clarified this model structure in the revised Methods section (Lines 250–254).10. L221: Can you explain here what Q3 and Q1 are?
[Response]: Thank you for your question. We have revised the text to define Q3 and Q1. Specifically, Q3 refers to the third quartile (the 75th percentile) and Q1 refers to the first quartile (the 25th percentile) of the trait distribution. The QCD formula thus represents a normalized measure of dispersion that is less sensitive to extreme values compared to the coefficient of variation. This clarification has been added to the revised Methods section (Lines 258–259).11. L251: This is an unclear statement.
[Response]: Thank you for your comment. We agree that the original sentence was unclear. We have revised it to more precisely describe the pattern in the PCA plot for P. jezoensis. The sentence now was revised as: "The individuals of P. jezoensis were distributed broadly along the PCA space, with no distinct clustering by elevation, suggesting that leaf trait variation in this species is less structured by elevation compared to B. ermanii (Figure 1b)." (Lines 288–291).
This revision clarified that P. jezoensis showed little elevational differentiation in trait composition based on PCA axes.12. L263-264: Adaptation to higher elevation (colder or drier conditions)?
[Response]: Thank you for your comment. We have clarified the ecological interpretation of these trait shifts. Specifically, for B. ermanii, the decrease in WUE and δ¹⁸O with elevation likely reflects a more liberal water-use strategy under colder but not necessarily drier conditions, as high-elevation sites on Changbai Mountain are characterized by low temperature but sufficient moisture. In contrast, for P. jezoensis, the increase in δ¹³C and decrease in SLA with elevation suggest enhanced water-use efficiency and thicker leaves, which are typical of adaptations to increased evaporative demand or lower temperatures.13. L266-267: The datapoints in each elevation are not independent, as they are from the same tree. Therefore, a simple linear regression is not appropriate here. Mean values per tree should be used, but this is also a problem as there is only 6 datapoints then. Thus, some other method than simple lm should be used. Maybe using a tree as a random factor (?)
[Response]: Thank you for the insightful comment. We agree that statistical analyses should account for the non-independence of data points from the same tree. In fact, we have already addressed this issue in Table 2, where we used linear mixed-effects models (LMMs) with tree identity as a random factor to evaluate the effects of elevation, species, and their interaction.
Figure 2 serves as a complementary visual summary of these trait–elevation relationships. The lines shown are not intended as formal linear models but as visual aids to highlight general trends, with statistical inference provided in Table 2. We have clarified this purpose in the revised figure caption and Methods section (lines 306-309).14. L269-270: Can you give a little bit more explanation here, e.g. was the other analysis done combining all the data together (both species)?
[Response]: Thank you for the comment. Yes, the linear mixed-effects model presented in Table 2 was performed by combining all the data across both species and elevations into a single model. This allowed us to assess the main effects of elevation, species, and their interaction (elevation × species) on each trait. One tree was included as a random effect to account for among-tree replication. We have clarified this in the revised caption of Table 2 and in the Methods section (Lines 310–316).15. L280-283: I am wondering how much of this variation is due to measurement challenges, as these traits, especially photosynthetic traits, are very sensible, and therefore their measurement usually need several repetitions? A discussion about this would be good to add.
[Response]: Thank you for raising this important point. We fully acknowledge that photosynthetic traits are generally sensitive to environmental fluctuations and may be prone to measurement variability. However, in our study, we addressed this issue by performing multiple replicate measurements per tree under carefully standardized conditions (e.g., fixed time frame, stable weather, and instrument calibration). Only stable and consistent readings were recorded for analysis. Therefore, we are confident that the observed intraspecific variation in photosynthetic traits reflects true biological differences rather than measurement artifacts (Lines 212–214).16. L312: Surprising that there is no relationship between A and SLA, and only a weak relationship between A and gs in P. jezwoensis and no relationship at all in B. ermanii ? There is a discussion about this in the discussion section., but it should be extended and made more clear.
[Response]: Thank you for this insightful observation. We agree that the lack of significant correlations between A and SLA or gs, particularly in B. ermanii, is unexpected given their commonly reported positive associations. We have expanded the discussion on this point. Specifically, we suggest that these decoupled relationships may reflect species-specific trait regulation strategies under environmental stress. For B. ermanii, strong coordination between nutrient-use traits (e.g., PNUE, PPUE) and A may override the influence of SLA or gs. In P. jezoensis, trait independence may reflect a modular response strategy where traits function more independently under competition and shade. These patterns underscore that trait coordination can vary not only across environments but also across species with different ecological strategies. This expanded explanation is now included in Discussion 4.3 (Lines 521–535).17. L313: This is confusing as the same color combination is used to separate the species.
[Response]: Thank you for pointing this out. To eliminate any confusion, we have revised the figure by using pink and green lines to represent positive and negative trait correlations, respectively. The updated figure caption now clearly explains the revised color coding (Figure 4).18. L317: Does this figure give significantly more information than fig 3 and 4? If so, can you explain a bit more what does each point (a, b and c) mean?
[Response]: Thank you for this comment. Figure 5 complements Figure 4 by quantifying the topological importance of individual traits in the network, rather than only showing their connections. Specifically, Figure 5a shows the degree centrality (number of connections) for each trait, 5b shows centrality closeness, reflecting how efficiently a trait is connected to all others in the network, and 5c shows betweenness centrality, indicating the extent to which a trait acts as a bridge in the network.
These metrics help identify hub traits and differentiate between the trait coordination structures of B. ermanii and P. jezoensis. We have clarified these points in the revised figure caption and discussion (Lines 370–376).19. L324-325: What does the radical water use strategy mean and how was that observed? This is discussed later on (in 4.2) a bit more, but if it is mentioned here it needs some explanation.
[Response]: Thank you for pointing this out. We agree that the term “radical water use strategy” requires clarification at its first mention. In the revised manuscript, we have explained that radical water use strategy refers to a more liberal water-use behavior, characterized by increased stomatal conductance (gs) and decreased water use efficiency (WUE) at higher elevations in B. ermanii. These changes suggest a strategy of prioritizing rapid gas exchange over water conservation, which may be advantageous during short alpine growing seasons. We have added this clarification in Section 4.1 (Lines 394–396).20. L328-329: But there was no connection between these traits, which is surprising (fig 4). This is discussed a bit more later on in 4.2, but same as above: if it is mentioned here, it needs some explanation.
[Response]: Thank you for pointing this out. We agree that the lack of observed connections between these traits in the network analysis (Figure 4) makes the interpretation in 4.1 potentially unclear. In the revised text, we have added a short clarification noting that although physiological linkages are expected between traits like gs, WUE, and A, the trait network showed weak or no correlations—possibly due to environmentally decoupled regulation or alternative coordination pathways (e.g., via nutrient-use traits like PNUE). This clarification now has been added alongside the original statement in Section 4.1 (Lines 390–392).21. L331-335: I think this is a bit controversial, as harsh treeline environments are not resource-rich. Can you explain better this idea?
[Response]: Thank you for this important observation. We agree that treeline environments are not resource-rich in the classical sense. We have revised the paragraph to clarify that B. ermanii’s “fast” strategy refers not to resource richness, but rather to its ability to rapidly acquire and utilize available resources during short, favorable periods, especially in a seasonally constrained and cold environment. This strategy enables it to take full advantage of transient growth opportunities. We have adjusted the texts in Section 4.1 to better reflect this interpretation (Lines 398–401).22. L337-339: What kind of mechanisms is this based on? Can you explain better, so that the next sentence would be more justified.
[Response]: Thank you for the comment. We agree that this statement needed further justification. We have clarified that the advantage of the fast resource-acquisition strategy under climate warming is based on several mechanisms:
(1) Warming extends the growing season, allowing B. ermanii to make better use of its rapid growth capacity;
(2) Increased temperature improves photosynthetic efficiency and nutrient uptake, both of which are aligned with B. ermanii's trait profile (e.g., high N, gs, PNUE);
(3) This strategy enables quick colonization of newly available niches above the current treeline, especially where competitors are slow-growing;
We have revised the text in Section 4.1 to reflect these points (Lines 404–409).23. L351-352: But on the other hand, the traits showed more intraspecific variation, indicating that there is potential.
[Response]: Thank you for this important comment. We agree that P. jezoensis exhibits a seemingly contrasting combination: a conservative water-use strategy (higher δ¹³C) alongside high intraspecific variation in photosynthetic traits. We interpret this as reflecting a modular response strategy, where core traits like δ¹³C are constrained by long-term adaptation, while other traits such as A, gs, or PNUE remain more plastic and responsive to microenvironmental variation. This decoupling may allow P. jezoensis to maintain a conservative baseline strategy while retaining adaptive flexibility. We have added this clarification to the revised Discussion (Lines 482–489).24. L355-356: It may limit a rapid expansion, but on the other hand, it can better ensure survival in harsh conditions. This should be also discussed.
[Response]: Thank you for this thoughtful comment. We fully agree that conservative trait strategies, while potentially limiting rapid range expansion, can also enhance survival and long-term persistence under harsh environmental conditions. We have added a sentence to the Discussion (Section 4.2) to reflect this dual role: P. jezoensis’s strategy may hinder rapid colonization but could buffer against environmental extremes and resource fluctuations, thereby ensuring stability in marginal habitats. This addition strengthens the balance of our interpretation (Lines 426–430).25. L362-363: Can you explain what is the reasoning behind this statement?
[Response]: Thank you for your question. We have clarified the reasons in the revised text. In harsher environments such as treeline areas, trait expression is more strongly affected by abiotic filtering, which can reduce phenotypic plasticity and allowable variation. Additionally, the cost of expressing highly plastic or deviant traits increases under stress, leading to trait convergence around stress-tolerant optima. We have added this explanation in Section 4.2 (Lines 444–445).26. L364-365: Is this due to a stronger adaptation to a certain type of environment (cold /dry)?
[Response]: Thank you for your question. Yes, we agree that reduced intraspecific variation in harsh environments likely reflects stronger directional selection and environmental filtering, leading to trait convergence around adaptive optima. In the case of B. ermanii, this is most likely associated with adaptation to cold and short-season alpine conditions, where trait combinations that maximize stress tolerance and survival are favored. We have added this clarification in Section 4.2 (Lines 446–449).27. L365: Is there something missing from this sentence?
[Response]: Thank you for pointing this out. We agree that the original sentence lacked an intermediate explanation. In the revised version, we have clarified that higher construction costs constrain the range of viable trait expressions, as producing and maintaining divergent phenotypes becomes more costly under stress. This leads to reduced trait variability through environmental filtering and selection for conservative strategies. The revised sentence now better explains the link between trait cost and intraspecific variation (Lines 446–449).28. L367-371: I think that this part is somehow aiming to give an explanation for my questions /comments (regarding e.g. the lines 324-325 and 328-329), but it could be a bit more clear, expanded and combined with the statements in the corresponding lines.
[Response]: Thank you very much for this insightful comment. We agree that the paragraph on climate filtering and trait variation (Lines 367–371) provides relevant context that should be more directly connected to earlier points regarding the water-use strategy (Lines 324–325) and trait linkages (Lines 328–329). We have revised the text to clarify that while B. ermanii shows lower intraspecific variation, its traits are more tightly coordinated in response to the harsh alpine environment. This combination reflects a stress-constrained yet functionally integrated strategy, where strong coordination among key traits (e.g., PNUE, PPUE, gs) supports rapid resource use during brief growing periods. We have added a clarification to bridge these points across Sections 4.2 (Lines 452–459).29. L374-376: I think that this statement is not so clear, especially what comes to photosynthesis, which is an old plant trait and thus expended to be quite fixed.
[Response]: Thank you for this important comment. We agree that photosynthesis, as a fundamental and evolutionarily conserved trait, is often considered genetically constrained. However, while the core biochemical processes of photosynthesis are conserved, many related traits we measured—such as stomatal conductance (gs), photosynthetic rate (A), and nutrient use efficiencies (PNUE, PPUE)—reflect ecophysiological responses to environmental variability and are influenced by microclimatic and nutrient conditions. These traits can exhibit substantial plasticity even within species.30. What do you mean by anisotropic traits?
[Response]: Thank you for pointing this out. We acknowledge that the term “anisotropic traits” is not appropriate in this ecological context and may be misleading. What we intended to refer to were structural or morphological traits, such as SLA or LDMC, which tend to have higher construction costs and are less reversible, and therefore typically exhibit lower phenotypic plasticity compared to physiological traits. We have revised the sentence accordingly to improve clarity (Lines 462–467).31. L377-378: Maybe some of them, but photosynthesis is an old trait and therefore expected to be more fixed.
[Response]: Thank you for this clarification. We agree that the core biochemical mechanisms of photosynthesis are highly conserved and not easily plastic. In our context, we referred specifically to physiological traits that modulate photosynthetic performance, such as stomatal conductance (gs), nutrient use efficiencies (PNUE, PPUE), and instantaneous photosynthetic rate (A). These traits, although related to photosynthesis, are more environmentally responsive and reversible, and thus fall into the more plastic category.32. L382-384: This makes sense, indicating that this is thus more as a species-specific difference than environmental adaptation.
[Response]: Thank you for your insightful comment. We agree that the observed differences in which trait groups show higher variation—hydraulic traits in B. ermanii and nutrient traits in P. jezoensis—likely reflect not only environmental influences but also species-specific functional strategies and evolutionary constraints. We have revised the discussion to acknowledge this interpretation, clarifying that both environmental adaptation and inherent species differences likely contribute to these trait variation patterns (Lines 472–475).33. L397-399: This sentence seems incomplete, or something is missing.
