Climate Change Effects in a Mediterranean Forest Following 21 Consecutive Years of Experimental Drought

Round 1

Reviewer 1 Report

In this study the authors analyse a very interesting dataset of a long-term experimental field study of water exclusion in Mediterranean Q. ilex forests. The experimental site is well known and has produce several important research papers. However, my feeling is that this manuscript is rather poor, particularly in the introduction and the statistical analysis sections. I admit that I am reluctant to suggest a rejection or a (major) revision of the paper.  The dataset itself is interesting and I believe that after a detailed and thorough analysis it can support the findings or even provide new important ones. Please find some comments below.

The introduction needs to be enhanced. Why  are these species important, which physiological adaptations might lead them to different responses to droughtetc.

L42-43 Shift to presentation of species is rather abrupt her. Please shortly describe why is this species import. See in conjunction with my next comment.

L55-L56. And why is this important> Please give an overview of how important species Q. ilex? Area that covers, marginal areas that it is established to.

L113. Is this a repeated measures ANOVA. Please specify?

L127-L128. Is soil moisture different at each period or at specific periods. Running a repeated measures ANOVA should inform you about that. Similarly for leaf water potential. Also consider the use of a mixed effects model here.

L129-L130 Italics for species names

L140. 3.2

Figure 3 and 4 needs fixing. They are too small to properly read.

For section 3.2 and 3.3I think an ANOVA  with a treatmentxspecies interaction, as in Ogaya et al.  Forest Ecology and Management (2003) could be informative to show differences in the response of the two species. An additional approach would be to perform an ANCOVA with treatment application and species as independent    factors,    annual    precipitation    as covariant, as in Ogaya and Penuelas 2007.

L147. “Negative polynomial relationship” what do you mean. The curve is initially increasing and the decreases. In the second part of Figure 5 (low canopy) it seems to me that the curves are increasing to aa asymptote and are not negative…. Also you are not describing what sort of model you fitted to the data in section 2.3.

L150-155. How can you say if the relationships are stronger? Do you mean the % of variation they account for? If not, are you estimating the rate of increase for each TBAI~Temp relationship and comparing between species? I think just looking at the graphs is oversimplifying.

L166 à 3.3

I also suggest that you could estimate an annual soil water availability index, or a drought index and use this as a predictor for BAI, BAL and TBAI. In this case potentially the relationships could be linear and thus you could use an ANCOVA to test for differences between species and treatment?

The Discussion is well written and points out the main findings of this work. However due to my concerns on the statistical analysis, I feel that it probably needs to be revised after a more thorough analysis.

Author Response

Regarding to reviewer 1:

-In this study the authors analyse a very interesting dataset of a long-term experimental field study of water exclusion in Mediterranean Q. ilex forests. The experimental site is well known and has produce several important research papers. However, my feeling is that this manuscript is rather poor, particularly in the introduction and the statistical analysis sections. I admit that I am reluctant to suggest a rejection or a (major) revision of the paper.  The dataset itself is interesting and I believe that after a detailed and thorough analysis it can support the findings or even provide new important ones. Please find some comments below.

Response: We thank the positive assessment of the referee to the interest of the data and the suggestions to improve the introduction and the statistical analyses.

-The introduction needs to be enhanced. Why are these species important, which physiological adaptations might lead them to different responses to drought etc.

Response: we have now added a description of the importance of the studied species and their main physiological adaptations to cope drought conditions.

-L42-43 Shift to presentation of species is rather abrupt her. Please shortly describe why is this species import. See in conjunction with my next comment.

Response: As explained in the previous comment, we have now added here the description of the importance of the two studied species and their physiological adaptations to drought.

-L55-L56. And why is this important> Please give an overview of how important species Q. ilex? Area that covers, marginal areas that it is established to.

Response: We added a brief description indicating that Q. ilex is the current dominant species of the Mediterranean forest, and its distribution area.

-L113. Is this a repeated measures ANOVA. Please specify?

