Growth Responses to Climate and Drought in Relict Cork Oak Populations as a Benchmark of the Species Tolerance

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors studied the response of cork oaks to climate under three different moisture gradients. By comparing multiple methods, it was demonstrated that moisture is a key factor limiting oak growth. The relevant results are scientifically significant and the results are reliable. Overall very good. Personally, I have the following comments.

1) The correlation coefficient was used to evaluate the relationship between the tree-ring chronologies and climate factors, while the moving correlation was used instead to calculate the relationship between tree-ring and NAO, what is the reason for doing so? Also, L247, "this association weakened in the 2010s-2020s". This is not obvious from Figure S2, how did you arrive at this result?

2) Are the daily climate data used in the calculation with “climwin” package? If yes, please clear it.

3) Please give r or r-square in Figure S3.

4）L100，from to 10 to 14 mature.

Author Response

Comments and Suggestions for Authors

The authors studied the response of cork oaks to climate under three different moisture gradients. By comparing multiple methods, it was demonstrated that moisture is a key factor limiting oak growth. The relevant results are scientifically significant and the results are reliable. Overall very good.

• We thank you for your positive comments on our study.

 

Personally, I have the following comments.

 

1) The correlation coefficient was used to evaluate the relationship between the tree-ring chronologies and climate factors, while the moving correlation was used instead to calculate the relationship between tree-ring and NAO, what is the reason for doing so? Also, L247, "this association weakened in the 2010s-2020s". This is not obvious from Figure S2, how did you arrive at this result?

• We used correlation analyses (standard or moving correlations) to assess relationships between climate and growth indices and between NAO and growth indices. The relationships between climate and the NAO are well documented in the Iberian Peninsula, so we were more interested on changes in stationarity (through time) of this major atmospheric pattern and tree growth.
• In Fig S2b you can see that the tree-ring-NAO correlation weakened in the 2010s-2020s and became non-significant.

 

2) Are the daily climate data used in the calculation with “climwin” package? If yes, please clear it.

• We used daily climate data in these analyses. This was indicated in the revised ms.

 

3) Please give r or r-square in Figure S3.

• Done, we provided it in the revised legend.

 

 

4）L100，from to 10 to 14 mature.

• Done, we corrected it.

Reviewer 2 Report

Comments and Suggestions for Authors

This study reveals growth of cork oak trees in three different Mediterranean regions and shows their responses to climate.  The findings have profound implications for forest management and conservation. I think the paper is suitable for publication in Forests, but now I still have a few comments and questions and suggest minor revision.

My major concern is about the crossdating. It is hard to distinguish the annual rings of cork oaks and the crossdating is challenging. The authors have made a lot of efforts, including slicing and staining of the samples. However, as shown in Fig. 3, the ring-width sequences of two adjacent samples, Bozoó and Zarautz, are almost completely reversed in interannual variation during the 1973-1983. I suggest checking the dating during this period and before.

Another issue is that the two analyses of climate growth relationships cannot match to some extent. For example, the author mentioned 'Precision shown similar impacts on growth indicators at Zarautz and Bozoó', which differs from the correlation in Fig. 4.

Moreover, why are the trees at the lower temperature limit in the cork oak ecological range (based on Fig.2) negatively correlated with the temperature in most months (including TMx and TMn) at Bozoó and Sestrica?

Minor issues:

1. I don't think this study only focuses on trees’ response to drought, so I suggest modifying the title.

2. The authors have mentioned in the manuscript that this oak responds to summer drought through changes in several traits. As a diffuse-porous species, do cork oak trees have smaller xylem vessels at drought site? This may be one of the important drought-avoiding strategies.

3. I strongly agree that the relationship between trees and climate should not be limited by monthly boundaries, and I support further exploration of the relationship between tree climate response at different time scales. But I think this should be based on daily climate data, rather than a combination of monthly data from multiple climate factors.

4. At Zarautz sampling site, younger trees did not show a significant growth trend. Besides the influence of climate, are there significant differences in the forest composition of these three sampling points? I hope the authors can provide additional explanations on the forest structure of these sampling points.

Also, is it possible for growth decline to occur at Bozoó and Sestrica? If it exists, does it mean that the cork oak of these related populations cannot adapt to the warming environment?

