The Effect of Season of the Year on the Frequency and Degree of Damage during Commercial Thinning in Black Alder Stands in Poland
Round 1
Reviewer 1 Report
My few comments are marked in the attached reviewed manuscript; mainly the following:
lines 104-104. If this measurement was done for each plot, it sounds as both numbers are the same thing, since each plot either was harvested in winter or in summer.
lines 140-142 (and also in conclusion 2). Table 2 shows the contrary (maybe the two column´s titles were exchanged). Clarify that or, correct it.
lines 140-150. Must specify that this is true just for extraction, not for felling.
line 154. Clarify that "vicinity", menas less than 1 m)
Comments for author File: 
Comments.pdf
Author Response
I would like to thank you very much for the valuable and substantive comments. The columns in Table 2 concerning depth of damage during felling and timber extraction had indeed been exchanged. All changes have been made in the revised manuscript according to the reviewer’s suggestions.
Reviewer 2 Report
This manuscript reports on the effects of season of harvest (summer vs. winter) on the size, location, frequency, and intensity of damage to black alder during a commercial thinning in Poland. The experimental design and analysis is adequate for this investigation. There are a few areas that could be revised to improve the overall readability of the paper:
Title: please include more details, such as: species' common name, country of study, mention of "commercial thinning", and more about the study ("frequency and degree of damage"). Title does not have to be long, but it could be much more descriptive with a little wordsmithing.
Abstract: species nomenclature (common name, scientific name with authorities) should be included in the abstract and possibly in the keywords.
L. 33: what is the difference between undergrowth and ground cover? Can one term be eliminated?
L. 36: suggesting a different term than "deciduous"...why not just call them "hardwoods"? There are deciduous conifers. Could also use the term "broadleaves" here (authors use it in line 174).
L. 42: need to include scientific name for grey alder as well.
L. 61: need to mention that the study is about a commercial thinning.
L. 70: indicate stand origins: natural, seeded, or planted? what is the topography of the site (slope, aspect)? what is the soil type and texture?
L. 72: what is the actual treatment that is being administered? for this, we need to know the current stand density (include both stems/ha and m2/ha basal area) and the target residual stand density. What type of thinning treatment... thinning from below to release dominants? Additionally, it would be good to know also what is the residual stand spacing, so the reader can judge the difficulty of conducting timber extraction without contacting residual trees.
L. 92-94: it would be helpful to know how the damage occurred during timber extraction, as I cannot picture the operation. Photographs would help. "Manual timber loading" suggests the crew hand lifted the 1 and 2.5 m logs; is movement of the forwarder (with a loading arm?) the cause of the damage attributable to "timber extraction"? this is unclear.
Figure 1: although this is "adequate" for showing the plot layout, I'd like to suggest these revisions to improve the presentation: (1) redraw with square, and not rectangular, plots., and (2) use regular numbering and not Roman numerals to refer to plots.
L. 108: distinguish between root collar and trunk.
L. 110: change second category to 1.1-7 m.
L. 116: I am not familiar with this test, but I did wonder why a factorial ANOVA wasn't used to test the effects of season, distance from skid trail, and their interaction. I don't see how the non-parametric test can be used to test these factors, and the authors do show results that indicate a significant interaction between these two factors.
L. 120: 196 trees per plot indicates a density of 784 trees/ha and a spacing of 3.6 m. This is very close spacing for a thinned stand. Does this actually release the residual trees? How typical is this thinning intensity in application? A higher thinning intensity would result in much less damage and a greater release response.
Table 1: why not report either as percentages of existing trees or as trees/ha? the same can be said for table 3.
L. 134: there is no description of how root damage was defined and field-identified.
L. 148-150: This sentence needs to be qualified because the result is limited to the timber extraction process and not relevant to the felling process.
L. 165-166: I am guessing that the level of damage is largely attributable to the close spacing of residual trees (3.6 m).
General: I wondered whether the authors could include at least 1 figure (could be a photograph of the operations). The authors could also consider including a histogram that shows how damage declined rapidly with distance from skid trails and also how this damage varied with season. Seems like that would add some interest to the paper to include a visual.
Author Response
First, I would like to thank you very much for the valuable and substantive comments. Below I will go on to answer the reviewer’s comments.
Title: please include more details, such as: species' common name, country of study, mention of "commercial thinning", and more about the study ("frequency and degree of damage"). Title does not have to be long, but it could be much more descriptive with a little wordsmithing.
Abstract: species nomenclature (common name, scientific name with authorities) should be included in the abstract and possibly in the keywords.
The suggested detailed information has been added in the title, abstract and keywords.
L. 33: what is the difference between undergrowth and ground cover? Can one term be eliminated?
Ground cover has been removed.
L. 36: suggesting a different term than "deciduous"...why not just call them "hardwoods"? There are deciduous conifers. Could also use the term "broadleaves" here (authors use it in line 174).
The term broadleaves has been used.
L. 42: need to include scientific name for grey alder as well.
It has been completed.