Response: Not originally, but we have now conducted a repeated measures ANOVA, and have described it in the revised text.

-L127-L128. Is soil moisture different at each period or at specific periods. Running a repeated measures ANOVA should inform you about that. Similarly for leaf water potential. Also consider the use of a mixed effects model here.

Response: Yes, as explained in this paragraph, both soil moisture and leaf water potential showed a typical seasonality: both variables decreased in summer (during the driest period), recovered in autumn, and had their highest values in winter and spring. We have now explained it more clearly in the revised text.

-L129-L130 Italics for species names

Response: Corrected everywhere in the manuscript.

-L140. 3.2

Response: Corrected.

-Figure 3 and 4 needs fixing. They are too small to properly read.

Response: We have now enlarged figures 3, 4, and 5.

-For section 3.2 and 3.3I think an ANOVA  with a treatmentxspecies interaction, as in Ogaya et al.  Forest Ecology and Management (2003) could be informative to show differences in the response of the two species. An additional approach would be to perform an ANCOVA with treatment application and species as independent    factors,    annual    precipitation    as covariant, as in Ogaya and Penuelas 2007.

Response: Following this suggestion, we have now changed the separate ANOVAs for Q. ilex and P. latifolia by other ANOVAs including species as independent factor, we have explained it in the revised Methods section, and highlighted the significant differences between species in the revised Results.

-L147. “Negative polynomial relationship” what do you mean. The curve is initially increasing and the decreases. In the second part of Figure 5 (low canopy) it seems to me that the curves are increasing to aa asymptote and are not negative…. Also you are not describing what sort of model you fitted to the data in section 2.3.

Response: We have now explained it more clearly in the new version of the manuscript in data section (2.4., lines 120-123) and in the Results section (lines 157-164).

-L150-155. How can you say if the relationships are stronger? Do you mean the % of variation they account for? If not, are you estimating the rate of increase for each TBAI~Temp relationship and comparing between species? I think just looking at the graphs is oversimplifying.

Response: Yes, that means the % of variation they account for. Now we have explained it better in the text: “mean annual temperature tended to be more determinant for TBAI in the tall canopy stand for Q. ilex (P<0.05) and all tree and shrub species together  (P<0.05), whereas annual rainfall was more determinant for TBAI in the low canopy stand for Q. ilex (P=0.05), P. latifolia (P<0.01) and all tree and shrub species together (P<0.05).”

-L166 à 3.3

Response: Corrected.

-I also suggest that you could estimate an annual soil water availability index, or a drought index and use this as a predictor for BAI, BAL and TBAI. In this case potentially the relationships could be linear and thus you could use an ANCOVA to test for differences between species and treatment?

Response: We have now performed your suggested analysis with SPEI index (Vicente-Serrano, S.M. et al. The response of vegetation to drought time-scales across global land biomes. PNAS) but we did not obtain better fits.

-The Discussion is well written and points out the main findings of this work. However due to my concerns on the statistical analysis, I feel that it probably needs to be revised after a more thorough analysis.

Response: We have now revised the Discussion section considering the new results obtained from the new statistical analyses you suggested.

Reviewer 2 Report

In the paper „Climate change effects in a Mediterranean forest following 21 consecutive years of experimental drought“ submitted to forests, the authors study the effect of a drought experiment mainly on 2 dominant species in the region. The authors could show that Quercus ilex is not able to cope with the changes in abiotic conditions as prognosed by climate change scenarios as well as the subdominant Phillyrea latifolia. This will have effects on future species compositions.

I liked the overall idea of the study and the dataset is very special, also given for how long the experiment is already up. The manuscript is nicely written, and it is pleasant to read. As I understand and know this is not the only publication from this long-term experiment, so I would have liked to hear more about the differences between this study and the already existing findings and the uniqueness of this paper. If it is just more data that is fine and valuable in itself but then has to be discussed.

Minor comments:

Introduction:

L 66: Please phrase you research questions and hypothesis clearer.

