The Variation in Water Consumption by Transpiration of Qinghai Spruce among Canopy Layers in the Qilian Mountains, Northwestern China

Round 1

Reviewer 1 Report

The reviewer was just read and checked the manuscript entitled as "The variation in water consumption by transpiration of Qinghai spruce among canopy layers in the Qilian Mountains, northwestern China", written by Wan et al.

The datasets obtained at the Qinghai spruce forest is highly estimated and interesting. However, some significant issues are found in it as follows.

The most serious (and possibly fatal) issue is that many of lack of data were found in daily transpiration in both 4-month periods (July-Oct.in 2015 and 2016), according to the lowermost panels of Figure 2, despite the authors determined the cumulative transpiration and mean daily value for each period. The authors must clearly describe how they reasonably interpolate the missed data and which tree(s) caused the lack of data.

Lack of sample number of trees is another great issue. 2 or 3 trees might not be enough to be represented the transpiration from each layer, even if 13 trees might be seemingly enough for estimating stand transpiration. The authors should refer to some previous studies to reconsider/examine the adequate sample number. One idea is to combine "Inter mediate" and "Suppressed" layers (to make Inter mediate- Suppressed category): Please consider.

Number of sensors for each tree was also large problem. Usually in previous studies, two or more sapflow sensors were installed for a tree having >2cm sapwood thickness, and sometimes sensors were installed from two directions, for example from north and south directions.

Although it is not the practical issue, there are some inconsistency in focus and background of this study, which may induce suspicions for worldwide readers.. From global scale, deforestation has been major problem and afforestation is considered the counter activity to it, to prevent environmental deterioration and/or increase CO₂ absorption. Afforestation may cause a problem, but it becomes each local issue and the background should not be confused with global problem.

Furthermore, from L60, Qinghai spruce forest is native vegetation (not afforestation vegetation) which is usually recognized better to be preserved if considering global environment. If there is some local necessity to manage such natural forest by clear cutting or thinning, what kind of local human activity conflicts to the existence of the forest should be described in detail. Anyway, what you (and we) should consider is how the precious forest will be kept under rapid climate change, and maybe, how one make human activities coordinate with the forest.

The followings are comments on details:

Introduction (L39-74) should be thoroughly modified to present reasonably why the authors and readers should grasp the manner of Qinghai spruce transpiration.

L144-149 Numerical threshold to classify a tree to a category should be declared, based on the results of Table 1. For example, the authors should show the reason that they determined No.5 was "Dominant" while No.7 was not.

The reviewer recommends to divide trees to 3 categories (e.g., 1-5, 6-9 and 10-13) to increase sample trees in each category.

L161-168 As above description, the missed data seemed occurred even in July and October in 2015 and August-September and October in 2016. describe the situation in detail and show a reasonable interpolation method if necessary.

L174-179 If the authors have an idea to estimate the sapflux in more than 20mm depth area, please add it. If not, the large trees' transpiration values should be clearly presented, as "the 0-20mm sapwood transpiration".

For all the instruments applied, manufacturers and their country must be presented consistently and carefully.

For example, description of SF-L was duplicated (L152, L165), while information for some instruments were not described (L104, L189-192). “LI-COR Inc.” and "Campbell Scientific Inc." are more accurate than "- Company " (L104, L192). One of "USA", "US and "United States" should be selected (L104, L192, L204-205, L219).

L225-230, L375-L406

The reviewer considers that the slope of regression equation does not necessarily mean the sensitivity to PET and VSM, since a transpiration from a larger tree usually become larger. To discuss the sensitivity, transpiration value should be normalized by sapwood area of each tree. (It would be clear when considering the comparison of sweat quantitiy of an animal with temperature. The slope of an elephant data would  larger than that of a small dog data, but it does not mean an elephant is more sensitive to temperature than a small dog. In this case, to discuss the sensitivity, the quantity of sweat should be normalized by each body surface area.)

As for VSM, please note that the transpiration of small trees almost stopped when the soil water content becomes 0.15, whereas the transpiration of large trees became smaller but continued.

This seems suggesting that, when severe drought occur, large trees can take water and keep photosynthesis activity with controlling transpiration to some extent, while small trees are more likely to die.

And this may mean that, if extreme drought occurs after cutting the large trees, the small tree will also die, and the precious forest itself will be destroyed, and soil erosion will occur easily. Please consider carefully with co-authors,

and rebuild the discussion.

L337-344 Please also examine which individual environmental factors consisting PET would more significantly affect on transpiration (VPD, Rn, U or T?)

