Maximum Rooting Depth of Pinus thunbergii Parl. Estimated with Depth at the Center Point of Rotation in a Tree-Pulling Experiment in a Coastal Forest in Japan

Round 1

Reviewer 1 Report

Title: In my opinion, the title of the research is a little inadequate. Which is the "main variable" and which is the "predictor variable"? and the type of forests is not mentioned.

Abstract: The purpose of research is repeated sentences from the introduction. The research method is not mentioned.

 

 

Introduction: The importance of research and the necessity of doing research are not well described.

Is the uprooting of trees a serious problem in the study area? And the prediction of “Depth at the center point of rotation” from “maximum rooting depth” will help forest management to eliminate the uprooting of trees by wind/snow?

Line 30: what is your mean by “critical turning moment”?

Line 32: stem volume is not equal with H × DBH².

Line 42: “In such cases, aboveground traits are not always predictors of tree resistance;” In my opinion, it needs more explanation and reference.

The previous sentence refers to the effect of management (lighting) on tree critical turning moment. In fact, according to which research the aboveground traits of tree is not a perfect predictor of tree resistance t uprooting?

Line 160: scientific name of tree should be italic. 

Study area and Methods:

Please well describe the study forest, such as tree height, canopy closure, basal area, species composition, crown width, crown length, wind condition and speed, heavy snowfall statistics, And any other environmental parameter that helps to understand the importance of this research and can affect the research variables.

How is the management history of these forests?   

What is meant by plot in this research? How were they chosen? How was the measurement with three trees extended to the whole plot with an area of about 2.16 hectares with a number of about 300-500 trees?

How the “Depth at the Center Point” was measured?

Why were more accurate methods not used to measure RSP? 

Why the physical and mechanical properties of the soil were not measured in three plots? Such as: bulk density, penetration resistance, shear strength, moisture content, etc.

 

Line 160: scientific name of tree should be italic. 

Is the average number of three samples sufficient for statistical comparison between plots?

Discussion and conclusion:

How can the results of this research contribute to a more appropriate management of these forests?

In the introduction, the effect of thinning is mentioned, do the results of this research have any suggestions for the reconciliation of these stands for possible wind damage?

Author Response

Responses to comments by Reviewer 1's report

 

*Italic letters show your comments.

*Normal letters (red) show our responses.

 

Title: In my opinion, the title of the research is a little inadequate. Which is the "main variable" and which is the "predictor variable"? and the type of forests is not mentioned.

(Response)

Thank you very much for your suggestions, We modified the title as “Maximum rooting depth of Pinus thunbergii estimated with depth at the center point of rotation in a tree-pulling experiment in a coastal forest in Japan”

 

Abstract: The purpose of research is repeated sentences from the introduction. The research method is not mentioned.

(Response)

We revised the sentence of purpose and added the methods to Abstract.

 

Abstract: Lines 15-19 of the revised MS

Our objective in this study was to clarify whether the Dcp in tree-pulling experiments can be estimated as the maximum root depth of Pinus thunbergii in sandy soils. We also estimated which position of displacement of the center of rotation (Cp) can be applied as the Dcp. We conducted tree-pulling experiments, and compared the Dcp obtained from images with the measured maximum root depth.

 

Introduction: The importance of research and the necessity of doing research are not well described. Is the uprooting of trees a serious problem in the study area? And the prediction of “Depth at the center point of rotation” from “maximum rooting depth” will help forest management to eliminate the uprooting of trees by wind/snow?

(Response)

According to your and the reviewer 2’s suggestions, we modified the introduction to make clear the importance and necessity of our research. In 2011, the Great East Japan Earthquake triggered a tsunami. Tsunami caused serious damages to the coastal P. thunbergii forests of Japan, such as uprooting and stem breakage. The maximum depth of vertical tap or sinker roots can reveal the resistance of tree uprooting. In Japan, tree pulling out tests without uprooting have been performed, but did not determine root system structure, because of legal limitation of tree uprooting. Estimates of the maximum rooting depth of P. thunbergii in this study can contribute to management of coastal forests against to tsunami.