Material and Methods:

L 79-82: here you mention many species apart from the 2 which were the focus of your research. Why did you not consider them too? Or why now mention them?

L 99: specify device.

Results:

L 129: make sure the species are in italics throughout the manuscript.

Figure 1: I think this is a formatting issue, I can’t see the labelling of the y axis on the right side. Please make that graph wider and lower. Then also x axis labels might be easier to read, the font size there is also quite big.

Fig 2: This is only 20 years of data… Which months were combined here in “winter” etc?

Fig 3: what is overall species? Those 2 combined with the other specie so on a plot level or just the others? Please include F statistics in the anova report and I think its better to have the statistical details in the description or in the text. I don’t see the point of the regression.

Fig. 4: this all looks like a formatting error…  If you’d label these 3 graphs in a, b, c it is easier to refer to them. In b the x-axis is labelled incorrectly.

According to these graphs the overall species seem to correlate well with Q. ilex. Is this because ilex is in there or could you elaborate on that?

Fig. 6… not sure why both 6 and 7 are needed, the information you look at is in 7. So I’d delete 6.

Fig. 7: cut off at the right side again.

Fig. 8: Please explain the abbreviations used. Why do some species only have error bars in one direction? Did they not occur in the other plots? What are the colours? Why drought/control in red there?

Discussion:

You again only discuss these 2 species, but neglecting the others. The first paragraph sounds like you are just confirming previous findings.

Appendix A1: What is all species?

Author Response

Regarding to reviewer 2:

-In the paper „Climate change effects in a Mediterranean forest following 21 consecutive years of experimental drought“ submitted to forests, the authors study the effect of a drought experiment mainly on 2 dominant species in the region. The authors could show that Quercus ilex is not able to cope with the changes in abiotic conditions as prognosed by climate change scenarios as well as the subdominant Phillyrea latifolia. This will have effects on future species compositions.

Response: Thanks for this precise summary of our work.

-I liked the overall idea of the study and the dataset is very special, also given for how long the experiment is already up. The manuscript is nicely written, and it is pleasant to read. As I understand and know this is not the only publication from this long-term experiment, so I would have liked to hear more about the differences between this study and the already existing findings and the uniqueness of this paper. If it is just more data that is fine and valuable in itself but then has to be discussed.

Response: thanks for the positive assessment of our work, dataset and manuscript. Yes, during all the duration of this experiment we have published several papers describing the rainfall exclusion effects on forest development. But the present manuscript is very different from the rest: first of all, the present manuscript contains the largest data of all of them, and this is a big difference when it is compared with the first papers. On the other hand, some previous papers described radial stem growth and biomass increment, whereas in this manuscript we describe basal area increment. Also, the topics of the last published papers were biomass allocation and tree mortality instead of tree growth described in the present manuscript. This manuscript highlights by the first time how Q. ilex is clearly replaced by P. latifolia in the low canopy forest stand, especially when submitted to the experimental rainfall exclusion. This manuscript also includes an extensive description of the most important ecosystem effects induced by the experimental rainfall exclusion described in previous papers (lines 269-277). To make it clearer and solve this concern you made us notice, we now describe these novelties more explicitly in the new version of the manuscript (lines 268-270).

Minor comments:

Introduction:

-L 66: Please phrase you research questions and hypothesis clearer.

Response: We have clarified it in the revised version as follows: “Here, our main objective was to determine the effects of climate change-mediated drought on the Mediterranean forest, paying special attention to the responses of the different species and their possible divergent evolution.”

Material and Methods:

-L 79-82: here you mention many species apart from the 2 which were the focus of your research. Why did you not consider them too? Or why now mention them?

Response: Q. ilex and P. latifolia are the dominant species of this forest, representing more than 80% of the total basal area of the studied forest (as you can see in table A1). The other species are scarcer, not present in all plots... and it was difficult to perform statistics with them. Now, we highlighted it in the study site description.

-L 99: specify device.

Response: Specified.

Results:

-L 129: make sure the species are in italics throughout the manuscript.

