The Composition and Height of Saplings Capturing Silvicultural Gaps at Two Long-Term Experiments in Managed Northern Hardwood Forests

Round 1

Reviewer 1 Report

It was a very complicated comparison of two case studies (DCGS vs YBLP) with different harvesting times, methods and opening sizes.

I would suggest following two points for improvement on discussion:

In the YBLP case study, the authors should explain in the discussion why there was not enough yellow birch regeneration or gap capturing (that was the original purpose of the study). The large yellow birch tree left in the centre of the opening should have provide enough seeds for regeneration and the harvesting operation should have provided enough site scarification to create suitable seedbed for yellow birch. Also for black cherry is different from yellow birch because black cherry seeds in most cases were dispersed by birds and how can it was more abundant in YBLP but not DCGS. It should be discussed. Page 8 line 312 " while sugar maple were only 47.2% (Figure 1b). That Figure 1b should be Figure 1a. This and the mistake labelling of (a) and (b) in Figure 1 caption has to be corrected because it casuses cunfusion to readers.

Finally I think that there is a mistake in Figure 1 caption: the labelling of (a) and (b) should be reversed. (a) Divide Canopy Gap Study (DCGS) and (b) Yellow Birch Legacy Project (YBLP).      

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Major comments:

Line 156 says that the experiment DCGS used a randomized complete block design. I do not find information on experimental design for YBLP experiment. The randomized block design refers to a nested experimental design. This needs to be treated by the mixed-effect modelling approach where the “block” should be considered as a random effect.

Lines 259-261: A simple one-way ANOVA without incorporating the nested design cannot be used. This can be a subject of pseudoreplication (Hurlbert 1984).

Line 265-268: here the authors used linear mixed-effect models and the “gaps” was considered as a random effect, but the authors did not consider the random effect of “block”?

Lines 285-286: Please use the mixed-effect modelling approach.

Line 288: Please do not use t-test or ANOVA when the data comes from a nested design.

Please use letters for statistical differences in Fig 1a and Fig 1b. These letters can be generated from the Tukey HSD multiple comparison test.

Please check your table size of Table 3. I cannot see the whole table.

I like Fig 4. Great idea!

I understood that the “gap-capturing saplings” are the new saplings excluding the advanced regeneration. I think for the abundance of “gap-capturing saplings”, the authors need to include the abundances of advanced regeneration as a predictor variable along the with gap area, and species. It will be interesting to include the interaction effect of advanced regeneration abundances and gap area on the abundances and height of gap-capturing saplings. I think the current analysis does not provide insightful information on the data. I would encourage the authors considering different candidate models (mixed-effect) while incorporating the additive and interaction effects of the gap area, species type, abundance of the advanced regeneration, and distance to edge for the abundances and height of “gap-capturing saplings”.

Minor comments:

The title is too long.

I would encourage authors of avoiding long sentences (e.g., the first three sentences of introduction).

Line 70, please cite some references that you meant.

Line 144, what is WI? Please write the full name.

Reference

_{Hurlbert SH. 1984. Pseudoreplication and the Design of Ecological Field Experiments Ecological Monographs. 54: 187-211.}

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The article reports the results of two long-term experimental tests that aim to analyze the reaction to group selection system of two American northern hardwood mixed forests.

This information is of great interest because it offers scientific support to a silvicultural approach, increasingly widespread throughout the world, which aims to increase the complexity of forest systems by improving their resistance, resilience and ability to adapt to the dramatic challenges posed by global change.

The results and their discussion confirm, in general, what is already known about the dynamics of multi-aged stands subject to disturbs induced by silvicultural treatments, and the relationships of competition among species with different tolerance to shading. However, some "certainties" are questioned by the outcome, in particular the role of gap size in the tree regeneration processes. Correctly, the authors call into question other factors such as ungulate grazing and past management, confirming the difficulty of modeling the dynamics of multi-aged forests, given the multiplicity of environmental and anthropic factors to be controlled and the general low predictability of complex systems. This low predictability is accentuated by the large number of species that characterize the North American forests compared to the European ones.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I do not have any further comments.