Seasonal and Daily Xylem Radius Variations in Scots Pine Are Closely Linked to Environmental Factors Affecting Transpiration
Round 1
Reviewer 1 Report
Please include some images during data collection or just show a photo of how Dendrometer record data. This should add more quality to your methodology.
Author Response
Corresponding photographs were added as Supplementary Material (Figure S1).
Reviewer 2 Report
Oberhuber et al. used point dendrometers, that are commonly used in the research community, to measure radial variation in Scots pine trees on a drought-prone site. They were able to establish the very important and innovative link between phloem and xylem radial irreversible variation. Understanding this process of how phloem and xylem interact is essential when trying to shed light on the effects of drought and heat on pine xylem radial variation besides the irreversible radial displacement during wood formation.
Overall the manuscript is perfectly written, has a truly innovative character and is an important step closer in the understanding of hydraulic functioning of trees. However, there are points, which could be improved: The statistical analysis could be improved to empower the findings. The structure of the results could be improved. 3.1 is fine but then in 3.2 appears the VPD and precipitation, but also in 3.3 environmental influence is the focus. Also the temporal level of the data is sometimes unclear. Daily, seasonal, detrended,.. This should be clarified in the method section: which parameters exist in this study in which resolution and what was done to them and which relationships have been tested.
The simple summary and the abstract nicely complement each other.
Introduction: The introduction is comprised well.
Methods: The methods section includes all necessary information.
The sample size is quite low… did you check the result power somehow?
2.3: The idea behind dividing xylem and phloem is innovative and definitely points into the right direction. By measuring radial displacement 2-3cm into the sapwood, you automatically exclude all developing xylem tissues (enlarging xylem cells) from your xylem data. However, exactly this tissue has the main potential for reversible xylem radial changes. How do you overcome this issue without having a bias in your results? I would not classify developing xylem cells as “inner bark”. You could consider dividing the “inner bark” data into bark and phloem tissue and developing xylem… this you could do by taking microcores during the vegetation period to get an idea of the changing share of bark tissue (including developing phloem) and developing xylem. Hpwever, since this would mean intensely more work you could consider in renaming our two measurements.
2.4: Just because the data is not normally distributed does not mean, that no other distribution or some data transformation could help in finding the right model. Maybe a gamma distribution could fit to your data.
Are the spearman corr coff determined for each environmental variable separately? Is there the option of an index to find out the potential mixed signal of temperature, vpd, p and swc?
Results:
L234ff: Unfortunately, the displayed dry and wet periods have not been on the same date. Therefore, the wood formation is in different status, and I don’t know, if you should compare both dates without including this information somehow.
L289 and Graph 6e: I really doubt the negative relationship is valid. Please consider the scattering of the data, the few influential points at the higher end of P and if you could really state, that (with great significance) the xylem daily amplitude is really decreasing with increasing precipitation.
Discussion:
L333ff: You tested not only the non-lignified living phloem cells in your study but also the non-living cambial zone and most importantly the non-lignified xylem cells, didn’t you?
How do you think, the shrinking and the subsequent re-swelling of the lignified xylem could also be due to the enlarging cells pushing the inner xylem cells?
Please also compare your findings to the study of Steppe et al 2015 (https://doi.org/10.1016/j.tplants.2015.03.015) and Stangler et al 2021 (https://doi.org/10.3390/f12030274).
L352: An interaction model could help here to rank the environmental variables influence maybe? Single spearman coefficients induce that the environmental variables can be separately discussed.
Author Response
Oberhuber et al. used point dendrometers, that are commonly used in the research community, to measure radial variation in Scots pine trees on a drought-prone site. They were able to establish the very important and innovative link between phloem and xylem radial irreversible variation. Understanding this process of how phloem and xylem interact is essential when trying to shed light on the effects of drought and heat on pine xylem radial variation besides the irreversible radial displacement during wood formation. Overall the manuscript is perfectly written, has a truly innovative character and is an important step closer in the understanding of hydraulic functioning of trees. However, there are points, which could be improved: The statistical analysis could be improved to empower the findings.
Response: Stepwise multiple regressions were run to predict XRV from several environmental variables given in Table 2 as requested. A new table (Table 3) showing relevant models and statistics was added in the manuscript.
The structure of the results could be improved. 3.1 is fine but then in 3.2 appears the VPD and precipitation, but also in 3.3 environmental influence is the focus.
Response: Yes, we agree and therefore we moved Figures 4 and 5 and the corresponding text to chapter 3.3. as suggested.
Also the temporal level of the data is sometimes unclear. Daily, seasonal, detrended,.. This should be clarified in the method section: which parameters exist in this study in which resolution and what was done to them and which relationships have been tested.
Response: We have reviewed the methods section for understanding and clarity and believe that the points made are already outlined:
Lines 110ff: All recorded environmental variables and their resolution used in the analysis are described.