Response: We checked it, and now the species are written in italics throughout the entire manuscript.

-Figure 1: I think this is a formatting issue, I can’t see the labelling of the y axis on the right side. Please make that graph wider and lower. Then also x axis labels might be easier to read, the font size there is also quite big.

Response: We enlarged this Figure 1 and checked that all axis labels are now easy to read.

-Fig 2: This is only 20 years of data… Which months were combined here in “winter” etc?

Response: January, February and March were combined in “winter”, April, May and June were combined in “spring”, July, August and September were combined in “summer”, and October, November and December were combined in “autumn”. Now, we have explained it in the Methods section (Data analysis).

-Fig 3: what is overall species? Those 2 combined with the other specie so on a plot level or just the others? Please include F statistics in the anova report and I think its better to have the statistical details in the description or in the text. I don’t see the point of the regression.

Response: Overall species means all the present species in the plots analyzed together (now we clarified it better). Q. ilex and P. latifolia are now analyzed in another ANOVA, following referee 1’s  suggestion (lines 123-125). Now, we have also included F statistics in the ANOVA reports and we have described the statistical analyses with more detail.

-Fig. 4: this all looks like a formatting error…  If you’d label these 3 graphs in a, b, c it is easier to refer to them. In b the x-axis is labelled incorrectly.

Response: We changed x-axis in the middle graph. We also enlarged figs. 3, 4, and 5 to to make reading easier.

-According to these graphs the overall species seem to correlate well with Q. ilex. Is this because ilex is in there or could you elaborate on that?

Response: Overall species includes all tree and shrub species present in the plots (Q. ilex and all other species), so it is not strange to see similar trends in graphs representing Q. ilex data and graphs representing the data corresponding to the overall species given its mentioned dominance.

-Fig. 6… not sure why both 6 and 7 are needed, the information you look at is in 7. So I’d delete 6.

Response: Despite the global results of figure 6 are represented in figure 7 (TBAI is BAI less BAL), we still think it is very interesting to show which fraction of BAI is finally accounted into TBAI and which fraction of BAI is lost by stem mortality. Seeing only figure 7, we don’t pay attention to tree mortality, one of the most important factors determining future species composition.

-Fig. 7: cut off at the right side again.

Response: Corrected.

-Fig. 8: Please explain the abbreviations used. Why do some species only have error bars in one direction? Did they not occur in the other plots? What are the colours? Why drought/control in red there?

Response: The abbreviations are now explained in figure caption. All species have error bars in the two directions, but some of them are so small that are not visible over the point. We have now explained the meaning of the different colours in figure caption. Control and Drought are now depicted in black.

Discussion:

-You again only discuss these 2 species, but neglecting the others. The first paragraph sounds like you are just confirming previous findings.

Response: All the manuscript is focused on Q. ilex and P. latifolia because these 2 species represent more than 80% of total basal area (as shown in table A1). Now, we made it clearer in the text (lines 83-84). We have also added the discussion of the new findings in the first paragraph. Thanks for making us notice.

-Appendix A1: What is all species?

Response: “All species” is now depicted as “overall species” to indicate we consider all of them.

Reviewer 3 Report

This paper aims on BasalAreaIncrement and BasalAreaLos of two main tree species at control and experimental sites in Catalonia in Spain. At experimental sites is manipulativelly lowered soil mosture due reducing the rainfall. This study clearly shows that increasing drought, which is trend simulated by climate scenarios for future decades, leads to increasing loss of total baral area (mortality) or dominant Quercus ilex, when compared with shruby tree species Phillyrea latifolia. Generally, presented research is well designed and very time consuming. Overall, the study is generally written and of high scientific importance. However, I still have several major and a few minor comment to this study.