5. The statement 'age at 1.3m' is not appropriate. It is recommended to use ‘estimated tree age’ and indicate in the method that the estimation of tree age is based on the trunk at a height of 1.3m.

6. Fig. S1: The relationship between tree age and growth rate is not similar at the three sampling sites, and it is not recommended to merge them.

7. Fig. S3: Does the Climate variable on the horizontal axis only refer to precipitation? Please also provide additional information on fitting lines.

Comments on the Quality of English Language

Small grammar errors, e.g. line 127-129.

Author Response

Open Review

( ) I would not like to sign my review report

(x) I would like to sign my review report

 

Quality of English Language

( ) I am not qualified to assess the quality of English in this paper

( ) English very difficult to understand/incomprehensible

( ) Extensive editing of English language required

( ) Moderate editing of English language required

(x) Minor editing of English language required

( ) English language fine. No issues detected

 

Yes    Can be improved       Must be improved      Not applicable

Does the introduction provide sufficient background and include all relevant references?

(x)    ( )    ( )    ( )

Are all the cited references relevant to the research?

(x)    ( )    ( )    ( )

Is the research design appropriate?

(x)    ( )    ( )    ( )

Are the methods adequately described?

(x)    ( )    ( )    ( )

Are the results clearly presented?

(x)    ( )    ( )    ( )

Are the conclusions supported by the results?

( )    (x)    ( )    ( )

 

Comments and Suggestions for Authors

This study reveals growth of cork oak trees in three different Mediterranean regions and shows their responses to climate.  The findings have profound implications for forest management and conservation. I think the paper is suitable for publication in Forests, but now I still have a few comments and questions and suggest minor revision.

• We thank you for your positive comments on our study.

 

My major concern is about the crossdating. It is hard to distinguish the annual rings of cork oaks and the crossdating is challenging. The authors have made a lot of efforts, including slicing and staining of the samples. However, as shown in Fig. 3, the ring-width sequences of two adjacent samples, Bozoó and Zarautz, are almost completely reversed in interannual variation during the 1973-1983. I suggest checking the dating during this period and before.

• We perfectly understand your concern because cross-dating cork oak is very difficult. This is why we took special care and made cross-sections so as to better detect and check tree-ring boundaries. We have checked the dating of the two mentioned sites in the period 1973-1983. We respectfully disagree on your comments because the two commented sites are not located far away from each other but they are subjected to very different climate conditions. Zarautz is located near the Atlantic Ocean, whereas Bozoó is located in a more continental area (see Fig. S2). Indeed, Bozoó is closer to Sestrica in the “climatic space” (Fig. 2d) than to Zarautz. Thius would justify that the matching of their ring-width series is not perfect. We have revised the samples and calculated a Cofecha output combining the measurements of Zarautz and Bozoó using 20-year intervals (overlapping every 10 years). The average series correlations for the commented periods 1970-1989 and 1980-1999 are 0.46 and 0.40, respectively. These values indicate coherence between sites’ series despite the commented climatic differences.

 

Another issue is that the two analyses of climate growth relationships cannot match to some extent. For example, the author mentioned 'Precision shown similar impacts on growth indicators at Zarautz and Bozoó', which differs from the correlation in Fig. 4.

• We think the correlation and climwin approaches to assess climate-growth relationships matched overall. Differences may arise because climwin selected the optimally fitted model and this was based on daily climate data (this has been indicated in the revised ms.).

 

Moreover, why are the trees at the lower temperature limit in the cork oak ecological range (based on Fig.2) negatively correlated with the temperature in most months (including TMx and TMn) at Bozoó and Sestrica?

• Negative associations between growth indices and temperatures during the growing season usually reflect the negative impact of elevated evapotranspiration rates and drought (see also the analyses with SPEI). Note that cork oak is restricted to mesic sites and does not tolerate wide thermal ranges, i.e. it does not tolerate continental climate conditions (as shown in Fig. 2d). However, at the “most continental” sites where cork oak is present (such as Sestrica) warmer conditions usually lead to water shortage.

 

Minor issues:

 

1. I don't think this study only focuses on trees’ response to drought, so I suggest modifying the title.

> We modified the title as suggested.