[Response]: Thank you for pointing this out. We have revised it to specify what the observed covariation and coordination patterns suggest about species adaptation strategies. The updated sentence now highlights that the differences in trait coordination between B. ermanii and P. jezoensis reflect contrasting strategies for dealing with environmental stress in their respective habitats (Lines 497–499).34. L417-418: What do you mean by "the central traits"?
[Response]: Thank you for your question. In trait network, “central traits” refer to those with high network centrality metrics, such as the degree, closeness, and betweenness, which indicate their importance in connecting and coordinating other traits. In our results (Figure 5), these central traits were primarily photosynthetic traits (e.g., PNUE, PPUE, A), which consistently showed the highest centrality scores in both species. We have clarified this in the revised text and figure caption (Lines 370–376).35. L421-427: Could you explain these interesting mechanisms a bit more. For example, why an increase in photosynthetic rate was not observed in B. ermanii when stomatal opening increased? How the independent regulation of water and carbon is expected to work?
[Response]: Thank you for this insightful comment. We have expanded our explanation of these physiological mechanisms. In B. ermanii, increased stomatal conductance (gs) might enhance CO₂ uptake capacity, but photosynthetic rate (A) might not increase proportionally due to biochemical limitations, such as Rubisco activity or nutrient availability, or feedback regulation under stress.
The negative correlation between gs and PNUE/PPUE suggests that when stomata open wider, photosynthetic N and P use efficiency decreases—possibly because more nutrients are required to maintain the same photosynthetic gain, reflecting a shift toward a resource-intensive strategy under alpine stress.
Similarly, the negative δ¹³C–δ¹⁸O correlation indicates that water-use efficiency (δ¹³C) and stomatal water loss (δ¹⁸O) may be regulated by different pathways or trade-offs, allowing B. ermanii to decouple water and carbon exchange. This might be an adaptive response to rapidly fluctuating alpine conditions. These clarifications have been added to the revised Discussion (Lines 522–542).36. L433-435: This was not tested with an experimental setting. So, it is difficult to claim that the differences were driven only by the environment and not by species-specific differences. And the difference in WUE seemed to be species-specific difference anyway. This aspect should not be forgotten from the discussion.
[Response]: Thank you for this important comment. We agree that our study, being observational along an elevational gradient, cannot fully disentangle environmental drivers from species-specific effects. While the positive correlation between δ¹³C and δ¹⁸O in P. jezoensis may suggest a coordinated response in more favorable conditions, we acknowledge that this pattern could also reflect inherent species-level physiological strategies.
In particular, the overall differences in WUE appeared to be largely species-driven, independent of elevation. We have revised the discussion to clarify that both environmental adaptation and species-specific trait syndromes likely contributed to the observed patterns (Lines 550–559).
Grime, P., J., Mackey, and L., J. M.: The role of plasticity in resource capture by plants, Evolutionary Ecology, 16, 299-307, 10.1023/A:1019640813676, 2002.
He, N., Li, Y., Liu, C., Xu, L., Li, M., Zhang, J., He, J., Tang, Z., Han, X., Ye, Q., Xiao, C., Yu, Q., Liu, S., Sun, W., Niu, S., Li, S., Sack, L., and Yu, G.: Plant Trait Networks: Improved Resolution of the Dimensionality of Adaptation, Trends in Ecology & Evolution, 35, 908-918, https://doi.org/10.1016/j.tree.2020.06.003, 2020.
Liao, H., Pal, R. W., Niinemets, Ü., Bahn, M., Cerabolini, B. E. L., and Peng, S.: Different functional characteristics can explain different dimensions of plant invasion success, Journal of Ecology, 109, 1524-1536, https://doi.org/10.1111/1365-2745.13575, 2021.
Rao, Q., Su, H., Ruan, L., Xia, W., Deng, X., Wang, L., Xu, P., Shen, H., Chen, J., and Xie, P.: Phosphorus enrichment affects trait network topologies and the growth of submerged macrophytes, Environmental Pollution, 292, 118331, https://doi.org/10.1016/j.envpol.2021.118331, 2022.
Reich, P. B.: The world‐wide ‘fast–slow’plant economics spectrum: a traits manifesto, Journal of ecology, 102, 275-301, 2014.
-
AC2: 'Reply on RC1', Hong He, 10 Jun 2025
-
RC2: 'Comment on egusphere-2025-369', Anonymous Referee #2, 23 Apr 2025
This study compares the leaf functional traits of Betula ermanii, a treeline species and Picea jezoensis, a non-treeline species, along an elevational gradient on Changbai Mountain to understand their adaptive strategies. B. ermanii exhibits traits favoring resource acquisition and adaptability at higher elevations, while P. jezoensis shows a resource-conserving strategy more suited to lower elevations.
I have a fundamental concern regarding the conceptual approach of comparing only two species in relation to the treeline. While the study aims to draw broader conclusions about vegetation stratification, analyzing just two species—each with distinct ecological strategies—limits the robustness of the findings. This approach risks confounding species-specific differences with general patterns, making it difficult to extrapolate broader ecological insights. Additionally, the manuscript often lacks sufficient detail and clarity, and contains multiple errors that hinder comprehension. It would benefit greatly from thorough proofreading and substantial rewriting to improve both readability and logical flow. I also have reservations about aspects of the data analysis and figure construction, which are at times unclear or insufficiently justified. These elements should be revised to enhance clarity and relevance.
Below are my specific comments:
Major issues:
- I have a conceptual concern with the study’s approach. The manuscript aims to explain vegetation stratification through functional traits, yet it compares two species that naturally occupy different elevations and likely have inherently different physiological thresholds. The comparison is further complicated by the contrasting functional types—Betula ermanii being a deciduous broadleaf and Picea jezoensis a conifer—each with fundamentally different life strategies. Moreover, the claim that P. jezoensis inhabits a resource-rich and less stressful environment is not well supported by empirical data. Since this species is at its own altitudinal limit, the level of stress it experiences is likely comparable—in relative terms—to that faced by B. ermanii at the treeline. Elevational limits represent ecological thresholds for both species, and interpreting lower elevation as inherently less stressful may overlook species-specific constraints. For instance, at comparable altitudes, the two species show markedly different trait values, and their trait responses to elevation differ substantially. These factors cast doubt on the validity of directly comparing their functional traits, and ultimately complicate the interpretation of adaptive strategies or elevation-driven processes. In my view, a more informative approach might have been to compare two co-occurring treeline species—or two non-treeline species—to identify shared adaptive strategies within each group.
- While I’m not a specialist in functional traits, I wonder if the analysis could be made more accessible by reducing redundancy among highly correlated traits—especially among leaf traits, which often show strong interdependence. Simplifying the dataset by selecting a representative trait from each highly correlated set could help clarify the figures and make them easier to interpret. Similarly, for trait group analyses (e.g., hydraulic, foliar), it might be useful to present either a single representative trait per group or an average value per group with associated uncertainty. This could streamline the presentation and highlight broader patterns more effectively.
- In the abstract you say that P. jezoensis is a treeline genus elsewhere. Does that mean that elsewhere it's on the treeline but not on your site? If so, it calls into question your study comparing a treeline and a non-treeline species.
- The manuscript requires careful proofreading, as it contains numerous typos, errors, and repeated phrases (e.g., lines 87, 97, 121, 160, 225, 237). The introduction lacks clarity and does not effectively communicate the main objectives or ecological significance of the study. Strengthening the justification for the research questions and explicitly outlining their broader relevance would greatly enhance this section. Figure legends are overly succinct and should be expanded to allow readers to interpret the figures without referring back to the main text. In the Methods section, essential parameters related to the trait network analysis—such as modularity and closeness—are not defined. While some of these are included in the supplementary materials, they should be clearly explained and contextualized within the main text. Additionally, the ecological relevance of these network metrics should be supported with appropriate references and further discussed in the Discussion section to better integrate them into ecological theory. The Results section would benefit from more explicit descriptions of key findings, including numerical values drawn from the figures to improve interpretability. The Discussion, meanwhile, needs editing for clarity and coherence, and should include a section addressing the limitations and potential weaknesses of the study’s approach. Finally, the authors should place greater emphasis on the broader ecological or practical implications of their findings to strengthen the study's impact.
- Several times throughout the manuscript—particularly in the Discussion—you suggest that conifers, such as Picea jezoensis, are less adapted for upward range shifts. However, I find this conclusion somewhat overreaching, given that it is based solely on a limited set of adult physiological traits. Key processes related to establishment, such as seed production, dispersal mechanisms, germination success, and seedling survival, are not addressed in the study. Without considering these life history stages, it is difficult to draw robust conclusions about a species' capacity to expand its range in response to climate change. I would recommend framing these interpretations more cautiously or supporting them with broader evidence.
- Your results highlight two distinct adaptive strategies—resource acquisition and resource conservation—across the two species studied. However, given the complexity and number of traits analyzed, it can be challenging to synthesize the findings as a whole. I suggest including a summary figure or diagram in the Discussion section that visually distills the main results for each species and clearly contrasts their respective strategies. This would greatly enhance the reader’s ability to grasp the overarching patterns and takeaways of the study.
Minor issues:
- it's best to avoid using acronyms in the abstract
- L84: ref ?
- L87: repetition of “functional traits”
- L96-99: lots of repetitions, improve the sentances
- L107-108: You say there are relatively few studies, but you don't cite any. Are you sure that no study has looked at the dynamics and physiological mechanisms of non-treeline species in response to climate change?
- L113: what do you mean by “integration”?
- L134: ‘a mean temperature of -7.3 to 4.9 °C in the growing season and annual precipitation of 800 to 1800 mm.’ Why do you give two figures each time? are they averages? quantiles? please specify.
- L140: You say B. ermanii is a treeline species, do you have a ref to support this?
- L142-145: Over what period were these figures calculated and what are the values for your sites?
- In methods (2.1.), could you specify if the area is grazed or mown. If yes is it still the case? And if the forest is managed? These informations are important to understand the structure and the responses to climate change.
- I think we're missing a study site map showing the sites sampled by species and their elevation. It is also in this figure that you could put the temperature values of the environments of each species (S1). You could add pictures of the two species.
- Figure 1: ‘environment’ corresponds to elevation ? the term needs to be changed. What's more, we don't know what altitude range it corresponds to, e.g. 1700-1800?
- I wonder if you run a PCA with the two species together, would that give two clusters corresponding to each species? Or would you still find altitudinal clusters?
- Table 2: What are F and p? You need to explain. What does ‘Elevation*species’ mean? It needs to be explained much better.
- Figure 2: equations and pvalues are displayed only for significant regressions ? specify.
- Figure 3: please add dotted ablines for certain % (25, 50, 75) to facilitate reading (panels a and b). For panel c, it's an average QCD per trait group? we don't understand what you've done here, please explain.
- Figure 4: I don't see the advantage of panels a and b over PCA. Justify it to me or put it in sup mat.
- Parameters of figures 4 and 5 need to be explained and ecological significances detailed and justified in methods.
- S1 : “DBH” meaning ? same, “High” what is high? Is it soil/air temperature? Specify.
- L368-369: ref ?
- Could you detail in methods the ecology of the two species: pioneer species? Place in the ecological succession, mode of dispersal.
- L420-432: not just a difference between conifer and broadleaved species?
- L434-435: “contrasting growth environments” in terms of what? Explain.
Citation: https://doi.org/10.5194/egusphere-2025-369-RC2 -
AC1: 'Reply on RC2', Hong He, 10 Jun 2025
### Reviewer #2:
This study compares the leaf functional traits of Betula ermanii, a treeline species and Picea jezoensis, a non-treeline species, along an elevational gradient on Changbai Mountain to understand their adaptive strategies. B. ermanii exhibits traits favoring resource acquisition and adaptability at higher elevations, while P. jezoensis shows a resource-conserving strategy more suited to lower elevations.I have a fundamental concern regarding the conceptual approach of comparing only two species in relation to the treeline. While the study aims to draw broader conclusions about vegetation stratification, analyzing just two species—each with distinct ecological strategies—limits the robustness of the findings. This approach risks confounding species-specific differences with general patterns, making it difficult to extrapolate broader ecological insights. Additionally, the manuscript often lacks sufficient detail and clarity, and contains multiple errors that hinder comprehension. It would benefit greatly from thorough proofreading and substantial rewriting to improve both readability and logical flow. I also have reservations about aspects of the data analysis and figure construction, which are at times unclear or insufficiently justified. These elements should be revised to enhance clarity and relevance.[Response]: We sincerely thank you for the thoughtful and constructive comments. We fully acknowledge the limitations of comparing only two species, each with distinct ecological strategies. Our goal was not to generalize treeline dynamics across all taxa, but to provide a focused, hypothesis-driven comparison between a dominant deciduous treeline species (B. ermanii) and a representative non-treeline conifer (P. jezoensis) that overlaps with the upper forest zone. We have revised the Introduction and Discussion to more clearly define the scope of inference and highlight the species-specific context of our findings.
We also appreciate the reviewer’s feedback on clarity and presentation. In response, we have carefully revised the manuscript for improved logical flow, language precision, and structural clarity. The figures and data analysis sections have been reorganized and more thoroughly explained, with clearer legends and justifications for the methods used.