L407-426 Descriptions about forest management must be reconsidered carefully.

Fig. 8

Instead of comparison of slope values, it is recommended that the result of each layer is presented in each panel. Standard deviation should be also presented with error bar in each panel.

Table2

It should be clearly divided which tree belongs to what category: Maybe No.6 was "Codominant"- right?/ Was No.9 "Codominant"?

Did "Sapwood area" contains >2cm area? If so, please add sapwood thickness. otherwise, please show it with a notification

The reviewer expects that those kind of issues reasonably revised. After that, this manuscript will be achieved the standard to be published “Forests”. Thanks.

Author Response

Response to comments

Dear Sir/Madam,

    Thank you so much for your kind encouragement on our manuscript entitled “The variation in water consumption by transpiration of Qinghai spruce among canopy layers in the Qilian Mountains, northwestern China” (Forests-826981) and sending us the review on our manuscript.

    We greatly appreciate you for your valuable comments and constructive suggestions and have carefully addressed these points in the new version.

    A point-to-point response to the comments has been detailed in our Response Letter, in which the comments are written in italics followed by our responses in regular text. In addition, the changes or revision were marked with red text in the manuscript.

    Now, we are resubmitting the latest vision of our manuscript to you.

Best wishes,

Pengtao Yu

Author Response File: Author Response.docx

Reviewer 2 Report

Review Forest 826981, Wan et al.

This paper addresses an interesting question of effects of forest structure on canopy transpiration with the purpose to guide thinning to save water loss in evapotranspiration and increase water yield. The case study from Northern China with uniform stand composition (single species) and a two years study is also advantageous.

However, the paper seems to oversimplify rather than properly develop the topic. I hope they can do this in revisions.

Firstly, it seems that this is a continuation of an earlier report of the same study of Wan et al 2017 in Acta Ecol Sin and this should be clarified noting what is the new information.

But more importantly, the main argument that this study investigates the transpiration (T) at different canopy layers (vertically) needs to be clarified. To me this study mainly shows that large trees with large canopy transpire more:

1. It is mainly a T vs size. Therefore, presentation of the relationships of T vs sapwood area, density, DBH, LAI etc. can be expected, but not presented or explored for hierarchy, interactions, etc..
2. Discussing vertical forest layers would require some information on vertical variations in conditions (PAR, VPD, temperature, LAI) across this vertical dimension but it seems there are only met data from on station in a nearby open site.

It seems that examining the deviations in the simple relationships in “A” that could be explained by data in “B” would allow the authors to develop their story. But I am not sure the current paper and data presented allow this.

It also shows that T is related to both soil moisture and VPD. This is of course expected. Quantifying these relationships is interesting if it would be compared to other species or climatic conditions. In fact, the interesting aspect of this part is the apparent interactions between soil moisture vs VPD effects across tree sizes. Small trees seem to be limited more by soil moisture and less sensitive to VPD. This transition from VPD dominance to SWC dominance could be further developed.

The issue of what causes the large differences in tree size is also not discussed or explored. Is this related to genetic variations, or to sub-surface spatial heterogeneity imposing moisture limitation on the trees? The authors seem to argue the inverse, that small tree take up less water and nutrients, without explaining this. It seems that this cause-effect issue could be further explored.

On a more technical level, the Abstract should be self-sustained and so what are dominant and co-dominant layer should be noted there. Also, in the abstract the link between the “results” and “conclusion” is not clear.

The introduction could develop more the discussion on forest structure, vertical gradient in conditions and in fluxes.

In the Methods, some definitions are missing like ‘canopy density’ ‘canopy thickness’.

Number of Sap flow measurements in the tree height groups is very small (1-2 in some cases) which limit the reliability of the results. Perhaps some validation, testing of the SF measurements or the hydrological budget would help.

In summary, I find the topic and the case study are interesting and appropriate for the journal. However, the paper in its present form does not develop the topic sufficiently, does not clearly indicate the addition over the previous report, and is missing some key data on vertical dimension. If it is a meant as a simple report for forest management for developing thinning strategies, it may be modified to avoid the problematic discussions noted above, but in that case I cannot judge if this journal is the appropriate venue.

Author Response

Response letter to comments

Dear Sir/Madam,

    Thank you so much for your kind encouragement on our manuscript entitled “The variation in water consumption by transpiration of Qinghai spruce among canopy layers in the Qilian Mountains, northwestern China” (Forests-826981) and sending us the review on our manuscript.