 

Introduction: Lines 30-40 of the revised MS

Coastal forests play an important role in preventing damage from strong salty winds caused by the sea and tsunamis [1,2]. Most coastal areas in Japan have been planted with Pinus thunbergii, which has been resistant to salt and wind for the past several hundred years [3,4]. In 2011, the Great East Japan Earthquake triggered a tsunami that severely damaged the coastal P. thunbergii forests, which caused uprooting and stem breakage in P. thunbergii in several coastal forests [5]. Although P. thunbergii originally had deep tap roots [6], the uprooted trees in the damaged coastal forests had plate root systems because of the shallower groundwater table [5]. In previous studies, we found that the contrasting systems of the plate or tap roots of P. thunbergii under different groundwater depths in a coastal forest have different levels of resistance against tsunami winds using tree-pulling experiments and the subsequent harvesting and measurement of root systems [7,8].

 

Line 30: what is your mean by “critical turning moment”?

(Response)

We used the definition of “critical turning moment by Nicol et al. (2006); ”The anchorage of trees can be expressed as the critical (maximum) resistive turning moment at the base of the stem during overturning. Turning moment is defined simply as force × length of a lever arm.

We added the explanation and the formula for critical tuning moment.

 

Introduction: Lines 41-45 of the revised MS

The resistance of trees to uprooting caused by strong winds or tsunamis can be estimated as the critical turning moment [7,9–15], which is defined as the force × length of the lever arm [16]. To evaluate critical turning moments in Japan, tree-pulling experiments have often been conducted without uprooting because of legal regulations regarding forest preservation [7,17–19].

 

Line 32: stem volume is not equal with H × DBH².

(Response)

As you pointed out, stem volume is not equal with H x DBH². H x DBH² is a ”stem volume related parameter”. We revised it.

 

Introduction: Lines 45-48 of the revised MS

The above-ground traits of trees that are directly affected by wind and tsunamis, namely stem diameter at breast height (DBH) and stem volume related parameter (height (H) × DBH²), have a strong relationship with the critical turning moment and are thus suitable predictors of tree resistance [7,13,14,20].

 

Line 42: “In such cases, aboveground traits are not always predictors of tree resistance;” In my opinion, it needs more explanation and reference.

The previous sentence refers to the effect of management (lighting) on tree critical turning moment. In fact, according to which research the aboveground traits of tree is not a perfect predictor of tree resistance to uprooting?

(Response)

Thank you for your suggestion. There was evidence that differences in root growth at the same aboveground biomass can contribute tree resistance to uprooting. We added the following sentences with references.

 

Introduction: Lines 51-59 of the revised MS

In a case study on the effect of below-ground traits, in the 17 years after thinning management, Cryptomeria japonica trees exhibited significantly higher critical turning moments than unthinned trees when trees with the same stem volumes were compared [17]. The reason for the higher critical turning moment in the thinned trees was the increased horizontal radius of the root-soil plate (RSP), which promoted horizontal root growth [17]. Cucchi et al. [13] also reported that the critical turning moments of trees at the border of a stand were higher than those of trees from inside the stand, probably because of their larger RSP volume. In such cases, the above-ground traits are not always predictors of tree resistance.

 

Line 160: scientific name of tree should be italic.

(Response)

This is the format of subheading in the journal of “forests”.

 

Study area and Methods:

Please well describe the study forest, such as tree height, canopy closure, basal area, species composition, crown width, crown length, wind condition and speed, heavy snowfall statistics, And any other environmental parameter that helps to understand the importance of this research and can affect the research variables.

(Response)

Environmental conditions such as wind speed and predicted tsunami wave heights were summarized in the MS as information on the study site. Tree height, tree species composition, and crown width are summarized in Table S1. There is no snowfall in this area. The description of the MS is as follows.