L. 61: need to mention that the study is about a commercial thinning.
It has been completed.
L. 70: indicate stand origins: natural, seeded, or planted? what is the topography of the site (slope, aspect)? what is the soil type and texture?
It has been completed.
L. 72: what is the actual treatment that is being administered? for this, we need to know the current stand density (include both stems/ha and m2/ha basal area) and the target residual stand density. What type of thinning treatment... thinning from below to release dominants? Additionally, it would be good to know also what is the residual stand spacing, so the reader can judge the difficulty of conducting timber extraction without contacting residual trees.
The necessary additions have been made. Unfortunately, we do not know the number of trees before the logging operation; we know the number of residual trees. Thinning from below was the main method used in the studied stands. The age and thinning intensity were typical for stands of this type in Polish forestry practice, and the trees to be removed were marked by a local forester.
L. 92-94: it would be helpful to know how the damage occurred during timber extraction, as I cannot picture the operation. Photographs would help. "Manual timber loading" suggests the crew hand lifted the 1 and 2.5 m logs; is movement of the forwarder (with a loading arm?) the cause of the damage attributable to "timber extraction"? this is unclear.
Photographs illustrating the studied stands and logging operations have been added. The description of the timber extraction method has been revised.
Figure 1: although this is "adequate" for showing the plot layout, I'd like to suggest these revisions to improve the presentation: (1) redraw with square, and not rectangular, plots., and (2) use regular numbering and not Roman numerals to refer to plots.
Figure 1 has been redrawn according to the reviewer’s suggestions.
L. 108: distinguish between root collar and trunk.
Root collar and trunk were treated as one category. It is not possible at this stage to distinguish between them.
L. 110: change second category to 1.1-7 m.
It has been revised.
L. 116: I am not familiar with this test, but I did wonder why a factorial ANOVA wasn't used to test the effects of season, distance from skid trail, and their interaction. I don't see how the non-parametric test can be used to test these factors, and the authors do show results that indicate a significant interaction between these two factors.
The groups of data do not have a normal distribution, and so the use of a non-parametric test was more appropriate.
L. 120: 196 trees per plot indicates a density of 784 trees/ha and a spacing of 3.6 m. This is very close spacing for a thinned stand. Does this actually release the residual trees? How typical is this thinning intensity in application? A higher thinning intensity would result in much less damage and a greater release response.
The stands were dense because they had not previously been harvested. The age at which the studied silvicultural operation was carried out is typical for swamp alder stands in Poland. The thinning intensity was 13.12% of stand volume and is typical for this kind of cutting. In forestry practice in Poland, intensive tree release in silvicultural treatment is avoided, especially in pine, oak, beech, and alder stands. At the next cutting in 5–7 years the number of residual trees will be reduced – the target number of trees per hectare in alder stands is about 300.
Table 1: why not report either as percentages of existing trees or as trees/ha? the same can be said for table 3.
Numbers of damaged trees and their percentages in relation to residual trees have been added in Table 1. Data on the number of damaged trees and percentage distribution of particular sizes of wounds have been added to Table 2, which thus now contains a comprehensive characterization of tree injures.
L. 134: there is no description of how root damage was defined and field-identified.
Only visible wounds on exposed roots below the root collar were classified as root damage. This information has been added to the text.
L. 148-150: This sentence needs to be qualified because the result is limited to the timber extraction process and not relevant to the felling process.
It has been revised.
L. 165-166: I am guessing that the level of damage is largely attributable to the close spacing of residual trees (3.6 m).
Yes, of course. One of the most important factors influencing the level of damage is tree density. In young stands (like those studied) the distances between trees are small. In the case of the studied stands, this was the first timber harvesting. The volume of one-off timber harvested is relatively small and usually does not exceed 15–20% of a growing stock.
Reviewer 3 Report
This paper is well written and easy to follow. My only substantive comment is that the analysis method used may not be the most informative. One question that arises is: why was it deemed that a t-test would suffice in some instances but a Mann-Whitney was needed. An explanation would be helpful. The t-test is problematic because it appears that you had blocking by stand (2) but that was not considered, including a block may provide even greater power. But to go further, a t-test is only an approximate method anyway and it relies on large sample properties. One could have chosen a glm approach with a logistic or probit link. A mixed model approach would allow for inclusion of a plot or block factor. I would recommend refitting this accordingly but I confess that it is quite likely the results and conclusions will remain the same.
Author Response
I would like to thank you for the very valuable and substantive comments concerning statistical analysis. The T-student test is commonly used to test significant differences between two groups of data. In our study it concerned numbers of damaged trees in seasons of the year and type of operation. When we compare three or more groups of independent data other tests need to be used – we decided on the non-parametric Mann–Whitney U test. The study was carried out on two neighboring stands, but they were treated as one block. The single compartments were too small to establish an adequate number of plots. You are correct that more sophisticated statistical tests could be used, but it is very likely that the results and conclusions would remain the same, as you acknowledge in your review.
