Spatial Variability and Optimal Number of Rain Gauges for Sampling Throughfall under Single Oak Trees during the Leafless Period

Round 1

Reviewer 1 Report

The presented study is focusing on measurements of TF under five oak trees in leafless period. Measurements were performed with 16 stationary TF collectors in order to estimate TF spatial variability and the required number of collectors needed for exact evaluation of TF. The idea is quite interesting as there is lack of such analysis, however a more detailed presentation of the used methods would be necessary, some other tree characteristics should be taken into account in analysis and discussion should be upgraded taking into account the differences between the trees. Additionally, I would reconsider the use of term “isolated tree”. All of my concerns are described in the comments.

L31: “and their expected when …” awkward, revise.

Keywords: consider adding “throughfall” and “leafless period”

L50: please check the definition of the term “isolated tree”. For example, Carlyle-Moses and Schooling (doi: 10.1002/hyp.10519) refer to isolated trees as trees “unobstructed within a field of view 35° from vertical, centred where the lowest branch met the bole”. Do trees in your case meet this requirement? The definitions of an isolated tree are different; however, I would argue that an isolated tree can grow in the forest. Therefore, I suggest to reconsider the use of this term and maybe refer to as “single trees” if the canopies really do not overlap.

Table 1: The very important tree characteristics for considering TF spatial distribution are the characteristic of the canopy above the observed point. I would suggest to add information on canopy openness or any similar variable above the points of each collector.

L113: “the Tf and Tf collectors” – is it meant the Pg and Tf? Please correct.

Figure 1: Later in the manuscript you refer to the difference between TF close to the tree trunk and at the edge of the canopy (Table 4). It would be useful to mark the collectors, considered in each category, with different colours or different signs.

Section 2.3: The methods, used for data analysis, should be described in more detail. The anisotropy should be better described and its implementation better presented. Which software was used for the analysis?  The abbreviation CV is not described (although it is commonly used). The equation for MAE should be added. The information, provided in L169-L177 should be presented in the scope of the Methods section. Also the kriging is not really described anywhere.

Section 3.1: Additional description of the observed events would be suitable.

L149: The SD value belongs to what, Pg or Tf? Rewrite.

L151: “marked by with high variation” awkward, revise.

L152-153: Mean TF per event? Per tree? In general? Awkward sentence, revise.

L156-157: Awkward, revise.

Figure 2: Add the name of the legend at the bottom of the figures. Which models were fitted to these data? Add description also in methods section.

Figure 3: The sills mentioned in the text are not clearly visible on the figures. Consider using a finer colour scale or adding a contour to the variograms.

L169-177: Move to the methods section.

Table 2: The titles in the second line are too close and it is hard to read the table. Consider using the abbreviations.

Table 3: Add presentation of these results in the methods section (description of RSS etc.). Some of the used methods are really quite well known and often implemented, however a bit better description should still be provided by the authors.

L185: Why did you choose the RMAE < 30% as threshold value? Add explanation.

L187: The analysis was performed only for three selected trees. It is obvious that the results are different for the trees as presented in the section 3.2. However, why these differences emerged and what they actually mean should also be described. As well as explanation on the selection of these three trees for further analysis should be added.

L187-L188: “when quantitatively estimated this” awkward, revise.

L189-191: The values are presented for all three trees together? This should be mentioned.

L195: According to the Figure 4 the differences under tree E are quite small, but according to the results in Table 4 the difference is much smaller for tree A. Add explanation.

L196: What about for trees C and D, were there also points with TF > Pg observed?

L212-224: This description is not necessary as all the results are better visible from the table itself. Rather consider describing some averages of the results or some comparison, not directly stated in the table.

L224: Table 4? Revise.

Figure 6: Mark the values on the x axis. Consider adding a description for the different results between the trees. Why in some cases the differences are expect to decrease faster etc.

Table 5: The columns are too close together. Consider at least inserting larger gap between the results for each tree.

L238: “or so collectors” – awkward. What dose “or so” mean? Revise.

L241: Add, which trees are these.

L273-274: Did you check, if those points were developed during all of the events or only during some of them? This comparison may offer additional explanation why such points occur.