Major issues:
1. I have a conceptual concern with the study’s approach. The manuscript aims to explain vegetation stratification through functional traits, yet it compares two species that naturally occupy different elevations and likely have inherently different physiological thresholds. The comparison is further complicated by the contrasting functional types—Betula ermanii being a deciduous broadleaf and Picea jezoensis a conifer—each with fundamentally different life strategies. Moreover, the claim that P. jezoensis inhabits a resource-rich and less stressful environment is not well supported by empirical data. Since this species is at its own altitudinal limit, the level of stress it experiences is likely comparable—in relative terms—to that faced by B. ermanii at the treeline. Elevational limits represent ecological thresholds for both species, and interpreting lower elevation as inherently less stressful may overlook species-specific constraints. For instance, at comparable altitudes, the two species show markedly different trait values, and their trait responses to elevation differ substantially. These factors cast doubt on the validity of directly comparing their functional traits, and ultimately complicate the interpretation of adaptive strategies or elevation-driven processes. In my view, a more informative approach might have been to compare two co-occurring treeline species—or two non-treeline species—to identify shared adaptive strategies within each group.
1. [Response]: We appreciate your insightful comment regarding the complexity of comparing a deciduous broadleaf species (B. ermanii) and a coniferous evergreen species (P. jezoensis). We agree that each species likely experiences environmental stress near its own elevational limit, and that these constraints may not be strictly hierarchical.
To address this concern, we have made the following clarifications in the revised manuscript:
(1) Both species are evaluated within their primary elevational ranges, where they are well established but approaching their current distributional limits.
(2) The interpretation of stress is therefore made relative to each species’ adaptive niche, rather than assuming lower elevations are universally less stressful.
(3) We acknowledge the inherent differences in functional types and strategies between conifers and deciduous broadleaves and emphasize that our analysis focuses on intraspecific trait responses and coordination patterns across elevation, rather than direct trait-by-trait comparisons.
These clarifications are now included in the Introduction (Lines 126–135), to better contextualize the adaptive significance of trait variability and coordination within each species.2. While I’m not a specialist in functional traits, I wonder if the analysis could be made more accessible by reducing redundancy among highly correlated traits—especially among leaf traits, which often show strong interdependence. Simplifying the dataset by selecting a representative trait from each highly correlated set could help clarify the figures and make them easier to interpret. Similarly, for trait group analyses (e.g., hydraulic, foliar), it might be useful to present either a single representative trait per group or an average value per group with associated uncertainty. This could streamline the presentation and highlight broader patterns more effectively.
[Response]: We thank you for this insightful comment. Indeed, functional traits—particularly within the same functional group—often exhibit high correlations. However, in our study, we intentionally retained all measured traits in the network and trait variation analyses to capture the full spectrum of trait covariation and integration. This comprehensive approach is important for several reasons:
(1) Trait coordination and network analysis: Our goal was to explore not only trait means but also the strength and topology of coordination among traits. Removing correlated traits would have risked oversimplifying the structure of trait networks and potentially omitting important hub or bridging traits (e.g., PNUE, PPUE, gs) that, although correlated, may serve distinct roles in physiological regulation.
(2) Intraspecific variation comparison: We aimed to assess intraspecific plasticity at the level of individual traits, which can differ significantly even within correlated trait groups due to trait-specific environmental sensitivity (e.g., PNUE vs. PPUE, δ¹³C vs. WUE). Aggregating traits would limit our ability to detect such patterns.
(3) Biological interpretability: While some traits are statistically correlated, they reflect different physiological mechanisms (e.g., gs reflects stomatal behavior; A represents carbon assimilation capacity). Including all traits allows a more nuanced discussion of functional strategies, particularly between the acquisitive and conservative species.To address potential concerns about clarity, we have revised the figure captions and main text interpretations to better highlight major trait patterns and to guide readers through the complexity of the trait relationships. We hope this balances analytical rigor with interpretability.
3. In the abstract you say that P. jezoensis is a treeline genus elsewhere. Does that mean that elsewhere it's on the treeline but not on your site? If so, it calls into question your study comparing a treeline and a non-treeline species.
[Response]: Thank you very much for this important and insightful comment. We acknowledge that our original statement may have led to confusion regarding the ecological role of P. jezoensis in our study region.
To clarify, while P. jezoensis is indeed a treeline-forming genus in other geographic regions such as Japan and the Russian Far East, it does not currently form the treeline on Changbai Mountain, where our study was conducted. Instead, B. ermanii dominates the upper treeline zone (1700–2200 m), while P. jezoensis is primarily found in the subalpine spruce-fir forest zone at lower elevations (1300–1800 m). Thus, in the context of this local elevational system, P. jezoensis functions ecologically as a non-treeline species.
Importantly, the objective of our study was not to compare species based on their functional types (e.g., conifer vs. broadleaf), nor to generalize treeline behavior across regions, but rather to explore how two locally dominant species occupying different elevational zones respond to environmental gradients near their upper distributional limits. We focus on how each species adjusts its leaf functional traits, intraspecific variation, and trait coordination in response to changing environmental conditions, especially in the context of climate-induced range shifts.4. The manuscript requires careful proofreading, as it contains numerous typos, errors, and repeated phrases (e.g., lines 87, 97, 121, 160, 225, 237). The introduction lacks clarity and does not effectively communicate the main objectives or ecological significance of the study. Strengthening the justification for the research questions and explicitly outlining their broader relevance would greatly enhance this section. Figure legends are overly succinct and should be expanded to allow readers to interpret the figures without referring back to the main text. In the Methods section, essential parameters related to the trait network analysis—such as modularity and closeness—are not defined. While some of these are included in the supplementary materials, they should be clearly explained and contextualized within the main text. Additionally, the ecological relevance of these network metrics should be supported with appropriate references and further discussed in the Discussion section to better integrate them into ecological theory. The Results section would benefit from more explicit descriptions of key findings, including numerical values drawn from the figures to improve interpretability. The Discussion, meanwhile, needs editing for clarity and coherence, and should include a section addressing the limitations and potential weaknesses of the study’s approach. Finally, the authors should place greater emphasis on the broader ecological or practical implications of their findings to strengthen the study's impact.
[Response]: Thank you for these detailed and constructive comments. We have carefully revised the manuscript to improve language clarity, eliminate repetitions, and enhance the overall logical flow. The Introduction has been refined to better articulate the study’s objectives and ecological significance, with improved justification for our research questions. Figure legends have been expanded and clarified, now providing sufficient context for interpretation without referring back to the main text. Key trait network metrics (e.g., modularity, closeness) are now clearly defined in the Methods, and their ecological relevance is further discussed in the Discussion with appropriate references. In the Results section, we added more numerical values to support key findings. Additionally, we included a brief paragraph addressing the study’s limitations and strengthened the discussion on broader ecological implications.5. Several times throughout the manuscript—particularly in the Discussion—you suggest that conifers, such as Picea jezoensis, are less adapted for upward range shifts. However, I find this conclusion somewhat overreaching, given that it is based solely on a limited set of adult physiological traits. Key processes related to establishment, such as seed production, dispersal mechanisms, germination success, and seedling survival, are not addressed in the study. Without considering these life history stages, it is difficult to draw robust conclusions about a species' capacity to expand its range in response to climate change. I would recommend framing these interpretations more cautiously or supporting them with broader evidence.
[Response]: Thank you for this important and insightful comment. We fully agree that drawing conclusions about species’ range expansion potential based solely on adult leaf functional traits has limitations, particularly when key life history processes—such as seed production, dispersal ability, germination success, and seedling establishment—are not directly measured.
In response, we have revised the relevant sections in Conclusion to frame our interpretation more cautiously, emphasizing that our conclusions about P. jezoensis reflect only its physiological trait-based responses near its upper elevational limit, rather than a comprehensive assessment of its migration capacity. We now explicitly acknowledge that successful upward range shifts depend on multiple factors beyond trait strategies, and we highlight this as a limitation of the current study. We have also added relevant references pointing to the importance of early life-stage processes in determining species distributional dynamics under climate change (Lines 563-570).6. Your results highlight two distinct adaptive strategies—resource acquisition and resource conservation—across the two species studied. However, given the complexity and number of traits analyzed, it can be challenging to synthesize the findings as a whole. I suggest including a summary figure or diagram in the Discussion section that visually distills the main results for each species and clearly contrasts their respective strategies. This would greatly enhance the reader’s ability to grasp the overarching patterns and takeaways of the study.
[Response]: Thank you for the valuable suggestion. In response, we have added a summary table (now included as Table 3) in the Discussion section to visually compare the key adaptive strategies of B. ermanii and P. jezoensis. This table synthesizes differences in resource use strategy, trait variation, coordination, and water-use patterns, making the main findings more accessible to readers.Minor issues:
1. it's best to avoid using acronyms in the abstract
[Response]: Thank you for the suggestion. We have revised the abstract to minimize the use of acronyms and now spell out key terms (e.g., “non-structural carbohydrates” instead of “NSC”) to ensure clarity and accessibility for a broader readership.2. L84: ref ?
[Response]: Thank you. We have added relevant references to support this statement. (Lines 93).3. L87: repetition of “functional traits”
[Response]: Done. Thank you very much.4. L96-99: lots of repetitions, improve the sentances
[Response]: Thank you for your suggestion. We have revised these sentences to improve clarity and eliminate repetitions, while maintaining the original meaning. The revised version now reads:
“In harsh environments such as polar and alpine regions, woody plants tend to exhibit lower connectivity and higher modularity in their trait networks (Rao et al., 2022), whereas in more favorable tropical conditions, higher connectivity and lower modularity have been observed (Flores-Moreno et al., 2019). These patterns suggest that reduced trait coordination and increased modular structure may be advantageous under environmental stress.” (Lines 102-107)5. L107-108: You say there are relatively few studies, but you don't cite any. Are you sure that no study has looked at the dynamics and physiological mechanisms of non-treeline species in response to climate change?
[Response]: Thank you for this important comment. In the revised manuscript, we have clarified the sentence to indicate that fewer studies have focused specifically on the responses of non-treeline species near their upper distribution limits, particularly in comparison to the extensive literature on treeline species. We also added relevant citations, such as Dong et al. (Dong et al., 2024), which addressed physiological mechanisms of P. jezoensis near its elevational limit. (Lines 115).6. L113: what do you mean by “integration”?
[Response]: Thank you for your question. In this paper, “integration” refers to the degree of coordination among multiple functional traits within a species. High trait integration implies strong interrelationships among traits that may function together as a coordinated strategy. To clarify this, we have revised the sentence in the manuscript to read:
“These comparisons will help us understand how treeline and non-treeline species respond to climate change in terms of trait means, trait plasticity, and coordination among traits (i.e., trait integration).” (Lines 118-120)7. L134: ‘a mean temperature of -7.3 to 4.9 °C in the growing season and annual precipitation of 800 to 1800 mm.’ Why do you give two figures each time? are they averages? quantiles? please specify.
[Response]: Thank you for the comment. Now,we have revised the sentence in the manuscript to clarify this as follows:
“The Changbai Mountain region experiences a growing season mean temperature ranging from −7.3 °C to 4.9 °C, and an annual precipitation between 800 and 1800 mm (Zhuang et al., 2017), reflecting typical climatic conditions across the mountain’s elevational zones.” (Lines 148-151).8. L140: You say B. ermanii is a treeline species, do you have a ref to support this?
[Response]: Thank you for your comment. We have added a reference (Du et al., 2018) to support the statement that B. ermanii is a treeline species on Changbai Mountain. (Lines 156).9. L142-145: Over what period were these figures calculated and what are the values for your sites?
[Response]: Thank you for this thoughtful comment. The lapse rate values cited (−0.68 °C per 100 m for temperature and +0.93% for relative humidity) are based on an empirical study by Reich et al. (1998), which reflects long-term average conditions in temperate montane forests. These values are not directly measured from our study plots but are used to provide general environmental context along the elevational gradient of Changbai Mountain. To avoid confusion, we have clarified this point in the revised manuscript and now state that these figures are regionally reported estimates rather than site-specific measurements. (Lines 163–167).10. In methods (2.1.), could you specify if the area is grazed or mown. If yes is it still the case? And if the forest is managed? These informations are important to understand the structure and the responses to climate change.
[Response]: Thank you for the insightful comment. We have added clarification in the Methods section (2.1) regarding land use and forest management. Specifically, we note that the study plots are located in a protected area of Changbai Mountain, where grazing, mowing, and active forest management are strictly prohibited. These areas are part of long-term ecological monitoring zones and remain under natural successional dynamics. This information has been included to better contextualize the vegetation structure and trait responses to climate variation. (Lines 167–172).11. I think we're missing a study site map showing the sites sampled by species and their elevation. It is also in this figure that you could put the temperature values of the environments of each species (S1). You could add pictures of the two species.
[Response]: Thank you for the helpful suggestion. We have added a new supplementary figure (Figure S1) showing the main distribution areas of B. ermanii and P. jezoensis, as well as the specific sampling sites and their corresponding elevations. This figure improves the spatial clarity of our sampling design and the context of species distribution.12. Figure 1: ‘environment’ corresponds to elevation ? the term needs to be changed. What's more, we don't know what altitude range it corresponds to, e.g. 1700-1800?
[Response]: Thank you for pointing this out. In the original figure, the term “environment” was used imprecisely to indicate sampling elevation. To improve clarity, we have replaced “environment” with the exact elevation ranges in the updated figure. In addition, we have revised the figure caption to specify the elevation intervals corresponding to each sampling group.13. I wonder if you run a PCA with the two species together, would that give two clusters corresponding to each species? Or would you still find altitudinal clusters?
