Spring Moisture Availability is the Major Limitation for Pine Forest Productivity in Southwest China

Round 1

Reviewer 1 Report

General comment

The article Yingfeng et al ”Temperature induced drought stress is the major limitation for pine forest productivity in southwest China” is on the scope to the journal. However, it should be substantially revised. I have two fundamental general comments.

1. The result cannot be a hypothesis, since the hypothesis confirms or refutes the proposed mechanism based on the result (lines 102-103).
2. There is no connection between the observed sharp changes in the productivity of trees and their climatic response to temperature and precipitation. It means that. From the materials (introduction and description of the study area) it follows that the authors study the radial growth of fairly young trees in the stands created by the natural insemination of the territory after clear-cutting. From the data of table 1 it also follows that the authors are dealing with approximately the even-age stands. Therefore, sharp changes in normalized productivity may have completely different reasons, such as achieving closeness and increasing competition for both light and competition of root systems for “living space” and, therefore, in deficient moisture in the soil. Such (and other forestry) reasons are not considered by the authors, and therefore are not taken into account in the analysis. Therefore, the second dendroclimatic part of the analysis, as it presented in the paper, hangs, which simply confirms the sensitivity of pine growth to precipitation (positive) and temperature (negative). By the way, the climatic response (in Fig. 4) is difficult for visual perception, and it was possible to show the significance of the leading climatic factors more clearly, using daily data as well see, for instance, Belokopytova et al., Contemp Problems of Ecol., 2018). The logical flaw in the article cannot be improved by any sophisticated statistics.

There are a lot of negligence in the text of the article. I especially note the one that occurs twice in the text: (line 32 and line 276-277: “…the productivity of both species begins to decline at the middle of this century…). But really it happen at the beginning of this century (2002 and 2006 yy.)

 

Specific comments:

Line 24: What is a difference between anthropogenic  and natural greenhouse gasses in the contest of " carbon sink ability of pine forest in "? It is better to delete "anthropogenic".

Line 139 and Figure 2: The Figure 2 should be clarified. For example, what is a variation of temperature in dry season during June-Oct (Fig. 2a)? The green curve simply disappears during these months.

Lines 188-197: Taking into account that the authors worked with young trees (see Table 1) how the authors remove the age-depended effect in case of using BAI time series?

Lines 198-215: The authors should clarify:

1. What was a specification of the model? Which were variables selected as dependent and independent variables?
2. How the authors estimated the number of independent variables (730 variables) and a number of observations?
3. Why the authors forgot about classsical response function analysis (Fritts et al., 1976) and principal component technigue taking into account "high autocorrelation among daily temperaturesand precipitation" (otherwords, high collinearity between independent variables)? Why is the classical approach in dendrochronology not appropriate?
4. Why did the authors use the PLS regression? Why the PLS is better than mixed models or simple multiple regression taking into account that the PSL technigue is not often used in dendrochronological research?
5. Having all required daily climate data and tree-ring indexes why the authors did not apply process-based simulations (e.g. VS-model) successfully used in different parts of China? Such simulation can get a clear answers on the most of MS questions.

I strongly recommend the authors to introduce a graph-chart (or other picture) explaining visually how PSL is used  in the data analysis. Moreover, by itself the PSL regression using in the work should be clearly explained with much more details!

Line 233 and Figure 3: Could the authors explain what is a difference between well-known age-related nonlinear trend for young trees of most conifer species and their results indicated in the Fig.3? Could the authors proof that the obtained BAI curves are not also related with age-depended effect?

Line 250 and Figure 4: Could the authors clarify what does VIP statistics mean  with corresponded formulas to estimate, distribution estimation, etc.? The potential readers should undestand why the value 0.8 is crutial! The reference provided is not enough.

My final suggestion the manuscript in the recent form should by revised with a possibility of re-submission.

Comments for author File: Comments.pdf

Author Response

Please kindly see the attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report

General comments

The manuscript is a focused work on the growth of two dominant species - P. yunnanensis and P. armandii - in a large-scale pine forest community of southwest China, predominantly from economic view. The authors sample and analyze the plant component in large detail, but in contrast, they make strong simplifications on the environmental aspect. High resolution (daily) precipitation and temperature data are included among the potential impact factors for plant growth, but all other enviromental variables are ignored, such as extra-climatic (local) properties. In addition, as also shown in Figure 3, precipitation and temperature values varied with different dynamics and characteristics in the long term trends. The authors do not reveal any possible relationship between the two climatic parameters at the model level, so I do not consider appropriate the term 'temperature induced' presented in the title. This is likely to be the case, but it should be substantiated in much more detail and analytical terms, that I do not currently find in the manuscript. I also note that the beta values, which can be the real output basis for comparability of the species and effectivity of parameters, are completely ignored in the regression analyses. I strongly suggest the authors re-viewing the analytical methods, the results and conclusions, and a more appropriate title for the manuscipt.

