Phenology Is Associated with Genetic and Stem Morphotype Variation in European Beech (Fagus sylvatica L.) Stands

Reviewer comment

Our response (with track changes in the manuscript)

- the research concerns stands created from planting and thus managed by foresters. Cultivation cuts, which can greatly influence the quality and form of trees, are a common business practice. Unfortunately, the authors are silent on this fact, although they note other factors potentially affecting the straightness of tree trunks. In my opinion, it is worth to refer to this in the discussion and devote one sentence in the methodology, informing the reader whether the authors have knowledge or not about the intensity and type of maintenance cuts.

OK, we added a sentence in MM on tending operations in these stands, see the 1^st paragraph of MM section.

- There is no information how technically the observations of spring budburst and leaf spread on the dominant stage over the crown were performed. In a mature stand with compact crowns it is difficult to make such observations objectively. Were only binoculars used or were fragments of crowns extracted and scored on the grond. Were the observations made once? If so, in what period of time? 

Ok we added a more detail description of phenology scoring. See the paragraph preceding Fig. 1 on page 5.

- In the case of stands from planting, genetic groups are not the result of natural processes. However, their determination is helpful in explaining morphological variation. Therefore, I don't quite see the point of using COLONY.  Rather, the population structure results from the selection of seedlings and their distribution. In my opinion, this requires a comment in the discussion.

We think COLONY results namely in plantations is interesting for comparison with natural stands, i.e., we get an additional factor in the series for estimating the effects of artificial planting. In addition to STRUCTURE, it returned the sibling stricture, Neff, which are interesting data. Therefore, we suggest keeping it. 

We added a comment on that in the discussion, see end of page 16

- No budburst observations were made in the NOR1 stand. This should not be passed over in the methodology. My doubts are also raised by the interchangeable use of observations of autumn leaf sentence stage in fig 2. However, growth was associated with the timing of growth cessation rather than with budburst.

We included information that budburst was not scored in NOR1 in MM section.

In Fig 2 for NOR1, we used leaf senescence as replacement of budburst, as an estimate of phenology, having in mind that budburst and leaf senescence are positively correlated. 

Reviewer comment

Our response (with track changes in the manuscript)

The introduction begins with a discussion on climate change but this isn’t revisited in the discussion in meaningful way. For example would you use these forests as a source of seed in the future or not?  What are the implications of the findings in the manuscript for climate change? Would you say these stands are adapted or was the transfer from the seed source to these planting sites too far? 

We agree and the following paragraph is inserted in the middle of page 15

“Considering the adaptive significance of phenology for spreading of European beech in Lithuania, we suggest deploying the seed sources by considering the frost hardness gra-dients in Lithuania. Our study showed that late budburst is a favorable feature for Euro-pean beech to be delayed in the seaside lowland of Lithuania. Furthermore, to disrupt the phenology-based groups of relatives in the new plantations of European beech, we rec-ommend collecting the seeds all over the area of a seed collection stand from the trees of variable phenology by avoiding the very early flushing mother trees.”

Secondly, what is the relationship between bud-break and flowering? You haven’t described or included observations about flower production, so I don’t understand how you can make statements about non-random mating in the population you sampled without sampling the progeny. The different genetic groupings are interesting but are likely a legacy of the seed sources used to establish the plantings.  If available, please provide supporting evidence to show that bud-break time is correlated with flowering and flower patterns.  If evidence is not available then focus on the connection between bud-break and morphotype and delete references to non-random mating.

Phenology traits within the annual cycle are strongly autocorrelated. Bud-break timing is positively correlated with flowering time, so that the trees that start active growth relatively later also are flowering relatively later. We included a paragraph with references on that in the discussion section, at the end of the first paragraph on page 14.

“The timing of phenology traits is tightly positively autocorrelated within the annual cycle of trees (Soularue and Kremer 2012, 2014). Therefore, the ranking in timing of budburst can be used as a reliable estimate of ranking in flowering time at the individual tree level (Shim et al. 2012; Vander Mijnsbrugge and Moreels 2020).” 

Soularue, J.P.; Kremer, A. Assortative mating and gene flow generate clinal phenological variation in trees. BMC Evol. Biol. 2012, 12, 79  

Soularue, J.P.; Kremer, A. Evolutionary responses of tree phenology to the combined effects of assortative mating, gene flow and divergent selection. Heredity 2014, 113, 485–494.  