[Response]: Thank you for this thoughtful comment. In this study, we conducted separate PCA analyses for B. ermanii and P. jezoensis to examine how leaf functional traits vary with elevation within each species, which aligns directly with our objective of understanding species-specific adaptive responses to elevational gradients. Since the two species differ markedly in functional type (deciduous vs. evergreen) and ecological strategy, a joint PCA would likely result in species-based clustering, which is less informative for our purpose.
Our focus is not on comparing species per se, but on assessing how traits shift within species across elevations. Therefore, we believe that separate PCA analyses better reflect the ecological questions posed and allow clearer interpretation of intra-species trait-environment relationships. For this reason, we did not conduct a combined PCA, which would not meaningfully advance the aims of this study.14. Table 2: What are F and p? You need to explain. What does ‘Elevation*species’ mean? It needs to be explained much better.
[Response]: Thank you for your valuable comment. In the revised manuscript, we have clarified the meaning of the F and p values, as well as the interaction term “Elevation × Species,” in the table caption. Specifically, F refers to the F-statistic from the linear mixed-effects model, and p represents the corresponding significance level. “Elevation × Species” indicates the interaction effect between elevation and species on each trait, showing whether the effect of elevation differs between the two species. We have updated the caption of Table 2 to include these definitions to improve clarity. (Lines 310–316).15. Figure 2: equations and p values are displayed only for significant regressions ? specify.
[Response]: Thank you for your comment. Yes, in Figure 2, regression lines, equations, and p-values are shown only for relationships that are statistically significant (p < 0.05), based on the results of the linear mixed-effects models presented in Table 2. We have now clarified this in the figure caption. (Lines 305–309).16. Figure 3: please add dotted ablines for certain % (25, 50, 75) to facilitate reading (panels a and b). For panel c, it's an average QCD per trait group? we don't understand what you've done here, please explain.
[Response]: Thank you for the constructive comment. We have updated Figure 3c by adding reference lines to improve interpretability. Additionally, we clarified in the figure legend that panel (c) presents the relative contribution of each trait group to the total intraspecific variation (QCD) within each species. This panel is intended to visually summarize group-level differences in trait plasticity, complementing the individual trait-level patterns shown in panels (a) and (b).17. Figure 4: I don't see the advantage of panels a and b over PCA. Justify it to me or put it in sup mat.
[Response]: Thank you for raising this important point. While PCA (Figure 1) provides a useful dimensionality reduction based on trait covariation, it summarizes multivariate structure without revealing the specific pairwise correlations or network topology among traits. In contrast, Figure 4a and 4b offer a trait network perspective, visualizing the significant pairwise correlations (edges) and highlighting how traits are functionally coordinated within each species.
This network approach reveals features not captured in PCA, such as:
(1) Inter-trait connectivity (i.e., average number of trait-trait links),
(2) Positive vs. negative correlations,
(3) Trait modularity and centrality, which are further explored in Figure 5.
Therefore, Figure 4 complements rather than duplicates PCA, and helps address our hypothesis on trait coordination (H2).18. Parameters of figures 4 and 5 need to be explained and ecological significances detailed and justified in methods.
[Response]: Thank you for the comment. We have added explanations of the parameters and their ecological meanings directly in the captions of Figures 4 and 5.19. S1 : “DBH” meaning ? same, “High” what is high? Is it soil/air temperature? Specify.
[Response]: Thank you for pointing this out. We have clarified that “DBH” refers to diameter at breast height and “High” refers to tree height. We also specified that the temperature was measured at canopy height. These clarifications have been added to the caption of Table S1.20. L368-369: ref ?
[Response]: Thank you for the suggestion. We have now added the appropriate references to support this statement (Lines 452).21. Could you detail in methods the ecology of the two species: pioneer species? Place in the ecological succession, mode of dispersal.
[Response]: Thank you for the helpful comment. We have added a brief description of the ecological characteristics of the two species in the Methods, Section 2.1 (Study Area). Specifically, we note that B. ermanii is a pioneer species often found in early-successional stages and is wind-dispersed, while P. jezoensis is a late-successional conifer with relatively limited seed dispersal, primarily via gravity and short distance of wind-bearing. These distinctions help contextualize their adaptive strategies and distribution patterns along the elevational gradient (Lines 156-163).22. L420-432: not just a difference between conifer and broadleaved species?
[Response]: Thank you for this insightful comment. We agree that the observed differences in trait correlations and regulation strategies may partially reflect the fundamental distinctions between coniferous (P. jezoensis) and deciduous broadleaf (B. ermanii) species. However, our analysis primarily aimed to explore how these species respond differently to elevational variations rather than to contrast functional types per se.
To address this concern, we have revised the text to explicitly acknowledge that part of the observed variation may result from differences in life-history attributes and phylogenetic differences, but we also emphasize that elevational gradients shape the strength and direction of trait coordination in both species (Lines 554-559).23. L434-435: “contrasting growth environments” in terms of what? Explain.
[Response]: Thank you for your thoughtful comment. In the revised manuscript, we have clarified that “contrasting growth environments” refer specifically to differences in temperature, light availability, and microclimatic stability across the elevational ranges of the two species. Betula ermanii, occurring near or above the treeline, is exposed to colder temperatures, stronger radiation, and greater environmental fluctuations. In contrast, Picea jezoensis inhabits lower elevations with milder temperatures, denser canopy cover, and more buffered microclimates.
However, we fully agree that the observed trait differences cannot be attributed to environmental filtering alone. As now stated in the revised Discussion, these patterns likely result from an interaction between external environmental constraints and intrinsic species-specific physiological strategies. This perspective emphasizes that species’ functional responses are shaped by both ecological context and evolutionary legacy—not simply by their position along an elevational gradient or by differences in leaf habit (Lines 554-559).
Dong, R., Li, N., Li, M.-H., Cong, Y., Du, H., Gao, D., and He, H. S.: Carbon allocation in Picea jezoensis: Adaptation strategies of a non-treeline species at its upper elevation limit, Forest Ecosystems, 11, 100188, https://doi.org/10.1016/j.fecs.2024.100188, 2024.
Du, H., Liu, J., Li, M. H., Büntgen, U., Yang, Y., Wang, L., Wu, Z., and He, H. S.: Warming‐induced upward migration of the alpine treeline in the Changbai Mountains, northeast China, Global Change Biology, 24, 1256-1266, 2018.
Flores-Moreno, H., Fazayeli, F., Banerjee, A., Datta, A., Kattge, J., Butler, E. E., Atkin, O. K., Wythers, K., Chen, M., Anand, M., Bahn, M., Byun, C., Cornelissen, J. H. C., Craine, J., Gonzalez-Melo, A., Hattingh, W. N., Jansen, S., Kraft, N. J. B., Kramer, K., Laughlin, D. C., Minden, V., Niinemets, Ü., Onipchenko, V., Peñuelas, J., Soudzilovskaia, N. A., Dalrymple, R. L., and Reich, P. B.: Robustness of trait connections across environmental gradients and growth forms, Global Ecology and Biogeography, 28, 1806-1826, https://doi.org/10.1111/geb.12996, 2019.
Rao, Q., Su, H., Ruan, L., Xia, W., Deng, X., Wang, L., Xu, P., Shen, H., Chen, J., and Xie, P.: Phosphorus enrichment affects trait network topologies and the growth of submerged macrophytes, Environmental Pollution, 292, 118331, https://doi.org/10.1016/j.envpol.2021.118331, 2022.
Status: closed
-
RC1: 'Comment on egusphere-2025-369', Anonymous Referee #1, 31 Mar 2025
The manuscript has an important scientific significance by giving new information about functional traits and their inter- and intraspecific variation of two tree species in an elevation gradient in Northeast China: a deciduous treeline species Betula ermanii and a conifer tree species Picea jezoensis ( non-treeline species), and a discussion on the species adaptive strategies and ability of upward range expansion. However, there are some important issues to be fixed, regarding the analyses, explanations of the obtained results and result interpretation. After these have been fixed, the manuscript is of good quality to be published in EGUsphere, according to my opinion. Below are my detailed comments, which are also marked as comments to the pdf file attached:
L23-24: What is expected from these in relation to treeline environment?
L33-34: This was against expectations, right?
L36-37: Why these traits are expected to relate adaptation to lower elevation? I would expect an opposite effect, as high elevation environments are sometimes also dry, and resource conservation strategy may be beneficial there due to harsh conditions.
L61: Do you refer here inter- or intraspecific variation in traits? Or both?
L71-73: Isn't resource conservation strategy something that would be beneficial at the treeline due to harsh environmental, not the resource acquisition strategy?
L83-85: How about the evolutionary aspect here? Physiological traits, such as photosynthesis, is an old trait in evolutionary perspective, and is therefore expected to be rather fixed, i.e. not very plastic, due to the conservative inheritance of the trait caused by its complex genetic basis.
L121-123: Why we are expecting reduced leaf trait connectivity when the construction costs need to be minimized? Could you give a short clarification for this.
L147-148: How much of leaf trait differences are explained by the fact that the other sp is conifer and other is deciduous? A short discussion on this would be good to have.
L216-218_ What were used here as fixed and random factors?
L221: Can you explain here what Q3 and Q1 are?
L251: This is an unclear statement.
L263-264: Adaptation to higher elevation (colder or drier conditions)?
L266-267: The datapoints in each elevation are not independent, as they are from the same tree. Therefore, a simple linear regression is not appropriate here. Mean values per tree should be used, but this is also a problem as there is only 6 datapoints then. Thus, some other method than simple lm should be used. Maybe using a tree as a random factor (?)
L269-270: Can you give a little bit more explanation here, e.g. was the other analysis done combining all the data together (both species)?
L280-283: I am wondering how much of this variation is due to measurement challenges, as these traits, especially photosynthetic traits, are very sensible, and therefore their measurement usually need several repetitions? A discussion about this would be good to add.
L312: Surprising that there is no relationship between A and SLA, and only a weak relationship between A and gs in P. jezwoensis and no relationship at all in B. ermanii ? There is a discussion about this in the discussion section., but it should be extended and made more clear.
L313: This is confusing as the same color combination is used to separate the species.
L317: Does this figure give significantly more information than fig 3 and 4? If so, can you explain a bit more what does each point (a, b and c) mean?
L324-325: What does the radical water use strategy mean and how was that observed? This is discussed later on (in 4.2) a bit more, but if it is mentioned here it needs some explanation.
L328-329: But there was no connection between these traits, which is surprising (fig 4). This is discussed a bit more later on in 4.2, but same as above: if it is mentioned here, it needs some explanation.
L331-335: I think this is a bit controversial, as harsh treeline environments are not resource-rich. Can you explain better this idea?
L337-339: What kind of mechanisms is this based on? Can you explain better, so that the next sentence would be more justified.
L351-352: But on the other hand, the traits showed more intraspecific variation, indicating that there is potential.
L355-356: It may limit a rapid expansion, but on the other hand, it can better ensure survival in harsh conditions. This should be also discussed.
L362-363: Can you explain what is the reasoning behind this statement?
L364-365: Is this due to a stronger adaptation to a certain type of environment (cold /dry)?
L365: Is there something missing from this sentence?
L367-371: I think that this part is somehow aiming to give an explanation for my questions /comments (regarding e.g. the lines 324-325 and 328-329), but it could be a bit more clear, expanded and combined with the statements in the corresponding lines.
L374-376: I think that this statement is not so clear, especially what comes to photosynthesis, which is an old plant trait and thus expended to be quite fixed.
What do you mean by anisotropic traits?
L377-378: Maybe some of them, but photosynthesis is an old trait and therefore expected to be more fixed.
L382-384: This makes sense, indicating that this is thus more as a species-specific difference than environmental adaptation.
L397-399: This sentence seems incomplete, or something is missing.
L417-418: What do you mean by "the central traits"?
L421-427: Could you explain these interesting mechanisms a bit more. For example, why an increase in photosynthetic rate was not observed in B. ermanii when stomatal opening increased? How the independent regulation of water and carbon is expected to work?
L433-435: This was not tested with an experimental setting. So, it is difficult to claim that the differences were driven only by the environment and not by species-specific differences. And the difference in WUE seemed to be species-specific difference anyway. This aspect should not be forgotten from the discussion.
-
AC2: 'Reply on RC1', Hong He, 10 Jun 2025
Reviewer #1:
The manuscript has an important scientific significance by giving new information about functional traits and their inter- and intraspecific variation of two tree species in an elevation gradient in Northeast China: a deciduous treeline species Betula ermanii and a conifer tree species Picea jezoensis (non-treeline species), and a discussion on the species adaptive strategies and ability of upward range expansion. However, there are some important issues to be fixed, regarding the analyses, explanations of the obtained results and result interpretation. After these have been fixed, the manuscript is of good quality to be published in EGUsphere, according to my opinion. Below are my detailed comments, which are also marked as comments to the pdf file attached:
[Response]: We sincerely thank you for the positive evaluation of our manuscript and the constructive comments provided. We have carefully addressed each of the issues raised, especially regarding the analyses, explanation of results, and their ecological interpretation. Below, we provided the point-to-point responses to all your comments.1. L23-24: What is expected from these in relation to treeline environment?