Major comments:

I fully understand, that this research is very time and money consuming. Therefore I understand, that authors wants to produce more than one study from this experiment. This is very good from several reasons. First, it allows to aim each study to one topic, paper is then easier to follow and more readable. Second, your research will get more visibility I this way. To this point I consider this approach correct (during my quick search I found two studies based on results from this experiment Bogdziewicz et al. 2020; doi: 10.1111/nph.16597, Liu et al. 2020; doi: 10.1111/jvs.12902, Ogaya and Peñuelas 2020; doi:10.3390/f11101094, Ogaya et al. 2020; doi: 10.3389/ffgc.2019.00089). However I do not consider as a correct approach using one figure in two studies without a proper citation. In this case it is about lower part of figure 2. Basically the same figure was used in the study Ogaya et al. (2020; doi: 10.3389/ffgc.2019.00089). From my point of view, this figure should be either properly cited or significantly changed (deleting of asterisk is not that case).

My second objection is related with measurement of soil moisture and leaf water potential. As I understand, you made 252 measurements per site (1 per month per 21 years). Presuming all sites are close to each other, this measurements is representative, when describing the difference in soil moisture between sites. However I would appreciate to quantify the consistency of those measurements over the entire period as well as quantification of difference between low and high canopy stands (both control and experimental). This approach could make significant change of figure 2 and it would be very good reaction to my previous comment. Last but not least, I would recommend installation of automatic devices to those sites if you want to continue with this experiment in the future.

After finishing reading the discussion, I still have a feeling that this section is missing the things.

• First is discussion about effect of the same climate difference on investigated parameters of differently sized trees. For inspiration I would recommend paper of Troullier et al. (2019; doi: 10.1007/s00468-018-1767-z), showing different response to climate of trees of different sizes.
• Second, I feel that your study is burdened by the effect of large trees. Especially when interpreting the absolute numbers of basal area loss. I am aware, that you are declaring no difference in basal area increment of both main tree species, however this is only considering all individuals at site (I hope I did not miss something). In the end, loss of one big tree have greater impact in your result than loss of several smaller individuals. For example study of Lutz et al. (2018; doi: 10.1111/geb.12747) clearly shows, that as the size of the tree grows, so does the amount of biomass.
• Third, study of Treml et al. (2019; doi: 10.1016/j.agrformet.2019.02.036) shows different microclimate characteristics of srub vegetation when compared with tall trees. Due shading effect of branches, the temperature root system of shrubs is lower when compared with tall trees. This study was conducted at treeline ecotone and on the first look may not be representative for your case. However, in your case lower temperature could be related with higher moisture. Therefore considering this, it would be again beneficial to compare soil moisture between all type of stands.

Minor comments:

Quality of figures could be improved. Axis labels are small (figure 3-7) or big (figures 1,2,8). Also color palette of figures 6-7 do not very much fit together. I am recommending to use a color picker to select more proper colors (e.g color.adobe.com)

Figures 3,4,5 – Have you tried to use a linear mixed effect model with a species as a fixed variable, when evaluating the trend for all species (Figures 3 and 4)? Similar approach could be done using generalized additive model to get a polynomial regression in figure 5. I believe that this analysis could provide more representative results.

P1 L14-15 –At those lines you are saying “our experiment resulted into 30% rainfall exclusion and 15% decrease in soil moisture”, but you cowered 30% of surface (P2 L95), which resulted into rainfall exclusion and 15% decrease in soil moisture (P3 L127).

Therefore I would recommend to change this statement: “performing 30% rainfall exclusion resulted in 15% decrease in soil moisture” or similar (I am not a native speaker).

P1 L32 – Personally I do not very much like expression global warming. Instead of it I prefer more general expression “Climate change”, which more general and also includes a decrease in annual precipitations, their intensity etc. Since difference in precipitation and soil moisture characteristics releted to basal growth of tree vegetation between two sites are the main aim of this paper, then expression "Climate change" could be slightly better here.

P11 L261 – I would avoid citations in conclusions.