L277-278: Comparison of the results for single tree and trees in a forest is not really meaningful in this case. This is why in the first place the presented analysis is important. Please add an explanation about it.

L288: Awkward, revise.

L328-331: Awkward, revise.

L235: What is meant by non-homogenous distribution of rain drops? What about the influence of barre canopy or leafs presence/absence? This should be an important variable regulating water redistribution, but is not mentioned in the paper at all.

L238: “channelled the intercepted water” – to where? To SF, to the floor? The explanation is very poor and inconsistent.

L239: “more branches and leaves intercepted the TF” – what leafs in leafless period? What “more branches” – do you have any information on canopy characteristics at discussed locations? These conclusions are not supported either by results nor by discussion.

L242: During the results and discussion the results for 15% error are continuously discussed; why did you state a different value here?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

In the manuscript the authors describe the throughfall characteristics of Quercus brantii during the leafless period. The study was conducted with five free-standing trees, which seems adequate.

The authors focus primarily on the spatial characteristics of throughfall using spatial statistics.  Some more detailed analysis regarding the temporal dynamics of throughfall of the dependence of throughfall on precipitation intensity would be interesting. Also, it would be interesting to relate the spatial pattern of throughfall to more detailed information of tree structure, e.g. from fisheye photos or from spatially explicit LAI/PAI measurements. The discussion provides a comparison with similar studies on other species.

Overall a nicely written manuscript.

 

Some specific comments:

lines 27-29: I did not find this statement supported later in the paper (results section)

line 139: What does CVt stand for, should be explained close to here.

lines 169-177: This paragraph would fit more to the methods than the results section.

Line 178: I would shortly explain Pg/Tf/MAE/RMAE/Skewness in the table caption.

Table 5: I would but a vertical line or a larger distance between trees to improve the readability of the table.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Generally, my comments were taken into account and sufficiently addressed. However, some analysis was performed and described in previous papers, but results were also presented in this paper. In this case I believe it is necessary to cite all the relevant papers when describing these results. A few comments were overlooked but I think that the questions are on point and I would like to see those things being addressed in the manuscript as I believe the topic would be more understandable and of more interest for wider community.

Figure 1: Please revise the legend.

L143-153: Please add mentioned reference to the description of the methods and clearly state that more information about the method is available there.

Section 3.1: No additional description on the events was prepared.

Figure 2: It is ok not to include the explanation in this case, but you have to cite the previous paper so that a reader who is not familiar with it knows where to get the information. It is very important.

L183: Add the reference and explanation, why you used this value. It is ok to cite the previous paper, but it should be mentioned! (According to the results of previous analysis, described in ***, only rainfall events with a RMAE < 30% …)

Comment about L187: The analysis was performed only for three selected trees. It is obvious that the results are different for the trees as presented in the section 3.2. However, why these differences emerged and what they actually mean should also be described. As well as explanation on the selection of these three trees for further analysis should be added.
Response: For trees D and C, there are no obvious variogram structure. So, we just present a variogram (Figure 4) for these trees: A, B, and E.
Yes, I understand this and it is obvious from the paper. But still, you are probably able to give some kind of explanation why there were differences between these trees? Please reconsider the comment once again.

Comment about L273-274: Did you check, if those points were developed during all of the events or only during some of them? This comparison may offer additional explanation why such points occur. Response: No, we didn’t measure fog, and this sentence deleted.
In the revised version the sentence is nor deleted. In my opinion you don't need to, just state that this may be part of future research at the study plot.

Comment about L277-278: Comparison of the results for single tree and trees in a forest is not really meaningful in this case. This is why in the first place the presented analysis is important. Please add an explanation about it. Response: Thanks for your useful comment. We added some explanation to clarify this section.
I don't see how this was addressed …

Comment about L235: What is meant by non-homogenous distribution of rain drops? What about the influence of barre canopy or leafs presence/absence? This should be an important variable regulating water redistribution, but is not mentioned in the paper at all. Response: In this manuscript, we just focused on Tf during leafless season.
Ok, but still please clarify what is meant by non-homogenous distribution of raindrops?

Author Response

Please see the attachment.

Author Response File: Author Response.docx