[Response]: Thank you for your insightful question. We agree that the original sentence did not clearly articulate our expectations regarding trait patternsin relation to treeline environments. In the revised manuscript, we have clarified this point by adding a sentence “Usually, treeline species would exhibit more acquisitive traits that enhance adaptability to low temperature and short growing seasons, while non-treeline species would display more conservative traits adapted to relatively stable, lower-elevation environments.” (Lines 24–27).2. L33-34: This was against expectations, right?
[Response]: Thank you for this insightful observation. We agree that the observed combination of high intraspecific variability in photosynthetic traits and weaker trait coordination in Picea jezoensis appears to contrast with the general expectation that stressful environments promote tighter trait integration and reduces trait variation. In this case, P. jezoensis is a non-treeline species experiencing conditions beyond its optimal range at higher elevations. We interpret this pattern not as maladaptive, but rather as an alternative compensatory strategy—potentially allowing individuals to explore diverse trait combinations to maintain function under novel environmental stress. This flexibility may reflect an attempt to cope with light competition, cold temperatures, and environmental heterogeneity in the upper margin of its range.3. L36-37: Why these traits are expected to relate adaptation to lower elevation? I would expect an opposite effect, as high elevation environments are sometimes also dry, and resource conservation strategy may be beneficial there due to harsh conditions.
[Response]: Thank you for raising this important point. We agree that resource-conserving strategies, including traits related to drought tolerance and water use efficiency, can be beneficial in high-elevation environments where water limitations exist. However, in our study, on Changbai Mountain, the treeline zone is characterized by plentiful precipitation, low temperatures, strong radiation, and short growing seasons, rather than drought stress. In contrast, the lower elevation zones where P. jezoensis dominates are characterized by denser canopies, lower light availability, and more stable microclimates.
The resource-conserving traits of P. jezoensis, such as lower SLA, higher δ¹³C, and relatively limited trait integration, align more closely with adaptation to shade, slower growth, and long-term structural investment—traits advantageous in stable, low-light environments. These traits may restrict its ability to cope with the high-light, low-temperature conditions near the treeline, limiting its potential for upward expansion.4. L61: Do you refer here inter- or intraspecific variation in traits? Or both?
[Response]: Thank you for your helpful comment. In this sentence, we refer to both interspecific and intraspecific variation in functional traits, as both levels of variation are shaped by environmental gradients such as elevation. To clarify this, we have revised the sentence in the manuscript to explicitly state that environmental variation can influence trait means between species (interspecific variation) as well as trait variability within species (intraspecific variation) (Lines 67–68).5. L71-73: Isn't resource conservation strategy something that would be beneficial at the treeline due to harsh environmental, not the resource acquisition strategy?
[Response]: Thank you for this important comment. You are quite right that, in general, resource-conservation strategies are often thought to be more beneficial under harsh conditions such as those founded at treelines. However, recent studies have shown that under some high-elevation conditions—especially where light is not limiting and the growing season is short—species that adopt a rapid resource acquisition strategy can gain a competitive advantage by maximizing resource use during limited favorable periods (Reich, 2014; Liao et al., 2021). In our study, B. ermanii exhibits such a strategy, with higher leaf N and P content and stomatal conductance at higher elevations, allowing for rapid carbon assimilation. This trait profile may support its dominance and potential upward expansion under warming scenarios.6. L83-85: How about the evolutionary aspect here? Physiological traits, such as photosynthesis, is an old trait in evolutionary perspective, and is therefore expected to be rather fixed, i.e. not very plastic, due to the conservative inheritance of the trait caused by its complex genetic basis.
[Response]: Thank you for raising this important evolutionary perspective. We agree that physiological traits such as photosynthesis are highly conserved over evolutionary timescales due to their fundamental role and complex genetic control. However, in this context, we were referring to ecological plasticity—i.e., the short-term, environmentally induced variation within species—rather than evolutionary lability. To clarify this distinction, we have revised the sentence to indicate that physiological traits may exhibit higher phenotypic plasticity at ecological timescales, even if they are evolutionarily conserved. The revised text now was added as "Although physiological traits such as photosynthesis are evolutionarily conserved due to complex genetic constraints, they can still exhibit high phenotypic plasticity at ecological timescales because of their reversible and low-cost adjustments (Grime et al., 2002)." (Lines 90–93)7. L121-123: Why we are expecting reduced leaf trait connectivity when the construction costs need to be minimized? Could you give a short clarification for this.
[Response]: Thank you for this thoughtful question. The expectation of reduced trait connectivity under harsh conditions is based on the idea that strong coordination among traits requires more developmental and physiological integration, which can be costly to maintain. In highly stressful environments such as treelines, plants may prioritize flexibility and survival by decoupling trait relationships, allowing independent adjustment of traits to minimize energy and resource expenditure. We have clarified this rationale in the revised text (Lines 135–138).8. L147-148: How much of leaf trait differences are explained by the fact that the other sp is conifer and other is deciduous? A short discussion on this would be good to have.
[Response]: Thank you for this important comment. We agree that the distinction between coniferous and deciduous leaf types may partly contribute to the observed differences in leaf functional traits. However, our study focused on intraspecific trait variation and trait coordination along elevational gradients within each species, rather than direct trait comparisons between functional types. In the revised manuscript, we have now included a short discussion acknowledging that some trait differences—such as SLA, LDMC, and photosynthetic rates—may inherently reflect differences between evergreen conifers and deciduous broadleaved trees. Nonetheless, we argue that the contrasting patterns in intraspecific variability and trait integration are primarily driven by environmental adaptation along the elevational gradient, as similar trends have also been observed within comparable functional groups in other studies (He et al., 2020; Rao et al., 2022). We have added this clarification in the Discussion (Lines 126–135).9. L216-218： What were used here as fixed and random factors?
[Response]: Thank you for your question. In our linear mixed models (LMMs), elevation, species, and their interaction (elevation × species) were used as fixed effects, and the individual tree was used as a random effect. We have clarified this model structure in the revised Methods section (Lines 250–254).10. L221: Can you explain here what Q3 and Q1 are?
[Response]: Thank you for your question. We have revised the text to define Q3 and Q1. Specifically, Q3 refers to the third quartile (the 75th percentile) and Q1 refers to the first quartile (the 25th percentile) of the trait distribution. The QCD formula thus represents a normalized measure of dispersion that is less sensitive to extreme values compared to the coefficient of variation. This clarification has been added to the revised Methods section (Lines 258–259).11. L251: This is an unclear statement.
[Response]: Thank you for your comment. We agree that the original sentence was unclear. We have revised it to more precisely describe the pattern in the PCA plot for P. jezoensis. The sentence now was revised as: "The individuals of P. jezoensis were distributed broadly along the PCA space, with no distinct clustering by elevation, suggesting that leaf trait variation in this species is less structured by elevation compared to B. ermanii (Figure 1b)." (Lines 288–291).
This revision clarified that P. jezoensis showed little elevational differentiation in trait composition based on PCA axes.12. L263-264: Adaptation to higher elevation (colder or drier conditions)?
[Response]: Thank you for your comment. We have clarified the ecological interpretation of these trait shifts. Specifically, for B. ermanii, the decrease in WUE and δ¹⁸O with elevation likely reflects a more liberal water-use strategy under colder but not necessarily drier conditions, as high-elevation sites on Changbai Mountain are characterized by low temperature but sufficient moisture. In contrast, for P. jezoensis, the increase in δ¹³C and decrease in SLA with elevation suggest enhanced water-use efficiency and thicker leaves, which are typical of adaptations to increased evaporative demand or lower temperatures.13. L266-267: The datapoints in each elevation are not independent, as they are from the same tree. Therefore, a simple linear regression is not appropriate here. Mean values per tree should be used, but this is also a problem as there is only 6 datapoints then. Thus, some other method than simple lm should be used. Maybe using a tree as a random factor (?)
[Response]: Thank you for the insightful comment. We agree that statistical analyses should account for the non-independence of data points from the same tree. In fact, we have already addressed this issue in Table 2, where we used linear mixed-effects models (LMMs) with tree identity as a random factor to evaluate the effects of elevation, species, and their interaction.
Figure 2 serves as a complementary visual summary of these trait–elevation relationships. The lines shown are not intended as formal linear models but as visual aids to highlight general trends, with statistical inference provided in Table 2. We have clarified this purpose in the revised figure caption and Methods section (lines 306-309).14. L269-270: Can you give a little bit more explanation here, e.g. was the other analysis done combining all the data together (both species)?
[Response]: Thank you for the comment. Yes, the linear mixed-effects model presented in Table 2 was performed by combining all the data across both species and elevations into a single model. This allowed us to assess the main effects of elevation, species, and their interaction (elevation × species) on each trait. One tree was included as a random effect to account for among-tree replication. We have clarified this in the revised caption of Table 2 and in the Methods section (Lines 310–316).15. L280-283: I am wondering how much of this variation is due to measurement challenges, as these traits, especially photosynthetic traits, are very sensible, and therefore their measurement usually need several repetitions? A discussion about this would be good to add.
[Response]: Thank you for raising this important point. We fully acknowledge that photosynthetic traits are generally sensitive to environmental fluctuations and may be prone to measurement variability. However, in our study, we addressed this issue by performing multiple replicate measurements per tree under carefully standardized conditions (e.g., fixed time frame, stable weather, and instrument calibration). Only stable and consistent readings were recorded for analysis. Therefore, we are confident that the observed intraspecific variation in photosynthetic traits reflects true biological differences rather than measurement artifacts (Lines 212–214).16. L312: Surprising that there is no relationship between A and SLA, and only a weak relationship between A and gs in P. jezwoensis and no relationship at all in B. ermanii ? There is a discussion about this in the discussion section., but it should be extended and made more clear.
[Response]: Thank you for this insightful observation. We agree that the lack of significant correlations between A and SLA or gs, particularly in B. ermanii, is unexpected given their commonly reported positive associations. We have expanded the discussion on this point. Specifically, we suggest that these decoupled relationships may reflect species-specific trait regulation strategies under environmental stress. For B. ermanii, strong coordination between nutrient-use traits (e.g., PNUE, PPUE) and A may override the influence of SLA or gs. In P. jezoensis, trait independence may reflect a modular response strategy where traits function more independently under competition and shade. These patterns underscore that trait coordination can vary not only across environments but also across species with different ecological strategies. This expanded explanation is now included in Discussion 4.3 (Lines 521–535).17. L313: This is confusing as the same color combination is used to separate the species.
[Response]: Thank you for pointing this out. To eliminate any confusion, we have revised the figure by using pink and green lines to represent positive and negative trait correlations, respectively. The updated figure caption now clearly explains the revised color coding (Figure 4).18. L317: Does this figure give significantly more information than fig 3 and 4? If so, can you explain a bit more what does each point (a, b and c) mean?
[Response]: Thank you for this comment. Figure 5 complements Figure 4 by quantifying the topological importance of individual traits in the network, rather than only showing their connections. Specifically, Figure 5a shows the degree centrality (number of connections) for each trait, 5b shows centrality closeness, reflecting how efficiently a trait is connected to all others in the network, and 5c shows betweenness centrality, indicating the extent to which a trait acts as a bridge in the network.
These metrics help identify hub traits and differentiate between the trait coordination structures of B. ermanii and P. jezoensis. We have clarified these points in the revised figure caption and discussion (Lines 370–376).19. L324-325: What does the radical water use strategy mean and how was that observed? This is discussed later on (in 4.2) a bit more, but if it is mentioned here it needs some explanation.
[Response]: Thank you for pointing this out. We agree that the term “radical water use strategy” requires clarification at its first mention. In the revised manuscript, we have explained that radical water use strategy refers to a more liberal water-use behavior, characterized by increased stomatal conductance (gs) and decreased water use efficiency (WUE) at higher elevations in B. ermanii. These changes suggest a strategy of prioritizing rapid gas exchange over water conservation, which may be advantageous during short alpine growing seasons. We have added this clarification in Section 4.1 (Lines 394–396).20. L328-329: But there was no connection between these traits, which is surprising (fig 4). This is discussed a bit more later on in 4.2, but same as above: if it is mentioned here, it needs some explanation.
[Response]: Thank you for pointing this out. We agree that the lack of observed connections between these traits in the network analysis (Figure 4) makes the interpretation in 4.1 potentially unclear. In the revised text, we have added a short clarification noting that although physiological linkages are expected between traits like gs, WUE, and A, the trait network showed weak or no correlations—possibly due to environmentally decoupled regulation or alternative coordination pathways (e.g., via nutrient-use traits like PNUE). This clarification now has been added alongside the original statement in Section 4.1 (Lines 390–392).21. L331-335: I think this is a bit controversial, as harsh treeline environments are not resource-rich. Can you explain better this idea?
[Response]: Thank you for this important observation. We agree that treeline environments are not resource-rich in the classical sense. We have revised the paragraph to clarify that B. ermanii’s “fast” strategy refers not to resource richness, but rather to its ability to rapidly acquire and utilize available resources during short, favorable periods, especially in a seasonally constrained and cold environment. This strategy enables it to take full advantage of transient growth opportunities. We have adjusted the texts in Section 4.1 to better reflect this interpretation (Lines 398–401).22. L337-339: What kind of mechanisms is this based on? Can you explain better, so that the next sentence would be more justified.