 

2. The authors have mentioned in the manuscript that this oak responds to summer drought through changes in several traits. As a diffuse-porous species, do cork oak trees have smaller xylem vessels at drought site? This may be one of the important drought-avoiding strategies.

> Yes, this was indicated in the original version of the ms (line 50).

 

3. I strongly agree that the relationship between trees and climate should not be limited by monthly boundaries, and I support further exploration of the relationship between tree climate response at different time scales. But I think this should be based on daily climate data, rather than a combination of monthly data from multiple climate factors.

> Yes, we agree and climwin analyses were based on daily climate data (this is indicated in the revised ms).

 

4. At Zarautz sampling site, younger trees did not show a significant growth trend. Besides the influence of climate, are there significant differences in the forest composition of these three sampling points? I hope the authors can provide additional explanations on the forest structure of these sampling points.

> Yes, we indicated the different forest compositions in the last column of Table 1. We are not aware on recent changes in forest use (e.g., thinning) which could explain that pattern.

 

Also, is it possible for growth decline to occur at Bozoó and Sestrica? If it exists, does it mean that the cork oak of these related populations cannot adapt to the warming environment?

> We did not observe any evidence of growth decline and canopy dieback (or higher mortality) in these sites. We think they tolerate well current warmer conditions.

 

5. The statement 'age at 1.3m' is not appropriate. It is recommended to use ‘estimated tree age’ and indicate in the method that the estimation of tree age is based on the trunk at a height of 1.3m.

> We changed it as you suggested; thanks for the useful comment.

 

6. Fig. S1: The relationship between tree age and growth rate is not similar at the three sampling sites, and it is not recommended to merge them.

> Done, we removed it from the figure.

 

7. Fig. S3: Does the Climate variable on the horizontal axis only refer to precipitation? Please also provide additional information on fitting lines.

> Done, we modified the legend and added R² values of the fitted models. The climate variable on horizontal axes refers to precipitation.

 

Comments on the Quality of English Language

Small grammar errors, e.g. line 127-129.

> Done, we corrected them.

Reviewer 3 Report

Comments and Suggestions for Authors

Cork oak is common in dendro studies, in part because trees are often impacted anthropogenically, an obvious result of periodic cork harvesting.  But, this study minimized that effect, and it is otherwise interesting to see how dendro performs on an uncommon species.  The sites chosen are even considered ideal (line 77), encouraging.

This manuscript is quite well done and looks nearly camera ready for publication.  Just a few comments for consideration in review:

Figure 1a: Are the background colors for Z, B, and S meaningful?  The colors aren’t mentioned in the caption.

Figure 1c: In which direction is tree growth here?

Figure 1 caption: Shouldn’t the Aronson et al. ref use the # system like all other refs in this journal?

Figure 2a-c: I find mean annual precipitation (MAP) to be of use when displaying climographs.  Pretty clearly, Zarautz gets more rainfall annually, but having the MAP numbers with the graphs would be nice.

Line 157: “cross-sections were prepared …”  I was startled by the word cross-section, which in forestry usually means cutting a tree down and getting a slab of wood.  In this case, it appears to mean taking a thin-section from a tree core.  I’d switch that word to thin-section, commonly understood in wood anatomical studies as thin sections.

Line 161: rig-width, should be ring-width.

Figure 3a: So, two sites are trending downward, with outermost rings being just under 0.5 mm in width.  Some might say the sky is falling, i.e., these trees are in decline.  Would that be the interpretation of the authors?  If not, I’d say something in the text to avoid that interpretation by readers.  Line 406 hints at this, but that seems a little late in the ms.

Figure 3a: Shouldn’t sample depth be expressed in terms of # of trees sampled, versus # of cores (radii)?  It’s the tree that counts, not how many times it was cored, which might be arbitrary.

Figure 3b: Dendrochronologies are easier to visualize when the 1.0 reference line is added.

4. Discussion: After reading nearly the entire ms, a caveat comes out (lines321-322): dendro of cork oak is hard because rigs are hard to distinguish, crossdate, and measure.  Hmmm.  But, this research is still trustable, right?

References: 58 listed, a good amount of scholarship.