 

Materials and Methods: Lines 138-149 of the revised MS

The study site was located in a coastal P. thunbergii forest stretching 8 km west of the Atsumi Peninsula in Tahara City, Japan [7,8,41,42]. In the event of a major earthquake in the region, tsunami wave heights of up to 4 m were predicted [43]. Most trees in the forest were replanted in the 1960s after being heavily damaged by Typhoon Vera in 1959 [44]; management has been performed only for the treatment of pine wilt dis-ease to remove dead trees [45]. The soil was sandy and classified as Regosol [46]. From 1981 to 2010, the mean annual temperature was 16 °C and the mean annual precipitation was 1603 mm [47]. At the nearest weather station, Irago (34°37′ N, 137°05′ E), the 30-year (1981–2010) annual mean wind speed was 3.8 m s-1 and the daily maximum wind speed was 26.2 m s-1 [38]. The groundwater level was almost constant from the shoreline inland [41]. However, the depth of the groundwater level from the ground surface deepened inland as elevation increased (Table S1).

 

How is the management history of these forests?  

(Response)

Thank you for your question. The most of this area has been planted Pinus thunbergii as a coastal forest in 1960’s after the heavy typhoon damage in 1959. The management has been done only for the treatments against Pine wilt disease to remove the dead trees.

 

Materials and Methods: Lines 140-143 of the revised MS

Most trees in the forest were replanted in the 1960s after being heavily damaged by Typhoon Vera in 1959 [44]; management has been performed only for the treatment of pine wilt disease to remove dead trees [45].

 

What is meant by plot in this research? How were they chosen? How was the measurement with three trees extended to the whole plot with an area of about 2.16 hectares with a number of about 300-500 trees?

(Response)

Thank you for your suggestion. The plots in this study are the areas where we obtained information on tree-pulling experiments and trees. The continuous P. thunbergii forest in this coastal area was divided into three sections based on the depth of groundwater table and the growth of P. thunbergii trees. Thus the plots are representative of these areas. Within the plots, there are areas where trees were dug up the root system. We also summarized the conditions of plots to Table S1. We added the description as follows.

 

Materials and Methods: Lines 158-169 of the revised MS

Within the continuous P. thunbergii forest, the stands had different tree growths, which was due to the different depths of the groundwater table (Tables 1 and S1) [7,8,41,42]. Therefore, we divided the stands into three sections, and surveyed representative sites as plots (Figure S1 [7,8,41,42]). Plot A was located 188–199 m from the shoreline [41,42]; plot B was also located close to the shoreline (150–270 m). Plot C was further from the shoreline (620–740 m [7]; Figure S1). The mean age of the P. thunbergii in the three plots was 45 years. The P. thunbergii stand density was lowest in plot A (100–200 individuals ha^–1), and mid-range in plot B (200–400 individuals ha^–1), and highest in plot C (400–1000 individuals ha^–1). Table S1 summarizes the other characteristics, such as soil water content and understory species among the plot. We established the locations where we dug the test trees within the plots. The areas where the P. thunbergii trees that we dug out were 400 × 15 m for plot A, 120 × 180 m for plot B, and 170 × 120 m for plot C.

 

How the “Depth at the Center Point” was measured?

(Response)

We measured Dcp (the depth of Cp from the ground surface) from images using video camera (Figure 1) and the pixels in images were converted to unit of cm.

 

Materials and Methods: Lines 207-224of the revised MS

We estimated the Dcp and Cp positions in the tree-pulling experiments using the method proposed by Morioka [36] (Figure 1a) and as described in detail by Todo et al. [7]. Briefly, we captured images of the stem during the tree-pulling experiment using a video camera (HDR-CX180, Sony Co., Tokyo, Japan) set perpendicular to the pull direction (Figure S2). We marked six points on the stem surface at vertical intervals of 20 cm from the ground level to a height of 100.0 cm to monitor positional changes [7]. We installed a measuring pole adjacent to the test tree for scaling (Figure 1). We fixed the position of the video camera when capturing the images of each test tree, and we did not use the zoom function.