Lines 151ff and 158ff: The method used to eliminate seasonal trends in dendrometer records and the tested relationships between environmental variables and xylem radial variations, respectively, are described in detail.
The simple summary and the abstract nicely complement each other. Introduction: The introduction is comprised well. Methods: The methods section includes all necessary information. The sample size is quite low… did you check the result power somehow?
Response: We agree that the sample size is low (n=4), but dendrometer studies frequently have a sample depth of less than 5 trees. In Figure 2 records of inner bark and xylem variations are depicted including standard errors. Individual records show high agreement in daily variations of inner bark and the xylem, which is supported by the close relationship between detrended inner bark and radial xylem variations (Figure 3). Absolute values are different among sample trees (see SE in Figure 2), which is, however, not unexpected for point measurements of radial variations, e.g., there are sometimes large differences in xylem formation among trees using microcores. Although daily xylem radial variations are rather low compared to radial growth (i.e., inner bark variations) and amounting to predominantly less than 50 µm (cf. Figure 4), we found close agreement of xylem radial variations with environmental variables (R2 of multiple regression models ranged between 0.608 to 0.700; see new Table 3) indicating that our data set has a high degree of significance.
2.3: The idea behind dividing xylem and phloem is innovative and definitely points into the right direction. By measuring radial displacement 2-3cm into the sapwood, you automatically exclude all developing xylem tissues (enlarging xylem cells) from your xylem data. However, exactly this tissue has the main potential for reversible xylem radial changes. How do you overcome this issue without having a bias in your results?
Response: Yes, of course the dendrometer sensor recording xylem radial variation (i.e., sensor positioned on a screw) excludes developing xylem tissues. The developing xylem (and phloem cells) were recorded by the dendrometer mounted on the bark (see also new Figure S1). By subtracting xylem radius variations (i.e., radial variations of the sapwood) from bark variations, variations in “inner bark tissue”, which included living phloem cells, the cambial zone and newly developed xylem and phloem tissue, were obtained. I am also not aware of any study reporting “shrinkage” of the xylem, i.e., the width of enlarging xylem cells is decreasing due to high transpiration during the growing season.
I would not classify developing xylem cells as “inner bark”. You could consider dividing the “inner bark” data into bark and phloem tissue and developing xylem… this you could do by taking microcores during the vegetation period to get an idea of the changing share of bark tissue (including developing phloem) and developing xylem. Hpwever, since this would mean intensely more work you could consider in renaming our two measurements.
Response: For reasons of consistency with the definitions used in the literature (e.g., Pfautsch et al. 2015; Sevanto et al. 2002, 2003; Mencuccini et al. 2013), we refrained from renaming them.
2.4: Just because the data is not normally distributed does not mean, that no other distribution or some data transformation could help in finding the right model. Maybe a gamma distribution could fit to your data.
Response: In our study we were primarily interested in determining environmental factors affecting xylem radial variations in Scots pine under drought. In our revised version of the manuscript we now also included multiple regression analyses as suggested. Both analyses revealed that environmental factors affecting transpiration are closely related to changes in xylem radius. We agree that data transformation might also be appropriate to develop a compelling model. Because multiple regression analyses (see results section, new Table 3) confirmed Spearman correlations and produced high R2 ranging from 0.608 to 0.700 (see Table 3), we refrained from data transformations.
Are the spearman corr coff determined for each environmental variable separately? Is there the option of an index to find out the potential mixed signal of temperature, vpd, p and swc?
Response: Yes, Spearman correlations were determined separately for each environmental variable. As requested, a stepwise multiple regression analysis was added (see new Table 3 and text).
Results: L234ff: Unfortunately, the displayed dry and wet periods have not been on the same date. Therefore, the wood formation is in different status, and I don’t know, if you should compare both dates without including this information somehow.
Response: As suggested we have added the following note to the legend of Figure 4: Both periods shown are within the growing season of Pinus sylvestris in the study area (see Gruber et al. 2010, reference in previous submission #42).
L289 and Graph 6e: I really doubt the negative relationship is valid. Please consider the scattering of the data, the few influential points at the higher end of P and if you could really state, that (with great significance) the xylem daily amplitude is really decreasing with increasing precipitation.
Response: We also doubted the close relationship found between precipitation and xylem amplitude, but the Spearman correlation shown in Fig. 6e (ρ=-0.477) was checked again and verified using two different statistical softwares (SPSS and Statistica).
Discussion: L333ff: You tested not only the non-lignified living phloem cells in your study but also the non-living cambial zone and most importantly the non-lignified xylem cells, didn’t you?
Response: Yes, you are absolutely right, and we thank you for bringing it to our attention. We corrected the sentence as follows:
…which can occur in the much more elastic non-lignified tissue including living phloem cells, the cambial zone and enlarging xylem cells.
