Evaluating the Likelihood of Tree Failure in Naples, Florida (United States) Following Hurricane Irma

Round 1

Reviewer 1 Report

This is a very good paper on the rates of tree damage after a hurricane in Florida.

To me, the most interesting findings are that surprisingly few trees were damaged by the hurricane and that larger trees were damaged less. Neither of these points is emphasised in the discussion. The first could just be an issue with wind speed measurements (see my points below) but the second seems to run contrary to expectation and is worthy of some emphasis.

Specific points / questions that arose while reading:

• Line 72 – did these trees evolve in a windier or calmer location?
• Line 105 – What does ‘medians’ mean here?
• Line 158 – Could the trees there were not found have been blown away by the hurricane? Were they generally small trees?
• Line 159 – 74% had no damage – that is a really surprising result given the description of the category 4 hurricane as ‘most trees will be snapped or uprooted’ (line 102). Are there any wind speed data available within the city?
• Line 161 – What does the term ‘whole tree failures’ mean here? Most readers would assume that the tree is uprooted or snapped at the trunk. In your methods section (line 122) it is defined as ‘severely damaged or whole tree failure’ so trees can be in this category without being uprooted or snapped – do you have any data on that? I think you should emphasise that again here for those who may have skipped / forgotten the definition. This actually makes your results more surprising because presumably even fewer than 15% of trees failed.
• Line 183 – this goes against the expectations in the introduction that larger trees will be at higher risk, but you don’t comment on it much in the discussion.
• Line 215 – I’m not sure that is correct. Couldn’t you use a multiple linear regression of damage likelihood against wind speed and add species as a random effect? That could be the subject of another paper, but I think the sample size would be sufficient.
• Figure 3 – why 229 km/h? This paper would be improved by an estimate of the range of wind speeds across the locations studied during the hurricane. There is no need to add the model above, just some estimate of whether it really was a category 4 hurricane by the time it reached the city.
• Line 243 – some explanation of where these wind speed values come from would be good. Are they the maxima for the hurricanes? Are they the maxima at the location at which the damage was measured?

Author Response

Please see the attached point by point response.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript is really interesting, well-written and suitable for publishing in the Forests journal. However, before publishing a few minor changes would be done in the text. In fact, most of my suggestions relates mostly to formal faults.

Abstract: The authors would explain briefly about meaning of wind damage degrees (especially minor and significant).

Keywords: I recommend deleting “tree” in “tree risk assessment” or replacing by “wind risk assessment”

Introduction (or alternatively in Discussion): Authors would write 2-3 sentences about potentially increasing risk of wind damage to trees due to the climate change and its inherent phenomena.

Material and Methods (but also other sections):

The text is often “too personal” mainly in the fourth paragraph of this section, e.g. We used…, We considered…, We computed… those sequences, would be rewritten, at least in some cases (reduce frequency of this kind of expression).     

Results

When referring to Tables and Figures, do not use simple – straight way, but rather indirect expression (a principle of academic writing). For instance, not: “Table 1 lists 15 species…” but rather: “Fifteen species with 30 individuals…something and something … were assessed (Table 1)”. Rewrite in all cases, please.

Figure 2. Both axes showing numbers in % would have its specific caption (e.g. Proportion of damage type). 

Discussion

Figure 3 and related text would be a part of the Results section!

Moreover, in Figure 3: Perhaps “Survivorship rate” would be better term than “Survivorship”?!

I suggest mentioning (even though it was not a part of research) in a few sentences about root system status as important attribute of tree resistance to wind damage specifically to uprooting. The authors can briefly explain that creating good conditions for root growth (by planting trees in spots with sufficiently bulky soil space) would be a part of complex preventive measures. Actually, limited space for tree rhizosphere is often significant problem for tree stability mainly in urban conditions… The authors may consider whether the short text about tree roots show as a part of the Discussion or Conclusions section.        

Author Response

Please see the attached point by point response.

Author Response File: Author Response.docx