Shim, D.; Ko, J.H.; Kim, W.C.; Wang, Q.J.; Keathley, D.E.; Han, K.H. A molecular framework for seasonal growth-dormancy regulation in perennial plants. Hortic. Res.-Engl. 2014, 1, 1–9. [Google Scholar] [CrossRef] [PubMed]

Vander Mijnsbrugge K, Moreels S. Varying Levels of Genetic Control and Phenotypic Plasticity in Timing of Bud Burst, Flower Opening, Leaf Senescence and Leaf Fall in Two Common Gardens of Prunus padus L. Forests. 2020; 11(10):1070. https://doi.org/10.3390/f11101070

L44. “emerging” (g is missing)

L46. No “a” needed before “large within-population”

L56. Not genetic “cause” but genetic “characterization.”

L58. Delete “the”

All done thanks

L71. Please describe a spike knot. Also, replace “early” with “premature” or “pre-empted” growth.

Here we suggest keeping “early”, because “premature” sounds as an endogenic defect. While “early” growth onset in spring is a normal phenomenon for some trees in forest tree population, given its vast diversity powered by geneflow.  

L73. A sentence is needed before “Gene flow…” in which you describe, and cite, that cold tolerance is usually highly heritable. 

Ok we included the following paragraph at the place suggested by the reviewer in the Introduction:

Owing to the adaptive significance, phenology rhythm and frost hardness traits are highly heritable in northerly forest tree species (Danusevicius et al. 1999, Elzinga et al. 2007, Pyhäjärvi et al. 2020).

Elzinga, J.A.; Atlan, A.; Biere, A.; Gigord, L.; Weis, A.E.; Bernasconi, G. Time after time: Flowering phenology and biotic interactions. Trends Ecol. Evol. 2007, 22, 432–439.

Danusevičius, D., Jonsson, A. and Eriksson, G. 1999. Variation among open-pollinated families of Picea abies (L.) Karst. in response to simulated frost desiccation treatment. Silvae Genetica 45 (3-4):  158-167.

L77-107. The information in these paragraphs is interesting, but these topics get little attention in the discussion. Place less emphasis on the points that are tangential to this project (insects).

We have touched the insect issue as a factor affecting stem morphotype. But OK we agree and shortened the paragraph about that in the introduction.

L118-20. Could you discuss whether the sites of probable seed origin are colder or more continental than the planting sites in Lithuania? That information would be interesting here.

OK thanks good suggestion. We did so, see the end of the Discussion section

The mountainous Bavarian and Carpathian sites are cooler than the planting sites in Lithuania (e.g., the mean annual temperature is 6.05-6.63°C, 5.05-5.61°C and 7.44-7.65°C for the Bavarian, Carpathian, and Lithuanian sites, respectively [78]. The Carpathian site is more continental than the other two sites: mean annual temperature amplitude for the Bavarian, Carpathian and Lithuanian sites is 18.4-18.5°C, 20.9-21.0°C and 19.7-20.7°C [78]. When transferred to a new location, broadleaved trees adapted to cooler sites require relatively less heat to budburst [10]. This may lead to a relatively stronger spring frost damage and lower stem quality.

L136-7. What hypothesis are you testing? Did you expect stands with lower heterozygosity to have more defects?  Why?

We explained that as follows (end of the discussion):

Owing to presumably lower vitality as a result of mating among relatives, the trees of low observed heterozygosity (Ho) may be less tolerant to environmental stresses. However, in our study the absence of relationship between fitness (high stem quality) and observed heterozygosity indicates that homozygotes are largely lost already in early ontogeny (review by [8]).

L144-45. Do you know the Mean Annual Temperature (MAT) or Plant Hardiness Zone (PHZ – or minimum temperature) of the seed origin – even an approximate range? How does it differ from the MAT and PHZ of the planting sites? This will help us understand the severity of the transfer distance of the original stands. Were the original seed sources from sources that were colder (higher elevation) or warmer than the planting sites?  This is important for a discussion about climate change and whether this is a model for what to do in the future or what NOT to do in the future.

OK, we responded to that by including the paragraph on the climatic data and the transfer effects (our response see above).

L146. Misspelling: beech instead of beach

L162. Very nice index!!

L166. Delete first use of “is.”

L252. “of” not “if”

L255. Phenology not phonology.  “consistent” instead of “consequent”

L289. “Traits” not “trats”

L302. “was more strongly associated” instead of “was stronger associated with”

L354. “beech” not “beach”

L355. “roughly” not “ruffly”

L357. “stands” not “stnads”

L441. “Differ” instead of “different”

OK we corrected them all

L213-4. What was the reason for testing heterozygosity? Did you anticipate potential for inbreeding? Include a sentence describing why you focused on Ho.

We inserted the following sentence at the end of the MM section:

Here we intended to test the hypothesis that low Ho values are associated with reduced tree vigour, which in turn leads to low stress tolerance that is reflected by stem defects.

L302-3. I’m confused as to why trends between height to first fork / budbreak score would be different from stem quality/ budbreak score.  Isn’t height to first fork part of the stem quality rating?