[Response]: Thank you for the comment. We agree that this statement needed further justification. We have clarified that the advantage of the fast resource-acquisition strategy under climate warming is based on several mechanisms:
(1) Warming extends the growing season, allowing B. ermanii to make better use of its rapid growth capacity;
(2) Increased temperature improves photosynthetic efficiency and nutrient uptake, both of which are aligned with B. ermanii's trait profile (e.g., high N, gs, PNUE);
(3) This strategy enables quick colonization of newly available niches above the current treeline, especially where competitors are slow-growing;
We have revised the text in Section 4.1 to reflect these points (Lines 404–409).23. L351-352: But on the other hand, the traits showed more intraspecific variation, indicating that there is potential.
[Response]: Thank you for this important comment. We agree that P. jezoensis exhibits a seemingly contrasting combination: a conservative water-use strategy (higher δ¹³C) alongside high intraspecific variation in photosynthetic traits. We interpret this as reflecting a modular response strategy, where core traits like δ¹³C are constrained by long-term adaptation, while other traits such as A, gs, or PNUE remain more plastic and responsive to microenvironmental variation. This decoupling may allow P. jezoensis to maintain a conservative baseline strategy while retaining adaptive flexibility. We have added this clarification to the revised Discussion (Lines 482–489).24. L355-356: It may limit a rapid expansion, but on the other hand, it can better ensure survival in harsh conditions. This should be also discussed.
[Response]: Thank you for this thoughtful comment. We fully agree that conservative trait strategies, while potentially limiting rapid range expansion, can also enhance survival and long-term persistence under harsh environmental conditions. We have added a sentence to the Discussion (Section 4.2) to reflect this dual role: P. jezoensis’s strategy may hinder rapid colonization but could buffer against environmental extremes and resource fluctuations, thereby ensuring stability in marginal habitats. This addition strengthens the balance of our interpretation (Lines 426–430).25. L362-363: Can you explain what is the reasoning behind this statement?
[Response]: Thank you for your question. We have clarified the reasons in the revised text. In harsher environments such as treeline areas, trait expression is more strongly affected by abiotic filtering, which can reduce phenotypic plasticity and allowable variation. Additionally, the cost of expressing highly plastic or deviant traits increases under stress, leading to trait convergence around stress-tolerant optima. We have added this explanation in Section 4.2 (Lines 444–445).26. L364-365: Is this due to a stronger adaptation to a certain type of environment (cold /dry)?
[Response]: Thank you for your question. Yes, we agree that reduced intraspecific variation in harsh environments likely reflects stronger directional selection and environmental filtering, leading to trait convergence around adaptive optima. In the case of B. ermanii, this is most likely associated with adaptation to cold and short-season alpine conditions, where trait combinations that maximize stress tolerance and survival are favored. We have added this clarification in Section 4.2 (Lines 446–449).27. L365: Is there something missing from this sentence?
[Response]: Thank you for pointing this out. We agree that the original sentence lacked an intermediate explanation. In the revised version, we have clarified that higher construction costs constrain the range of viable trait expressions, as producing and maintaining divergent phenotypes becomes more costly under stress. This leads to reduced trait variability through environmental filtering and selection for conservative strategies. The revised sentence now better explains the link between trait cost and intraspecific variation (Lines 446–449).28. L367-371: I think that this part is somehow aiming to give an explanation for my questions /comments (regarding e.g. the lines 324-325 and 328-329), but it could be a bit more clear, expanded and combined with the statements in the corresponding lines.
[Response]: Thank you very much for this insightful comment. We agree that the paragraph on climate filtering and trait variation (Lines 367–371) provides relevant context that should be more directly connected to earlier points regarding the water-use strategy (Lines 324–325) and trait linkages (Lines 328–329). We have revised the text to clarify that while B. ermanii shows lower intraspecific variation, its traits are more tightly coordinated in response to the harsh alpine environment. This combination reflects a stress-constrained yet functionally integrated strategy, where strong coordination among key traits (e.g., PNUE, PPUE, gs) supports rapid resource use during brief growing periods. We have added a clarification to bridge these points across Sections 4.2 (Lines 452–459).29. L374-376: I think that this statement is not so clear, especially what comes to photosynthesis, which is an old plant trait and thus expended to be quite fixed.
[Response]: Thank you for this important comment. We agree that photosynthesis, as a fundamental and evolutionarily conserved trait, is often considered genetically constrained. However, while the core biochemical processes of photosynthesis are conserved, many related traits we measured—such as stomatal conductance (gs), photosynthetic rate (A), and nutrient use efficiencies (PNUE, PPUE)—reflect ecophysiological responses to environmental variability and are influenced by microclimatic and nutrient conditions. These traits can exhibit substantial plasticity even within species.30. What do you mean by anisotropic traits?
[Response]: Thank you for pointing this out. We acknowledge that the term “anisotropic traits” is not appropriate in this ecological context and may be misleading. What we intended to refer to were structural or morphological traits, such as SLA or LDMC, which tend to have higher construction costs and are less reversible, and therefore typically exhibit lower phenotypic plasticity compared to physiological traits. We have revised the sentence accordingly to improve clarity (Lines 462–467).31. L377-378: Maybe some of them, but photosynthesis is an old trait and therefore expected to be more fixed.
[Response]: Thank you for this clarification. We agree that the core biochemical mechanisms of photosynthesis are highly conserved and not easily plastic. In our context, we referred specifically to physiological traits that modulate photosynthetic performance, such as stomatal conductance (gs), nutrient use efficiencies (PNUE, PPUE), and instantaneous photosynthetic rate (A). These traits, although related to photosynthesis, are more environmentally responsive and reversible, and thus fall into the more plastic category.32. L382-384: This makes sense, indicating that this is thus more as a species-specific difference than environmental adaptation.
[Response]: Thank you for your insightful comment. We agree that the observed differences in which trait groups show higher variation—hydraulic traits in B. ermanii and nutrient traits in P. jezoensis—likely reflect not only environmental influences but also species-specific functional strategies and evolutionary constraints. We have revised the discussion to acknowledge this interpretation, clarifying that both environmental adaptation and inherent species differences likely contribute to these trait variation patterns (Lines 472–475).33. L397-399: This sentence seems incomplete, or something is missing.
[Response]: Thank you for pointing this out. We have revised it to specify what the observed covariation and coordination patterns suggest about species adaptation strategies. The updated sentence now highlights that the differences in trait coordination between B. ermanii and P. jezoensis reflect contrasting strategies for dealing with environmental stress in their respective habitats (Lines 497–499).34. L417-418: What do you mean by "the central traits"?
[Response]: Thank you for your question. In trait network, “central traits” refer to those with high network centrality metrics, such as the degree, closeness, and betweenness, which indicate their importance in connecting and coordinating other traits. In our results (Figure 5), these central traits were primarily photosynthetic traits (e.g., PNUE, PPUE, A), which consistently showed the highest centrality scores in both species. We have clarified this in the revised text and figure caption (Lines 370–376).35. L421-427: Could you explain these interesting mechanisms a bit more. For example, why an increase in photosynthetic rate was not observed in B. ermanii when stomatal opening increased? How the independent regulation of water and carbon is expected to work?
[Response]: Thank you for this insightful comment. We have expanded our explanation of these physiological mechanisms. In B. ermanii, increased stomatal conductance (gs) might enhance CO₂ uptake capacity, but photosynthetic rate (A) might not increase proportionally due to biochemical limitations, such as Rubisco activity or nutrient availability, or feedback regulation under stress.
The negative correlation between gs and PNUE/PPUE suggests that when stomata open wider, photosynthetic N and P use efficiency decreases—possibly because more nutrients are required to maintain the same photosynthetic gain, reflecting a shift toward a resource-intensive strategy under alpine stress.
Similarly, the negative δ¹³C–δ¹⁸O correlation indicates that water-use efficiency (δ¹³C) and stomatal water loss (δ¹⁸O) may be regulated by different pathways or trade-offs, allowing B. ermanii to decouple water and carbon exchange. This might be an adaptive response to rapidly fluctuating alpine conditions. These clarifications have been added to the revised Discussion (Lines 522–542).36. L433-435: This was not tested with an experimental setting. So, it is difficult to claim that the differences were driven only by the environment and not by species-specific differences. And the difference in WUE seemed to be species-specific difference anyway. This aspect should not be forgotten from the discussion.
[Response]: Thank you for this important comment. We agree that our study, being observational along an elevational gradient, cannot fully disentangle environmental drivers from species-specific effects. While the positive correlation between δ¹³C and δ¹⁸O in P. jezoensis may suggest a coordinated response in more favorable conditions, we acknowledge that this pattern could also reflect inherent species-level physiological strategies.
In particular, the overall differences in WUE appeared to be largely species-driven, independent of elevation. We have revised the discussion to clarify that both environmental adaptation and species-specific trait syndromes likely contributed to the observed patterns (Lines 550–559).
Grime, P., J., Mackey, and L., J. M.: The role of plasticity in resource capture by plants, Evolutionary Ecology, 16, 299-307, 10.1023/A:1019640813676, 2002.
He, N., Li, Y., Liu, C., Xu, L., Li, M., Zhang, J., He, J., Tang, Z., Han, X., Ye, Q., Xiao, C., Yu, Q., Liu, S., Sun, W., Niu, S., Li, S., Sack, L., and Yu, G.: Plant Trait Networks: Improved Resolution of the Dimensionality of Adaptation, Trends in Ecology & Evolution, 35, 908-918, https://doi.org/10.1016/j.tree.2020.06.003, 2020.
Liao, H., Pal, R. W., Niinemets, Ü., Bahn, M., Cerabolini, B. E. L., and Peng, S.: Different functional characteristics can explain different dimensions of plant invasion success, Journal of Ecology, 109, 1524-1536, https://doi.org/10.1111/1365-2745.13575, 2021.
Rao, Q., Su, H., Ruan, L., Xia, W., Deng, X., Wang, L., Xu, P., Shen, H., Chen, J., and Xie, P.: Phosphorus enrichment affects trait network topologies and the growth of submerged macrophytes, Environmental Pollution, 292, 118331, https://doi.org/10.1016/j.envpol.2021.118331, 2022.
Reich, P. B.: The world‐wide ‘fast–slow’plant economics spectrum: a traits manifesto, Journal of ecology, 102, 275-301, 2014.
-
AC2: 'Reply on RC1', Hong He, 10 Jun 2025
-
RC2: 'Comment on egusphere-2025-369', Anonymous Referee #2, 23 Apr 2025
This study compares the leaf functional traits of Betula ermanii, a treeline species and Picea jezoensis, a non-treeline species, along an elevational gradient on Changbai Mountain to understand their adaptive strategies. B. ermanii exhibits traits favoring resource acquisition and adaptability at higher elevations, while P. jezoensis shows a resource-conserving strategy more suited to lower elevations.
I have a fundamental concern regarding the conceptual approach of comparing only two species in relation to the treeline. While the study aims to draw broader conclusions about vegetation stratification, analyzing just two species—each with distinct ecological strategies—limits the robustness of the findings. This approach risks confounding species-specific differences with general patterns, making it difficult to extrapolate broader ecological insights. Additionally, the manuscript often lacks sufficient detail and clarity, and contains multiple errors that hinder comprehension. It would benefit greatly from thorough proofreading and substantial rewriting to improve both readability and logical flow. I also have reservations about aspects of the data analysis and figure construction, which are at times unclear or insufficiently justified. These elements should be revised to enhance clarity and relevance.
Below are my specific comments:
Major issues:
- I have a conceptual concern with the study’s approach. The manuscript aims to explain vegetation stratification through functional traits, yet it compares two species that naturally occupy different elevations and likely have inherently different physiological thresholds. The comparison is further complicated by the contrasting functional types—Betula ermanii being a deciduous broadleaf and Picea jezoensis a conifer—each with fundamentally different life strategies. Moreover, the claim that P. jezoensis inhabits a resource-rich and less stressful environment is not well supported by empirical data. Since this species is at its own altitudinal limit, the level of stress it experiences is likely comparable—in relative terms—to that faced by B. ermanii at the treeline. Elevational limits represent ecological thresholds for both species, and interpreting lower elevation as inherently less stressful may overlook species-specific constraints. For instance, at comparable altitudes, the two species show markedly different trait values, and their trait responses to elevation differ substantially. These factors cast doubt on the validity of directly comparing their functional traits, and ultimately complicate the interpretation of adaptive strategies or elevation-driven processes. In my view, a more informative approach might have been to compare two co-occurring treeline species—or two non-treeline species—to identify shared adaptive strategies within each group.
- While I’m not a specialist in functional traits, I wonder if the analysis could be made more accessible by reducing redundancy among highly correlated traits—especially among leaf traits, which often show strong interdependence. Simplifying the dataset by selecting a representative trait from each highly correlated set could help clarify the figures and make them easier to interpret. Similarly, for trait group analyses (e.g., hydraulic, foliar), it might be useful to present either a single representative trait per group or an average value per group with associated uncertainty. This could streamline the presentation and highlight broader patterns more effectively.
- In the abstract you say that P. jezoensis is a treeline genus elsewhere. Does that mean that elsewhere it's on the treeline but not on your site? If so, it calls into question your study comparing a treeline and a non-treeline species.