Author Response

Regarding to reviewer 3:

-This paper aims on BasalAreaIncrement and BasalAreaLos of two main tree species at control and experimental sites in Catalonia in Spain. At experimental sites is manipulativelly lowered soil mosture due reducing the rainfall. This study clearly shows that increasing drought, which is trend simulated by climate scenarios for future decades, leads to increasing loss of total baral area (mortality) or dominant Quercus ilex, when compared with shruby tree species Phillyrea latifolia. Generally, presented research is well designed and very time consuming. Overall, the study is generally written and of high scientific importance. However, I still have several major and a few minor comment to this study.

 Response: Thanks for the positive assessment on the experiment and the manuscript and for the suggestions on how to improve the manuscript.

Major comments:

-I fully understand, that this research is very time and money consuming. Therefore I understand, that authors wants to produce more than one study from this experiment. This is very good from several reasons. First, it allows to aim each study to one topic, paper is then easier to follow and more readable. Second, your research will get more visibility I this way. To this point I consider this approach correct (during my quick search I found two studies based on results from this experiment Bogdziewicz et al. 2020; doi: 10.1111/nph.16597, Liu et al. 2020; doi: 10.1111/jvs.12902, Ogaya and Peñuelas 2020; doi:10.3390/f11101094, Ogaya et al. 2020; doi: 10.3389/ffgc.2019.00089). However I do not consider as a correct approach using one figure in two studies without a proper citation. In this case it is about lower part of figure 2. Basically the same figure was used in the study Ogaya et al. (2020; doi: 10.3389/ffgc.2019.00089). From my point of view, this figure should be either properly cited or significantly changed (deleting of asterisk is not that case).

Response: We agree with your considerations. Despite the lower panel of figure 2 did not contain exactly the same data extend than the figure represented in Ogaya et al. 2020 (now there is one year more included), it is true that they are very similar. Therefore, we have now changed it, also following your next suggestion.

-My second objection is related with measurement of soil moisture and leaf water potential. As I understand, you made 252 measurements per site (1 per month per 21 years). Presuming all sites are close to each other, this measurements is representative, when describing the difference in soil moisture between sites. However I would appreciate to quantify the consistency of those measurements over the entire period as well as quantification of difference between low and high canopy stands (both control and experimental). This approach could make significant change of figure 2 and it would be very good reaction to my previous comment. Last but not least, I would recommend installation of automatic devices to those sites if you want to continue with this experiment in the future.

Response: We have now changed figure 2 following your suggestion. Soil moisture is now represented in four graphics (one for each annual season), indicating the soil moisture in the different canopy stands both in control and experimental. We have also clarified in the Methods section that leaf water potential measurements were conducted seasonally instead of monthly (line 101).

After finishing reading the discussion, I still have a feeling that this section is missing the things.

• First is discussion about effect of the same climate difference on investigated parameters of differently sized trees. For inspiration I would recommend paper of Troullier et al. (2019; doi: 10.1007/s00468-018-1767-z), showing different response to climate of trees of different sizes.

Response: We appreciate this suggestion and we incorporated this reference in the Discussion. Otherwise, the size of stems are similar in control and drought plots, and almost in low canopy plots. The two studied species have a similar range of stem diameter sizes. So the drought effect on TBAI and the progressive replacement of Q. ilex by P. latifolia (precisely observed in low canopy plots) could not be explained by the stem size effect.

• Second, I feel that your study is burdened by the effect of large trees. Especially when interpreting the absolute numbers of basal area loss. I am aware, that you are declaring no difference in basal area increment of both main tree species, however this is only considering all individuals at site (I hope I did not miss something). In the end, loss of one big tree have greater impact in your result than loss of several smaller individuals. For example study of Lutz et al. (2018; doi: 10.1111/geb.12747) clearly shows, that as the size of the tree grows, so does the amount of biomass.

Response: It is true that bigger trees have a larger effect on basal area variation than smaller trees. However, as commented in the site description, this forest is not very old (the forest was not disturbed during the last 80 years, but before these 80 years period it was cut to obtain charcoal). Also, it is described that the canopy reaches 10 meters as a maximum height, which means that there are not any very big tree mainly affecting the global trend of one plot. The stem density is very high in this forest, so when there is a large dead basal area of Q. ilex one can be sure that it is not produced by the effect of one big tree.