We tracked the Cp position in 1 s intervals. We also extracted the x and y coordinates of the marks on the tree in 1 s intervals. We created a straight line from all images using the two marks that were clearest and easiest to capture, and we then calculated Cp (Figure 1b). We measured the horizontal displacement (Figure 1c) parallel to the ground surface of each tree as the distance from the start of the tree-pulling experiment, which we defined as 0. We measured the vertical displacement (Figure 1c) at the Cp for each tree as the distance from the ground surface. We thus determined the position of the Cp at the time of the critical turning moment exertion, and we identified the depth of Cp at that time from the ground surface as the Dcp. We measured the length per pixel on a scale.

 

Why were more accurate methods not used to measure RSP?

(Response)

Thank you for pointing this out. Because of the legal forest protection in Japan in this coastal area, it is not easy to uproot and cut down for several trees. Therefore, the measurement of RSP was done by the method described.

 

Materials and Methods: Lines 195-199 of the revised MS

Owing to logging restrictions at the study site, we could not measure the RSP by complete uprooting. We defined the RSP radius as the average distance from the arcs at the edges of the formed RSP (i.e., cracks) to the center of the stem [7]. We measured ten points of the RSP radius for individual trees [7]. We calculated the ratio of the RSP radius to Dcp (RSP radius/Dcp) as an indicator of the RSP shape.

 

Why the physical and mechanical properties of the soil were not measured in three plots? Such as: bulk density, penetration resistance, shear strength, moisture content, etc.

(Response)

Thank you for your suggestion. Basically the soils in this coastal forests are sandy but different groundwater table depth. Descriptions of soil moisture and other factors were added to Table S1.

 

Line 160: scientific name of tree should be italic.

(Response)

This is the format of subheading in the journal of “forests”.

 

Is the average number of three samples sufficient for statistical comparison between plots?

(Response)

Due to limited number of harvesting tree roots by legal limitation, we modified as no statistical comparisons among plots in the revised manuscript.

 

Discussion and conclusion:

How can the results of this research contribute to a more appropriate management of these forests? In the introduction, the effect of thinning is mentioned, do the results of this research have any suggestions for the reconciliation of these stands for possible wind damage?

(Response)

When we estimate the maximum root depth, and thus tree resistance to tsunami, we will predict the susceptible areas. Then, we can propose the management such as drainage and embarkment which are necessary to ensure the availability of sufficient space for root growth in depth direction to achieve a coastal forest which have higher tree stability. We added the description to the discussion.

 

Discussion: Lines 428 - 434 in the revised manuscript

Knowing the maximum root depth of P. thunbergii in coastal forests can provide insights into the resistance of P. thunbergii to tsunamis. Whether coastal forest areas will be prone to collapse or resistant to future tsunamis can be predicated. Fraser [54] suggested that drainage induced deeper rooting in P. sitchensis, which increased the resistance to uprooting by 25%. Management practices such as drainage and embankments, which are necessary to ensure the availability of sufficient space for root growth in the depth direction [49], can be incorporated to create a coastal forest with higher tree stability.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The topic of the manuscript is relevant and with high scientific interest, as authors have focused on the age-long problem of facilitating root evaluation with minimized effort and investments. The study offers an approach to assess the rooting distribution without time consuming excavations or tree uprooting by winching. In general, trees just need to be pulled to reach the maximum resistive turning moment. In this regard, the necessity of applying such approach stayed unclear, as tree damage is not prevented anyway. I would suggest authors to turn this insight into the manner of non-destructive testing.