- The manuscript requires careful proofreading, as it contains numerous typos, errors, and repeated phrases (e.g., lines 87, 97, 121, 160, 225, 237). The introduction lacks clarity and does not effectively communicate the main objectives or ecological significance of the study. Strengthening the justification for the research questions and explicitly outlining their broader relevance would greatly enhance this section. Figure legends are overly succinct and should be expanded to allow readers to interpret the figures without referring back to the main text. In the Methods section, essential parameters related to the trait network analysis—such as modularity and closeness—are not defined. While some of these are included in the supplementary materials, they should be clearly explained and contextualized within the main text. Additionally, the ecological relevance of these network metrics should be supported with appropriate references and further discussed in the Discussion section to better integrate them into ecological theory. The Results section would benefit from more explicit descriptions of key findings, including numerical values drawn from the figures to improve interpretability. The Discussion, meanwhile, needs editing for clarity and coherence, and should include a section addressing the limitations and potential weaknesses of the study’s approach. Finally, the authors should place greater emphasis on the broader ecological or practical implications of their findings to strengthen the study's impact.
- Several times throughout the manuscript—particularly in the Discussion—you suggest that conifers, such as Picea jezoensis, are less adapted for upward range shifts. However, I find this conclusion somewhat overreaching, given that it is based solely on a limited set of adult physiological traits. Key processes related to establishment, such as seed production, dispersal mechanisms, germination success, and seedling survival, are not addressed in the study. Without considering these life history stages, it is difficult to draw robust conclusions about a species' capacity to expand its range in response to climate change. I would recommend framing these interpretations more cautiously or supporting them with broader evidence.
- Your results highlight two distinct adaptive strategies—resource acquisition and resource conservation—across the two species studied. However, given the complexity and number of traits analyzed, it can be challenging to synthesize the findings as a whole. I suggest including a summary figure or diagram in the Discussion section that visually distills the main results for each species and clearly contrasts their respective strategies. This would greatly enhance the reader’s ability to grasp the overarching patterns and takeaways of the study.
Minor issues:
- it's best to avoid using acronyms in the abstract
- L84: ref ?
- L87: repetition of “functional traits”
- L96-99: lots of repetitions, improve the sentances
- L107-108: You say there are relatively few studies, but you don't cite any. Are you sure that no study has looked at the dynamics and physiological mechanisms of non-treeline species in response to climate change?
- L113: what do you mean by “integration”?
- L134: ‘a mean temperature of -7.3 to 4.9 °C in the growing season and annual precipitation of 800 to 1800 mm.’ Why do you give two figures each time? are they averages? quantiles? please specify.
- L140: You say B. ermanii is a treeline species, do you have a ref to support this?
- L142-145: Over what period were these figures calculated and what are the values for your sites?
- In methods (2.1.), could you specify if the area is grazed or mown. If yes is it still the case? And if the forest is managed? These informations are important to understand the structure and the responses to climate change.
- I think we're missing a study site map showing the sites sampled by species and their elevation. It is also in this figure that you could put the temperature values of the environments of each species (S1). You could add pictures of the two species.
- Figure 1: ‘environment’ corresponds to elevation ? the term needs to be changed. What's more, we don't know what altitude range it corresponds to, e.g. 1700-1800?
- I wonder if you run a PCA with the two species together, would that give two clusters corresponding to each species? Or would you still find altitudinal clusters?
- Table 2: What are F and p? You need to explain. What does ‘Elevation*species’ mean? It needs to be explained much better.
- Figure 2: equations and pvalues are displayed only for significant regressions ? specify.
- Figure 3: please add dotted ablines for certain % (25, 50, 75) to facilitate reading (panels a and b). For panel c, it's an average QCD per trait group? we don't understand what you've done here, please explain.
- Figure 4: I don't see the advantage of panels a and b over PCA. Justify it to me or put it in sup mat.
- Parameters of figures 4 and 5 need to be explained and ecological significances detailed and justified in methods.
- S1 : “DBH” meaning ? same, “High” what is high? Is it soil/air temperature? Specify.
- L368-369: ref ?
- Could you detail in methods the ecology of the two species: pioneer species? Place in the ecological succession, mode of dispersal.
- L420-432: not just a difference between conifer and broadleaved species?
- L434-435: “contrasting growth environments” in terms of what? Explain.
Citation: https://doi.org/10.5194/egusphere-2025-369-RC2 -
AC1: 'Reply on RC2', Hong He, 10 Jun 2025
### Reviewer #2:
This study compares the leaf functional traits of Betula ermanii, a treeline species and Picea jezoensis, a non-treeline species, along an elevational gradient on Changbai Mountain to understand their adaptive strategies. B. ermanii exhibits traits favoring resource acquisition and adaptability at higher elevations, while P. jezoensis shows a resource-conserving strategy more suited to lower elevations.I have a fundamental concern regarding the conceptual approach of comparing only two species in relation to the treeline. While the study aims to draw broader conclusions about vegetation stratification, analyzing just two species—each with distinct ecological strategies—limits the robustness of the findings. This approach risks confounding species-specific differences with general patterns, making it difficult to extrapolate broader ecological insights. Additionally, the manuscript often lacks sufficient detail and clarity, and contains multiple errors that hinder comprehension. It would benefit greatly from thorough proofreading and substantial rewriting to improve both readability and logical flow. I also have reservations about aspects of the data analysis and figure construction, which are at times unclear or insufficiently justified. These elements should be revised to enhance clarity and relevance.[Response]: We sincerely thank you for the thoughtful and constructive comments. We fully acknowledge the limitations of comparing only two species, each with distinct ecological strategies. Our goal was not to generalize treeline dynamics across all taxa, but to provide a focused, hypothesis-driven comparison between a dominant deciduous treeline species (B. ermanii) and a representative non-treeline conifer (P. jezoensis) that overlaps with the upper forest zone. We have revised the Introduction and Discussion to more clearly define the scope of inference and highlight the species-specific context of our findings.
We also appreciate the reviewer’s feedback on clarity and presentation. In response, we have carefully revised the manuscript for improved logical flow, language precision, and structural clarity. The figures and data analysis sections have been reorganized and more thoroughly explained, with clearer legends and justifications for the methods used.
Major issues:
1. I have a conceptual concern with the study’s approach. The manuscript aims to explain vegetation stratification through functional traits, yet it compares two species that naturally occupy different elevations and likely have inherently different physiological thresholds. The comparison is further complicated by the contrasting functional types—Betula ermanii being a deciduous broadleaf and Picea jezoensis a conifer—each with fundamentally different life strategies. Moreover, the claim that P. jezoensis inhabits a resource-rich and less stressful environment is not well supported by empirical data. Since this species is at its own altitudinal limit, the level of stress it experiences is likely comparable—in relative terms—to that faced by B. ermanii at the treeline. Elevational limits represent ecological thresholds for both species, and interpreting lower elevation as inherently less stressful may overlook species-specific constraints. For instance, at comparable altitudes, the two species show markedly different trait values, and their trait responses to elevation differ substantially. These factors cast doubt on the validity of directly comparing their functional traits, and ultimately complicate the interpretation of adaptive strategies or elevation-driven processes. In my view, a more informative approach might have been to compare two co-occurring treeline species—or two non-treeline species—to identify shared adaptive strategies within each group.
1. [Response]: We appreciate your insightful comment regarding the complexity of comparing a deciduous broadleaf species (B. ermanii) and a coniferous evergreen species (P. jezoensis). We agree that each species likely experiences environmental stress near its own elevational limit, and that these constraints may not be strictly hierarchical.
To address this concern, we have made the following clarifications in the revised manuscript:
(1) Both species are evaluated within their primary elevational ranges, where they are well established but approaching their current distributional limits.
(2) The interpretation of stress is therefore made relative to each species’ adaptive niche, rather than assuming lower elevations are universally less stressful.
(3) We acknowledge the inherent differences in functional types and strategies between conifers and deciduous broadleaves and emphasize that our analysis focuses on intraspecific trait responses and coordination patterns across elevation, rather than direct trait-by-trait comparisons.
These clarifications are now included in the Introduction (Lines 126–135), to better contextualize the adaptive significance of trait variability and coordination within each species.2. While I’m not a specialist in functional traits, I wonder if the analysis could be made more accessible by reducing redundancy among highly correlated traits—especially among leaf traits, which often show strong interdependence. Simplifying the dataset by selecting a representative trait from each highly correlated set could help clarify the figures and make them easier to interpret. Similarly, for trait group analyses (e.g., hydraulic, foliar), it might be useful to present either a single representative trait per group or an average value per group with associated uncertainty. This could streamline the presentation and highlight broader patterns more effectively.
[Response]: We thank you for this insightful comment. Indeed, functional traits—particularly within the same functional group—often exhibit high correlations. However, in our study, we intentionally retained all measured traits in the network and trait variation analyses to capture the full spectrum of trait covariation and integration. This comprehensive approach is important for several reasons:
(1) Trait coordination and network analysis: Our goal was to explore not only trait means but also the strength and topology of coordination among traits. Removing correlated traits would have risked oversimplifying the structure of trait networks and potentially omitting important hub or bridging traits (e.g., PNUE, PPUE, gs) that, although correlated, may serve distinct roles in physiological regulation.
(2) Intraspecific variation comparison: We aimed to assess intraspecific plasticity at the level of individual traits, which can differ significantly even within correlated trait groups due to trait-specific environmental sensitivity (e.g., PNUE vs. PPUE, δ¹³C vs. WUE). Aggregating traits would limit our ability to detect such patterns.
(3) Biological interpretability: While some traits are statistically correlated, they reflect different physiological mechanisms (e.g., gs reflects stomatal behavior; A represents carbon assimilation capacity). Including all traits allows a more nuanced discussion of functional strategies, particularly between the acquisitive and conservative species.To address potential concerns about clarity, we have revised the figure captions and main text interpretations to better highlight major trait patterns and to guide readers through the complexity of the trait relationships. We hope this balances analytical rigor with interpretability.
3. In the abstract you say that P. jezoensis is a treeline genus elsewhere. Does that mean that elsewhere it's on the treeline but not on your site? If so, it calls into question your study comparing a treeline and a non-treeline species.
[Response]: Thank you very much for this important and insightful comment. We acknowledge that our original statement may have led to confusion regarding the ecological role of P. jezoensis in our study region.
To clarify, while P. jezoensis is indeed a treeline-forming genus in other geographic regions such as Japan and the Russian Far East, it does not currently form the treeline on Changbai Mountain, where our study was conducted. Instead, B. ermanii dominates the upper treeline zone (1700–2200 m), while P. jezoensis is primarily found in the subalpine spruce-fir forest zone at lower elevations (1300–1800 m). Thus, in the context of this local elevational system, P. jezoensis functions ecologically as a non-treeline species.
Importantly, the objective of our study was not to compare species based on their functional types (e.g., conifer vs. broadleaf), nor to generalize treeline behavior across regions, but rather to explore how two locally dominant species occupying different elevational zones respond to environmental gradients near their upper distributional limits. We focus on how each species adjusts its leaf functional traits, intraspecific variation, and trait coordination in response to changing environmental conditions, especially in the context of climate-induced range shifts.4. The manuscript requires careful proofreading, as it contains numerous typos, errors, and repeated phrases (e.g., lines 87, 97, 121, 160, 225, 237). The introduction lacks clarity and does not effectively communicate the main objectives or ecological significance of the study. Strengthening the justification for the research questions and explicitly outlining their broader relevance would greatly enhance this section. Figure legends are overly succinct and should be expanded to allow readers to interpret the figures without referring back to the main text. In the Methods section, essential parameters related to the trait network analysis—such as modularity and closeness—are not defined. While some of these are included in the supplementary materials, they should be clearly explained and contextualized within the main text. Additionally, the ecological relevance of these network metrics should be supported with appropriate references and further discussed in the Discussion section to better integrate them into ecological theory. The Results section would benefit from more explicit descriptions of key findings, including numerical values drawn from the figures to improve interpretability. The Discussion, meanwhile, needs editing for clarity and coherence, and should include a section addressing the limitations and potential weaknesses of the study’s approach. Finally, the authors should place greater emphasis on the broader ecological or practical implications of their findings to strengthen the study's impact.
[Response]: Thank you for these detailed and constructive comments. We have carefully revised the manuscript to improve language clarity, eliminate repetitions, and enhance the overall logical flow. The Introduction has been refined to better articulate the study’s objectives and ecological significance, with improved justification for our research questions. Figure legends have been expanded and clarified, now providing sufficient context for interpretation without referring back to the main text. Key trait network metrics (e.g., modularity, closeness) are now clearly defined in the Methods, and their ecological relevance is further discussed in the Discussion with appropriate references. In the Results section, we added more numerical values to support key findings. Additionally, we included a brief paragraph addressing the study’s limitations and strengthened the discussion on broader ecological implications.5. Several times throughout the manuscript—particularly in the Discussion—you suggest that conifers, such as Picea jezoensis, are less adapted for upward range shifts. However, I find this conclusion somewhat overreaching, given that it is based solely on a limited set of adult physiological traits. Key processes related to establishment, such as seed production, dispersal mechanisms, germination success, and seedling survival, are not addressed in the study. Without considering these life history stages, it is difficult to draw robust conclusions about a species' capacity to expand its range in response to climate change. I would recommend framing these interpretations more cautiously or supporting them with broader evidence.
[Response]: Thank you for this important and insightful comment. We fully agree that drawing conclusions about species’ range expansion potential based solely on adult leaf functional traits has limitations, particularly when key life history processes—such as seed production, dispersal ability, germination success, and seedling establishment—are not directly measured.