We explained what we meant by this in the text as follows (page 10, 1^st paragraph):

Stem quality index was more strongly associated with phenology stages than height to first fork, indicating that the stem quality index is a better estimate when studying the causal phenology – stem quality associations.

L305-6. I don’t see the trend of diameter association with phenology in the supplemental figures. 

We revised this sentence to better reflect the figures as follows (the last sentence of 3.3 subsection on page 10):

On the contrary, stem diameter tended to be positively associated with phenology, especially in VIES stand (Figures S3 to S6). 

L402-3. I don’t follow the logic of this sentence: …”in the case of winter frost tolerance, we would have observed a better stem quality of the trees with early growth onset and cessation.”  Do you mean trees with more conservative phenology would be more hardy to winter frosts?  I think you can delete this sentence.  

Yes, the trees that early start/cease active growth develop higher frost hardness …

We rephrased is as follows:

If adaptive significance of winter frost tolerance exceeds spring frost avoidance, then the trees of namely early growth onset/cessation get an adaptive advantage and contain better stem quality over the late flushing trees in a population [10, 18].

L406. How does the stocking of these stands compare with other commercial European beech stands?  Please include a reference to indicate if it is higher, or lower than typical stands.

By stocking effects in this sentence, we meant the random error effects such as e.g. overstory openings or glades on phenology (more sun) and stem straightness, when trees tend to be curvy when seeking for the opening that is nor directly above a tree.  

We revised it to be more understandable

Of course, we cannot exclude the effects of random environmental stochasticity such as overstory openings, glades or game and pest damage on the phenology and stem morphology estimates.

L425-426. I don’t agree with this statement.  I believe the difference in phenology and genetic groups is a legacy of the original seed source. This whole discussion about the phenology and causes for its mismatch are irrelevant without data on flowering that might reveal connections with bud-break and flowering. Other forces might affect flowering which would drive non-random mating.  The rest of this paragraph should be shortened only to include relevant points.

OK we shorted it.

Nevertheless, we strongly belief as also supported by many authors that in northerly tree species timing of budburst is strongly connected with timing of flowering.

Soularue, J.P.; Kremer, A. Assortative mating and gene flow generate clinal phenological variation in trees. BMC Evol. Biol. 2012, 12, 79  

Soularue, J.P.; Kremer, A. Evolutionary responses of tree phenology to the combined effects of assortative mating, gene flow and divergent selection. Heredity 2014, 113, 485–494.  

Shim, D.; Ko, J.H.; Kim, W.C.; Wang, Q.J.; Keathley, D.E.; Han, K.H. A molecular framework for seasonal growth-dormancy regulation in perennial plants. Hortic. Res.-Engl. 2014, 1, 1–9. [Google Scholar] [CrossRef] [PubMed]

Vander Mijnsbrugge K, Moreels S. Varying Levels of Genetic Control and Phenotypic Plasticity in Timing of Bud Burst, Flower Opening, Leaf Senescence and Leaf Fall in Two Common Gardens of Prunus padus L. Forests. 2020; 11(10):1070. https://doi.org/10.3390/f11101070

L443-444. Replace “marker” with “indicator.” I would recommend saying it “may” be used as an indicator of genetic population structure – other factors such as stocking and insects/diseases may cause forking in other stands. 

We changed ‘marker” to “indicator”, may be indicator of pop structure is too strong?

L446-465. I don’t understand what a “visually evident spatial genetic structures (SGS)” might be, so I don’t follow the logic in this paragraph. Do you mean that naturally regenerated stands produce clusters of related families?  Would this occur from suckering or short-distance seed dispersal?

Ok we revised to be more clear as follows:

In our study, the European beech stands were artificially established and contained no immediately evident spatial clustering of related individuals as indicated by the spatial arrangement of the genetic groups in the studied stands (Figure S7, S10, S11).

L457. “stands” instead of “sands”

OK

L461. “though” instead of “tough”

OK

L467-471. Please reflect on the difference in climate between Carpathian (origin of NOR1 seed source) and the planting site. How is it different – define your expectation of this type of transfer.

Ok we did that, see end of page 16.

L473. I would not refer to stem morphotype as a “fitness” trait because it does not necessarily influence survival.

OK, we changed “fitness” to “stem quality”, see 2^nd paragraph on page 17

Table 1. Please include Mean Annual Temperature and Plant Hardiness Zone (minimum temperature) of the plantings and the origin (if available) in this table. 

OK, we described the climatic zone of Lithuania where the plantations are located and included mean annual temperature of the original regions and the planting sites from interactive maps of climatic data from [78]

Table 4. Categories “Leaf” “Findties” and “Ties” are not described in caption.  

Corrected

Figure 5. This figure can probably get placed with supplemental figures.

OK