Neverhteless, these suggestions of the referee are very interesting, so we have now incorporated them in the Discussion (lines 238-242).

• Third, study of Treml et al. (2019; doi: 10.1016/j.agrformet.2019.02.036) shows different microclimate characteristics of srub vegetation when compared with tall trees. Due shading effect of branches, the temperature root system of shrubs is lower when compared with tall trees. This study was conducted at treeline ecotone and on the first look may not be representative for your case. However, in your case lower temperature could be related with higher moisture. Therefore considering this, it would be again beneficial to compare soil moisture between all type of stands.

Response: Despite in our case the difference between tall and low canopy stands seems to be smaller than that described in Treml et al., we agree that it is interesting to show separately the soil moisture of both canopy stands, now figure 2 is arranged following this suggestion.

Minor comments:

-Quality of figures could be improved. Axis labels are small (figure 3-7) or big (figures 1,2,8). Also color palette of figures 6-7 do not very much fit together. I am recommending to use a color picker to select more proper colors (e.g color.adobe.com)

Response: We enlarged figures 3, 4, and 5, and decreased the size of figures 1 and 8. We also improved the quality of all figures in general.  Thanks for making us notice it.

-Figures 3,4,5 – Have you tried to use a linear mixed effect model with a species as a fixed variable, when evaluating the trend for all species (Figures 3 and 4)? Similar approach could be done using generalized additive model to get a polynomial regression in figure 5. I believe that this analysis could provide more representative results.

Response: A mixed effect model with a species as a fixed variable, could improve the significance of the relationships in Q. ilex and P. latifolia, but excludes the effect of the other species when you are evaluating precisely the trend for all species. We have though conducted new statistical analyses in the line suggested by the referee, we have now conducted a repeated measures ANOVA. See responses to referee 1.

-P1 L14-15 –At those lines you are saying “our experiment resulted into 30% rainfall exclusion and 15% decrease in soil moisture”, but you cowered 30% of surface (P2 L95), which resulted into rainfall exclusion and 15% decrease in soil moisture (P3 L127).

Response: That’s true, we covered 30% of plot surface, and that means we excluded 30% of rainfall, but soil moisture only decreased 15%. That is not strange because under rainfall periods, soil moisture could go to soil layers located under the plastic strips due to soil capillarity from other areas receiving rainfall: despite some soil areas never received rainfall directly, their soil water content was never 0.

-Therefore I would recommend to change this statement: “performing 30% rainfall exclusion resulted in 15% decrease in soil moisture” or similar (I am not a native speaker).

Response: We agree with this suggestion, so we have incorporated it in the Abstract (lines 17-18).

-P1 L32 – Personally I do not very much like expression global warming. Instead of it I prefer more general expression “Climate change”, which more general and also includes a decrease in annual precipitations, their intensity etc. Since difference in precipitation and soil moisture characteristics releted to basal growth of tree vegetation between two sites are the main aim of this paper, then expression "Climate change" could be slightly better here.

Response: We changed “global warming” by “climate change”.

-P11 L261 – I would avoid citations in conclusions.

Response: We now deleted all citations in Conclusions.

Round 2

Reviewer 1 Report

I would like to thank the authors for taking the time to respond to most of my comments. I think the ms is in a much better state now. I would include the analysis of BAI vs SPEI int he methods and just report in the results that it did not give any significant results

Author Response

Thank you for your last comments. Now it is depicted in Materials and Methods (line 123) how we tested also the relationship between BAI measurements and SPEI climatic indices, and in Results (lines 172-173) how we did not obtain significative results using SPEI indices.

Reviewer 3 Report

Performed changes improved the quality of the manuscript. I am satisfied with all the changes as well as with the responses to my comments. I appreciate the change of figure 2. It now shows consistent difference between both types of sites and significantly differs from previously used figures. The updated version of the manuscript is now suitable for publication in the iForest journal.

Author Response

Thank you for your comments.