In response, we have revised the relevant sections in Conclusion to frame our interpretation more cautiously, emphasizing that our conclusions about P. jezoensis reflect only its physiological trait-based responses near its upper elevational limit, rather than a comprehensive assessment of its migration capacity. We now explicitly acknowledge that successful upward range shifts depend on multiple factors beyond trait strategies, and we highlight this as a limitation of the current study. We have also added relevant references pointing to the importance of early life-stage processes in determining species distributional dynamics under climate change (Lines 563-570).6. Your results highlight two distinct adaptive strategies—resource acquisition and resource conservation—across the two species studied. However, given the complexity and number of traits analyzed, it can be challenging to synthesize the findings as a whole. I suggest including a summary figure or diagram in the Discussion section that visually distills the main results for each species and clearly contrasts their respective strategies. This would greatly enhance the reader’s ability to grasp the overarching patterns and takeaways of the study.
[Response]: Thank you for the valuable suggestion. In response, we have added a summary table (now included as Table 3) in the Discussion section to visually compare the key adaptive strategies of B. ermanii and P. jezoensis. This table synthesizes differences in resource use strategy, trait variation, coordination, and water-use patterns, making the main findings more accessible to readers.Minor issues:
1. it's best to avoid using acronyms in the abstract
[Response]: Thank you for the suggestion. We have revised the abstract to minimize the use of acronyms and now spell out key terms (e.g., “non-structural carbohydrates” instead of “NSC”) to ensure clarity and accessibility for a broader readership.2. L84: ref ?
[Response]: Thank you. We have added relevant references to support this statement. (Lines 93).3. L87: repetition of “functional traits”
[Response]: Done. Thank you very much.4. L96-99: lots of repetitions, improve the sentances
[Response]: Thank you for your suggestion. We have revised these sentences to improve clarity and eliminate repetitions, while maintaining the original meaning. The revised version now reads:
“In harsh environments such as polar and alpine regions, woody plants tend to exhibit lower connectivity and higher modularity in their trait networks (Rao et al., 2022), whereas in more favorable tropical conditions, higher connectivity and lower modularity have been observed (Flores-Moreno et al., 2019). These patterns suggest that reduced trait coordination and increased modular structure may be advantageous under environmental stress.” (Lines 102-107)5. L107-108: You say there are relatively few studies, but you don't cite any. Are you sure that no study has looked at the dynamics and physiological mechanisms of non-treeline species in response to climate change?
[Response]: Thank you for this important comment. In the revised manuscript, we have clarified the sentence to indicate that fewer studies have focused specifically on the responses of non-treeline species near their upper distribution limits, particularly in comparison to the extensive literature on treeline species. We also added relevant citations, such as Dong et al. (Dong et al., 2024), which addressed physiological mechanisms of P. jezoensis near its elevational limit. (Lines 115).6. L113: what do you mean by “integration”?
[Response]: Thank you for your question. In this paper, “integration” refers to the degree of coordination among multiple functional traits within a species. High trait integration implies strong interrelationships among traits that may function together as a coordinated strategy. To clarify this, we have revised the sentence in the manuscript to read:
“These comparisons will help us understand how treeline and non-treeline species respond to climate change in terms of trait means, trait plasticity, and coordination among traits (i.e., trait integration).” (Lines 118-120)7. L134: ‘a mean temperature of -7.3 to 4.9 °C in the growing season and annual precipitation of 800 to 1800 mm.’ Why do you give two figures each time? are they averages? quantiles? please specify.
[Response]: Thank you for the comment. Now,we have revised the sentence in the manuscript to clarify this as follows:
“The Changbai Mountain region experiences a growing season mean temperature ranging from −7.3 °C to 4.9 °C, and an annual precipitation between 800 and 1800 mm (Zhuang et al., 2017), reflecting typical climatic conditions across the mountain’s elevational zones.” (Lines 148-151).8. L140: You say B. ermanii is a treeline species, do you have a ref to support this?
[Response]: Thank you for your comment. We have added a reference (Du et al., 2018) to support the statement that B. ermanii is a treeline species on Changbai Mountain. (Lines 156).9. L142-145: Over what period were these figures calculated and what are the values for your sites?
[Response]: Thank you for this thoughtful comment. The lapse rate values cited (−0.68 °C per 100 m for temperature and +0.93% for relative humidity) are based on an empirical study by Reich et al. (1998), which reflects long-term average conditions in temperate montane forests. These values are not directly measured from our study plots but are used to provide general environmental context along the elevational gradient of Changbai Mountain. To avoid confusion, we have clarified this point in the revised manuscript and now state that these figures are regionally reported estimates rather than site-specific measurements. (Lines 163–167).10. In methods (2.1.), could you specify if the area is grazed or mown. If yes is it still the case? And if the forest is managed? These informations are important to understand the structure and the responses to climate change.
[Response]: Thank you for the insightful comment. We have added clarification in the Methods section (2.1) regarding land use and forest management. Specifically, we note that the study plots are located in a protected area of Changbai Mountain, where grazing, mowing, and active forest management are strictly prohibited. These areas are part of long-term ecological monitoring zones and remain under natural successional dynamics. This information has been included to better contextualize the vegetation structure and trait responses to climate variation. (Lines 167–172).11. I think we're missing a study site map showing the sites sampled by species and their elevation. It is also in this figure that you could put the temperature values of the environments of each species (S1). You could add pictures of the two species.
[Response]: Thank you for the helpful suggestion. We have added a new supplementary figure (Figure S1) showing the main distribution areas of B. ermanii and P. jezoensis, as well as the specific sampling sites and their corresponding elevations. This figure improves the spatial clarity of our sampling design and the context of species distribution.12. Figure 1: ‘environment’ corresponds to elevation ? the term needs to be changed. What's more, we don't know what altitude range it corresponds to, e.g. 1700-1800?
[Response]: Thank you for pointing this out. In the original figure, the term “environment” was used imprecisely to indicate sampling elevation. To improve clarity, we have replaced “environment” with the exact elevation ranges in the updated figure. In addition, we have revised the figure caption to specify the elevation intervals corresponding to each sampling group.13. I wonder if you run a PCA with the two species together, would that give two clusters corresponding to each species? Or would you still find altitudinal clusters?
[Response]: Thank you for this thoughtful comment. In this study, we conducted separate PCA analyses for B. ermanii and P. jezoensis to examine how leaf functional traits vary with elevation within each species, which aligns directly with our objective of understanding species-specific adaptive responses to elevational gradients. Since the two species differ markedly in functional type (deciduous vs. evergreen) and ecological strategy, a joint PCA would likely result in species-based clustering, which is less informative for our purpose.
Our focus is not on comparing species per se, but on assessing how traits shift within species across elevations. Therefore, we believe that separate PCA analyses better reflect the ecological questions posed and allow clearer interpretation of intra-species trait-environment relationships. For this reason, we did not conduct a combined PCA, which would not meaningfully advance the aims of this study.14. Table 2: What are F and p? You need to explain. What does ‘Elevation*species’ mean? It needs to be explained much better.
[Response]: Thank you for your valuable comment. In the revised manuscript, we have clarified the meaning of the F and p values, as well as the interaction term “Elevation × Species,” in the table caption. Specifically, F refers to the F-statistic from the linear mixed-effects model, and p represents the corresponding significance level. “Elevation × Species” indicates the interaction effect between elevation and species on each trait, showing whether the effect of elevation differs between the two species. We have updated the caption of Table 2 to include these definitions to improve clarity. (Lines 310–316).15. Figure 2: equations and p values are displayed only for significant regressions ? specify.
[Response]: Thank you for your comment. Yes, in Figure 2, regression lines, equations, and p-values are shown only for relationships that are statistically significant (p < 0.05), based on the results of the linear mixed-effects models presented in Table 2. We have now clarified this in the figure caption. (Lines 305–309).16. Figure 3: please add dotted ablines for certain % (25, 50, 75) to facilitate reading (panels a and b). For panel c, it's an average QCD per trait group? we don't understand what you've done here, please explain.
[Response]: Thank you for the constructive comment. We have updated Figure 3c by adding reference lines to improve interpretability. Additionally, we clarified in the figure legend that panel (c) presents the relative contribution of each trait group to the total intraspecific variation (QCD) within each species. This panel is intended to visually summarize group-level differences in trait plasticity, complementing the individual trait-level patterns shown in panels (a) and (b).17. Figure 4: I don't see the advantage of panels a and b over PCA. Justify it to me or put it in sup mat.
[Response]: Thank you for raising this important point. While PCA (Figure 1) provides a useful dimensionality reduction based on trait covariation, it summarizes multivariate structure without revealing the specific pairwise correlations or network topology among traits. In contrast, Figure 4a and 4b offer a trait network perspective, visualizing the significant pairwise correlations (edges) and highlighting how traits are functionally coordinated within each species.
This network approach reveals features not captured in PCA, such as:
(1) Inter-trait connectivity (i.e., average number of trait-trait links),
(2) Positive vs. negative correlations,
(3) Trait modularity and centrality, which are further explored in Figure 5.
Therefore, Figure 4 complements rather than duplicates PCA, and helps address our hypothesis on trait coordination (H2).18. Parameters of figures 4 and 5 need to be explained and ecological significances detailed and justified in methods.
[Response]: Thank you for the comment. We have added explanations of the parameters and their ecological meanings directly in the captions of Figures 4 and 5.19. S1 : “DBH” meaning ? same, “High” what is high? Is it soil/air temperature? Specify.
[Response]: Thank you for pointing this out. We have clarified that “DBH” refers to diameter at breast height and “High” refers to tree height. We also specified that the temperature was measured at canopy height. These clarifications have been added to the caption of Table S1.20. L368-369: ref ?
[Response]: Thank you for the suggestion. We have now added the appropriate references to support this statement (Lines 452).21. Could you detail in methods the ecology of the two species: pioneer species? Place in the ecological succession, mode of dispersal.
[Response]: Thank you for the helpful comment. We have added a brief description of the ecological characteristics of the two species in the Methods, Section 2.1 (Study Area). Specifically, we note that B. ermanii is a pioneer species often found in early-successional stages and is wind-dispersed, while P. jezoensis is a late-successional conifer with relatively limited seed dispersal, primarily via gravity and short distance of wind-bearing. These distinctions help contextualize their adaptive strategies and distribution patterns along the elevational gradient (Lines 156-163).22. L420-432: not just a difference between conifer and broadleaved species?
[Response]: Thank you for this insightful comment. We agree that the observed differences in trait correlations and regulation strategies may partially reflect the fundamental distinctions between coniferous (P. jezoensis) and deciduous broadleaf (B. ermanii) species. However, our analysis primarily aimed to explore how these species respond differently to elevational variations rather than to contrast functional types per se.
To address this concern, we have revised the text to explicitly acknowledge that part of the observed variation may result from differences in life-history attributes and phylogenetic differences, but we also emphasize that elevational gradients shape the strength and direction of trait coordination in both species (Lines 554-559).23. L434-435: “contrasting growth environments” in terms of what? Explain.
[Response]: Thank you for your thoughtful comment. In the revised manuscript, we have clarified that “contrasting growth environments” refer specifically to differences in temperature, light availability, and microclimatic stability across the elevational ranges of the two species. Betula ermanii, occurring near or above the treeline, is exposed to colder temperatures, stronger radiation, and greater environmental fluctuations. In contrast, Picea jezoensis inhabits lower elevations with milder temperatures, denser canopy cover, and more buffered microclimates.
However, we fully agree that the observed trait differences cannot be attributed to environmental filtering alone. As now stated in the revised Discussion, these patterns likely result from an interaction between external environmental constraints and intrinsic species-specific physiological strategies. This perspective emphasizes that species’ functional responses are shaped by both ecological context and evolutionary legacy—not simply by their position along an elevational gradient or by differences in leaf habit (Lines 554-559).
Dong, R., Li, N., Li, M.-H., Cong, Y., Du, H., Gao, D., and He, H. S.: Carbon allocation in Picea jezoensis: Adaptation strategies of a non-treeline species at its upper elevation limit, Forest Ecosystems, 11, 100188, https://doi.org/10.1016/j.fecs.2024.100188, 2024.
Du, H., Liu, J., Li, M. H., Büntgen, U., Yang, Y., Wang, L., Wu, Z., and He, H. S.: Warming‐induced upward migration of the alpine treeline in the Changbai Mountains, northeast China, Global Change Biology, 24, 1256-1266, 2018.
Flores-Moreno, H., Fazayeli, F., Banerjee, A., Datta, A., Kattge, J., Butler, E. E., Atkin, O. K., Wythers, K., Chen, M., Anand, M., Bahn, M., Byun, C., Cornelissen, J. H. C., Craine, J., Gonzalez-Melo, A., Hattingh, W. N., Jansen, S., Kraft, N. J. B., Kramer, K., Laughlin, D. C., Minden, V., Niinemets, Ü., Onipchenko, V., Peñuelas, J., Soudzilovskaia, N. A., Dalrymple, R. L., and Reich, P. B.: Robustness of trait connections across environmental gradients and growth forms, Global Ecology and Biogeography, 28, 1806-1826, https://doi.org/10.1111/geb.12996, 2019.
Rao, Q., Su, H., Ruan, L., Xia, W., Deng, X., Wang, L., Xu, P., Shen, H., Chen, J., and Xie, P.: Phosphorus enrichment affects trait network topologies and the growth of submerged macrophytes, Environmental Pollution, 292, 118331, https://doi.org/10.1016/j.envpol.2021.118331, 2022.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 248 | 55 | 20 | 323 | 28 | 23 | 42 |
- HTML: 248
- PDF: 55
- XML: 20
- Total: 323
- Supplement: 28
- BibTeX: 23
- EndNote: 42
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1