How do you think, the shrinking and the subsequent re-swelling of the lignified xylem could also be due to the enlarging cells pushing the inner xylem cells?
Response: In our opinion, it is highly unlikely that the cell turgor in expanding (non-lignified) xylem cells will cause lignified xylem cells to be compressed. On the other hand, it is well known that xylem shrinkage is caused by transpiration and changes in xylem water potential (please see text for more details).
Please also compare your findings to the study of Steppe et al 2015 (https://doi.org/10.1016/j.tplants.2015.03.015) and Stangler et al 2021 (https://doi.org/10.3390/f12030274).
Response: Thank you for pointing us to the publication by Steppe et al. (2015). This important opinion article has been included in the manuscript. On the other hand, Stangler et al. (2021) is not an appropriate study to compare with our analysis for the following reasons: different species (Norway spruce), elevational transect studied rather than drought stress (our study), point dendrometer mounted over bark rather than on the xylem (our study). Furthermore, we would like to note that xylem development (xylogenesis) of Pinus sylvestris was already analyzed within the study area using microcores (see Gruber et al. 2010, reference in previous submission #42).
L352: An interaction model could help here to rank the environmental variables influence maybe? Single spearman coefficients induce that the environmental variables can be separately discussed.
Response: As requested, a stepwise multiple regression analysis was added (see new Table 3 and text).
Reviewer 3 Report
The manuscript entitled ‘Seasonal and daily xylem radius variations in Scots pine are closely linked to environmental factors affecting transpiration’ presents analyses of the correlation of xylem radius variations based on studies using a dendrometer.
Merits
The manuscript is written coherently and logically.
The aim of the work and the hypothesis are clearly presented and enough supported by results and discussion.
Discussion briefly explains the variations in the results obtained.
Critique and suggestions
All the figures are low resolution. Reading them may not be comfortable, but it is still possible. For the final version I recommend including high resolution figures.
The study area shows the total precipitation (724 mm l. 97), while the precipitation in Table 1 refers to the studied period (growing seasons, April-September). It is hard to equate the studied period (Ap-Sep) to the mean precipitation from all the meteorological records. I suggest also presenting full-year precipitation for 2019-2021, for informational purposes.
Figure 1–in soil water contents are gaps in missing data (blue line), it is worth explaining what they are due to.
Based on the arrangement of the points in the figure, correlation (-0.477) seems unlikely (it is quite high), I suppose it should be lower. Please verify as this is just a speculation.
Table 2 shows a summary of individual correlation coefficients. Such an overview is valuable information, but to draw the more general conclusion that the title of the manuscript implies, it is worth constructing an overall model. To get an accurate understanding of the impact of a variable, the model must take into account the interactions and contributions of each variable to the overall model. Such a model could be, for example, a backward or stepwise regression. This is just a proposal for you to think about.
Author Response
The manuscript entitled ‘Seasonal and daily xylem radius variations in Scots pine are closely linked to environmental factors affecting transpiration’ presents analyses of the correlation of xylem radius variations based on studies using a dendrometer.
Critique and suggestions
All the figures are low resolution. Reading them may not be comfortable, but it is still possible. For the final version I recommend including high resolution figures.
Response: High resolution figures were submitted as requested.
The study area shows the total precipitation (724 mm l. 97), while the precipitation in Table 1 refers to the studied period (growing seasons, April-September). It is hard to equate the studied period (Ap-Sep) to the mean precipitation from all the meteorological records. I suggest also presenting full-year precipitation for 2019-2021, for informational purposes.
Response: As suggested, total precipitation sums in study years 2019-2021 were added in the text.
Figure 1–in soil water contents are gaps in missing data (blue line), it is worth explaining what they are due to.
Response: The following note has been added to the legend of Figure 1: Missing soil water content data are due to logger failure.
Based on the arrangement of the points in the figure, correlation (-0.477) seems unlikely (it is quite high), I suppose it should be lower. Please verify as this is just a speculation.
Response: We also doubted the close relationship found between precipitation and xylem amplitude, but the Spearman correlation shown in Fig. 6e (ρ=-0.477) was checked again and verified using two different statistical softwares (SPSS and Statistica).
Table 2 shows a summary of individual correlation coefficients. Such an overview is valuable information, but to draw the more general conclusion that the title of the manuscript implies, it is worth constructing an overall model. To get an accurate understanding of the impact of a variable, the model must take into account the interactions and contributions of each variable to the overall model. Such a model could be, for example, a backward or stepwise regression. This is just a proposal for you to think about.
Response: Stepwise multiple regressions were run to predict xylem radial variations from several environmental variables given in Table 2 as requested. A new table (Table 3) showing relevant models and statistics was added in the manuscript.
Round 2
Reviewer 2 Report
Dear authors,
thank you very much for this detailed rebuttal. All comments and questions from my side have been answered and uncertainties have been explained.
Kind regards.