    We greatly appreciate the reviewers for your valuable comments and constructive suggestions. We have carefully addressed these points in our revised manuscript and a point-to-point response to the comments has been detailed in our Response Letter, in which the comments are written in italics followed by our responses in regular text. In addition, the changes or revision were marked with red text in new version of the manuscript.

    Now, we are resubmitting the latest vision of our manuscript to you.

Best wishes,

Pengtao Yu

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

It can be recognized that the authors revised well with sincere attitude, and the MS is now close to be accepted.

However, the large problem still remains in the way to assume 20-40 mm sapflux. The reviewer is afraid that the "simply double" assumption which the authors adopted in the revised MS may set a bad precedent for future observation studies.

Further, the reviewer considers the objective of this study is still apart from what was done in this study, even after the revision. The details and a suggestion are as follows.

The authors insisted "We couldn’t agree with you more that natural forest should be conserved rather than artificial management, e.g., thinning."

However, the reviewer does not hope discussion about it but hope consistency in the Introduction. Consistency is important anyway: Surely overdense planted forest is problematic because it may induce water use conflict in the semi-arid region, but the Qinghai spruce forest itself is not planted forest and not usually considered as "overdense". Thus, if the authors insist as such, they must clearly describe one of the followings:

-- Reasonable background that the Qinghai forest is better to be managed (e.g., the forest has been degraded recently under rapid climate change; or the forest water consumption seems increased recently to threaten human life/activity), with some research results.

-- The relationship between layer-classified transpiration of the Qinghai forest and management of overdense forest.

Or, another reasonable aim of this study may be adopted: For example, the last sentence of Introduction can be changed as:
"Our work will contribute to providing information to consider whether the natural Qinghai spruce forest should be preserved as it is or partially managed, and how it is managed if it should be done so, under rapid climate change".
In this case, the later half of discussion 4.3 can be start as "To decrease water consumption by transpiration in forests, thinning of artificial forest has been the first choice and strongly advocated in the dryland regions of China."

Please reconsider.

L117 Please add the model number of the plant canopy imager (or "plant canopy analyzer"?) .

L178-184 Please rebuilt the sentence, for example " The azimuthal variation in sap flux density of each tree was related with tree crown architecture [35] and crown exposure to the sun [36,37], both of which were thought to be closely related to the canopy density [5]. When the canopy was highly closed, the azimuthal variation in sap flux density was thought to be negligible [5]. In this study, highly closed forests (with canopy densities of 0.55) was chosen as research plot, thus we assumed the azimuthal variation was small for each tree."

Sensor insertion direction was not needed since it was already explained in L167-168. Plus, please describe the definition of "canopy densities" value.

L185-190 The reviewer consider it is unreasonable that Js in the 20-40mm was equal to that in the 0-20 mm. If some study support the assumption, please present it as a reference.Or, please apply "Gaussian distribution" or clearly declare as "only 0-20 mm transpiration was considered".

L368-371 Please note that lower trees might be also sensitive to PET, if PET is determined from the meteorological factors measured just above the lower trees.(The meteorological factors were measured at the open area whose condition would be similar to over the large trees' canopy).

Figure 2
The interpolated values of stand transpiration should be presented in the lowermost panels by using dotted lines.

Figure 8
The reviewer intended to prepare for "one panels for one layer's result", Then total 6 panels would be included in Figure 8. It will make comparison of Kd clear. Please consider.

Figure X
The result of linear regressions to interpolate daily transpiration must be shown by some figures which may be included in Appendix.

(That's all.)

Author Response

Dear Sir/Madam,

    Thank you so much for your kind encouragement on our manuscript entitled “The variation in water consumption by transpiration of Qinghai spruce among canopy layers in the Qilian Mountains, northwestern China” (Forests-826981) and sending us the review on our manuscript.

    We greatly appreciate the reviewers for your valuable comments and constructive suggestions. We have carefully addressed these points in our revised manuscript and a point-to-point response to the comments has been detailed in our Response Letter, in which the comments are written in italics followed by our responses in regular text. In addition, the changes or revision were marked with red text in new version of the manuscript.

    Now, we are resubmitting the latest vision of our manuscript to you.

Best wishes,

Pengtao Yu

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

The reviewer was just checked the revised manuscript.

It is well modified and the reviewer agrees it to be accepted.

Only one part is better to be changed in Line 434: "we will pay more attention ..." --> "more attention should be paid..." 

Anyway, Congratulations!