Authors mention the importance of mixed forests and emphasize their significance in forest ecosystems. These statements and priorities are really important, but the reviewer asks: 1) Is this really the case with the two studied (contrasted) tree species, and 2) How do the authors consider possible biotic and abiotic interactions and their implications for interpreting the phenomena? A further question that arises is whether a small-scale intensive sampling of the pine forests presenting a large distributional area was more appropriate than a wider selection of individuals that represents the environmental diversity more appropriate? I suggest that the authors reconsider the work along the lines of the issues indicated and improve the content of the manuscript accordingly.

Detailed comments:

Line 2 – need to overview the term ’temperature induced drought stress’ (see general comments) and also throughout the whole manuscript, consistently

Lines 32-39 – primary results and conclusions are questionable in the light along the separate analyses on precipitation and temperature changes and also without of standardized beta values (see also general notes)

Lines 64-65 - this statement is not entirely true and cannot be linked to the subject of the manuscript, I suggest deleting it

Lines 101-104 – in the working hypothesis, it is worthwhile to make a definite distinction between the options for forests and for each tree species, please specify this statements

Line 178 – please specify the ’high agreement’ (e.g. fitting level, correlation, etc.)

Line 182 – please specify the meaning ’better’ (e.g. reliable, adaptable, etc.)

Line 199 – please specify the detailed meaning ’high autocorrelation’
Line 212 – ’Due to PLS only identified …’ did YOU identify the critical periods? If so, please state it clearly

Line 225 – please specify the kind/type of ’information’ in Table 1

Lines 228-231 – I find it necessary to refine and improve the results reported in Figure 3, additionally taking into account the standardized beta values

Line 232 – report the standardized beta values for each regression, and rescale the X axis appropriately, and please correct the scientific name P. armandii

Lines 237-248 - please restructure this chapter more consistent, and it is sufficient to mentions the figure references once in each paragraph (Fig. 4a, Fig. 4b) referring to the main differencies between the species

Line 249 - refine subfigure caption 'effects of temperature' and 'effects of precipitation' - it does not express the content shown

Line 272 – check the match between Line 5 in Table 2 and middle line subfigure in Figure 4, and correct the mismatch

Lines 273-337 – the chapter discussion needs a major overview: figure references are not need to be indicated, but adequate scientific references need to be inserted

Lines 342-343 – this statement is incorrect because it is a frequently studied ecological phenomenon in deciduous forests (e.g. in Europe) - please interpret it in case of pine forests at low latitude

Lines 349-350 – I guess that not only the growth depression is more severe in Pinus armandii but also the previous increasing (see the first sections in Figure 2 and also apply the standardized beta). Consequently, this species can be more climate-dependent than P. yunnanensis

Lines 340-354 - please give a possible explanation of the differencies in the trends of series threshold values of the studied species (2002 versus 2006)

07/03/2020

Author Response

Please kindly see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors accepted and fulfilled the critical recommendations of the reviewer: increasing the number of sampling sites (for example, P. yunnanensis with two experimental sites), refining the analyzes (presenting the beta values for comparing), refining the illustrations (Figure 2), explanation extending (new Figure 3) and modification to be more illustrative (Figure 5). The current title is more relevant to present the most important conclusions of the research. The English language of the manuscript requires further minor changes throughout the manuscript.

Some further suggestions to check and modify:

Lines 153-155 - mentioning Southeast China is enough in one place (e.g. in section a), explain abbrevatiations of sampling sites in the figure legend (BSA, BSY, LJY)

Line 287 – Figure 3 - the structure of Figure 3 (analitical framework) is too complicated, indicate the most important steps with simplified explanations, I suggest the linear form; if the term 'precipitation variable' means 'precipitation data', correct it accordingly, so do it for 'temperature variables'; please clearly mark (highlight) the three analytical (analytical) steps (PCA, CCA, PLS)

Lines 300-301 - explain abbrevatiations of sampling sites in the figure legend (BSA, BSY, LJY)

Line 311 – it is sufficient to indicate the abbreviation (BAI) of the dependent variable on the y-axis of the figure (detailed explanation is included in the figure legend)

Line 358 – Figure 6 - it is sufficient to indicate the abbreviation (BAI, PDSI) of the dependent variables on the y-axis of the figure (detailed explanation is included in the figure legend); if dotted line (in section a) refers to the critical (threshold) value of correlation coefficient, please indicate it in the figure legend; in section b-d: PDSI and BAI trend lines are easy to confuse, try to contrast them for a better identification

Line 359 – tree ring width (?) – if so, please refer that plots indicate the tree ring data (eg. mean)

Lines 365-369 – Table 2 - please unify the abbreviations for temperature (TEMP) and precipitation (PREC) along the table and footnote, I suggest using capital letters for a better identification; I suggest using the 3-letter abbreviations of the months (similar to Fig. 6) for easier overwiev; in line 369 check and correct Fig.4 as the reference

Lines 418-419 and 431-432 - if possible, please clarify chapter titles for higher consistency

28/03/2020

Author Response

Please see the attachment

Author Response File: Author Response.pdf