Author Response

Open Review

(x) I would not like to sign my review report

( ) I would like to sign my review report

 

Quality of English Language

( ) I am not qualified to assess the quality of English in this paper

( ) English very difficult to understand/incomprehensible

( ) Extensive editing of English language required

( ) Moderate editing of English language required

( ) Minor editing of English language required

(x) English language fine. No issues detected

 

Yes    Can be improved       Must be improved      Not applicable

Does the introduction provide sufficient background and include all relevant references?

(x)    ( )    ( )    ( )

Are all the cited references relevant to the research?

(x)    ( )    ( )    ( )

Is the research design appropriate?

(x)    ( )    ( )    ( )

Are the methods adequately described?

(x)    ( )    ( )    ( )

Are the results clearly presented?

( )    (x)    ( )    ( )

Are the conclusions supported by the results?

(x)    ( )    ( )    ( )

 

Comments and Suggestions for Authors

Cork oak is common in dendro studies, in part because trees are often impacted anthropogenically, an obvious result of periodic cork harvesting.  But, this study minimized that effect, and it is otherwise interesting to see how dendro performs on an uncommon species.  The sites chosen are even considered ideal (line 77), encouraging.

 

This manuscript is quite well done and looks nearly camera ready for publication.  Just a few comments for consideration in review:

 

• We thank you for your positive comments on our study. We avoided sites where cork debarking was recent so as to circumvent the impacts of cork harvesting on cork-climate relationships.

 

Figure 1a: Are the background colors for Z, B, and S meaningful?  The colors aren’t mentioned in the caption.

• Thank you. We indicated the background colors to differentiate sites. The colors are meaningful because they are the same as in Fig. 2d.

 

Figure 1c: In which direction is tree growth here?

• The bark is located to the right. We indicated it in the legend.

 

Figure 1 caption: Shouldn’t the Aronson et al. ref use the # system like all other refs in this journal?

• Done, we modified it.

 

Figure 2a-c: I find mean annual precipitation (MAP) to be of use when displaying climographs.  Pretty clearly, Zarautz gets more rainfall annually, but having the MAP numbers with the graphs would be nice.

• The precipitation data are shown in Table 1.

 

Line 157: “cross-sections were prepared …”  I was startled by the word cross-section, which in forestry usually means cutting a tree down and getting a slab of wood.  In this case, it appears to mean taking a thin-section from a tree core.  I’d switch that word to thin-section, commonly understood in wood anatomical studies as thin sections.

• Done, we modified it.

 

Line 161: rig-width, should be ring-width.

• Done, we corrected it.

 

Figure 3a: So, two sites are trending downward, with outermost rings being just under 0.5 mm in width.  Some might say the sky is falling, i.e., these trees are in decline.  Would that be the interpretation of the authors?  If not, I’d say something in the text to avoid that interpretation by readers.  Line 406 hints at this, but that seems a little late in the ms.

• The decline in ring width as trees age and enlarge is a common pattern observed in many tree species and sites. This does not mean that sites show growth decline because this is the result of ontogenetic changes. We commented on this when describing Fig. 3.

 

Figure 3a: Shouldn’t sample depth be expressed in terms of # of trees sampled, versus # of cores (radii)?  It’s the tree that counts, not how many times it was cored, which might be arbitrary.

• We agree; trees count rather than cores. These numbers are given in Table 2 and in many tree-ring papers the number of cores/radii is usually shown because it is common to take and measure 2 cores per tree. Nevertheless, we modified it as you suggested.

 

Figure 3b: Dendrochronologies are easier to visualize when the 1.0 reference line is added.

• Done, we added it.

 

4. Discussion: After reading nearly the entire ms, a caveat comes out (lines 321-322): dendro of cork oak is hard because rings are hard to distinguish, crossdate, and measure. Hmmm. But, this research is still trustable, right?

• This is a very good point. There are a few tree-ring studies on cork oak, but it is a very difficult species to cross-date. We have a long experience on dendro studies in similar Mediterranean tree and shrub species and the use of thin cross-sections (as proposed in our study) is a very promising complementary tool to improve cross-dating reliability. So yes, we hope and think our research is trustable.

 

References: 58 listed, a good amount of scholarship.

 

• Thank you.