The most important problem is unconvincing data statistical analysis, which is based on insufficient sample size and repetition – 10 trees in one soil type in same forest stand, as well as soil conditions (mechanical properties, moisture) are not considered. Sampling in three different plots of same forest was not justified as a variation with growing conditions. This study needs larger sample size, which includes different soil and site conditions, to exclude uncertainties for the statistical analysis and thus comparability. The manuscript has poor wording, which hinders the readability, thus serious work is needed for re-writing. This would help to better understand the intentions of authors, as introduction lacks a clearly stated justification of study aim and objectives.

Considering the insufficiency in sample size and necessity for improvement in writing, the manuscript is not acceptable for publication in such a state. Nevertheless, please see minor comments bellow, which might give some practical hints for improving this manuscript for eventual resubmission.

 

1.        Lines 29-30: Please introduce yourself with newest literature in this field.

2.        Lines 30-34: With above-ground traits I would expect you to talk about tree parts, such as trunk, canopy etc. Is the formation of DBH and stem volume directly affected thus dependent on the wind or the resistance against wind loading of a tree is dependent on DBH and stem volume? Convoluted sentence – verbiage. No need to emphasize winds and tsunamis again. The cause of effect is clear from the first sentence. Please make clear statement. This sentence can be split in two. Please reference each fact of the sentence separately, instead of placing all references at the end. This hinders reading. Please do this for all sentences in the text from now on.

3.        Lines 34-38: Maybe with the root system structure you meant all those characteristics mentioned further? Convoluted sentence that is difficult to perceive. Please improve the wording.

4.        Lines 39-41: Please clarify the connection of this sentence with the previous one. Leap into the stream of thought.

5.        Lines 42-44: What exactly indicates that aboveground parts are not always predictors of tree resistance. Such explanation (connection) is missing.

6.        Lines 46-47: Citation needed.

7.        Lines 49-53: Please consider to use another preposition instead of “in” for starting the sentences.

8.        Lines 52-57: I did not understand how this derives from the previous part of the paragraph. Please provide the connection. Citations are needed.

9.        Lines 55-56: “Although destructive measurements of maximum root depth are important...” on what exactly this statement is based?

10.    Lines 55-58: Citation needed.

11.    Line 60: Pleas introduce readers what exactly is the centre point of rotation (Dcp) and how it can be determined. Also, what does Dcp indicate?

12.    Lines 61-62: Citation needed.

13.    Lines 66-68: This thought must be placed somewhere at the beginning for introducing reader with the Dcp.

14.    Lines 69-72: Convoluted sentence that is difficult to perceive. Please improve the wording.

15.    Lines 73-77: All this can be written shorter, avoiding verbosity.

16.    Lines 78-80: What exactly this is important. Please justify the need for such comparison and explain the contribution of that in the field of tree stability research or any practical application.

17.    Line 81: Please use previously defined abbreviation if Dcp is meant by “depth at the center point of rotation” here.

18.    Lines 81-82: The explanation of the concepts of both terms of center point of rotation (Dcp) and the center of rotation (Cp) is crucially necessary somewhere at the begging. The difference between them remains unclear. Otherwise this text is highly confusing considering poor wording.

19.    Lines 87: At which particular edge? Upper, lower, on the radii?

20.    Lines 88-90: What exact improvements tree-pulling experiments will gain by monitoring the displacement of Cp? This has to be clarified.

21.    Lines 100-104 and Table 1: This belongs to Materials and Methods.

22.    Lines 118-136: This belongs to Introduction.

23.    Figure 1: This should be integrated in the description of methods.

24.    Figure 2: This is redundant and tells nothing important. This information can be placed in the text.

25.    Line 150: Are there other plots as well? What are these plots – long-term monitoring e.g.?

26.    Lines 161-164; On what tree selection was based?

27.    Lines 163-164: What exactly limited the sample size?

28.    Lines 164-165: Every abbreviation must be explained first.

29.    Lines 166-169: Detailed description of different parameters among A, B, and C plots is required if this notable differences in, e.g. growing conditions, tree dimensions are present. Otherwise, this is verbosity, which inflates text hindering to follow it.

30.    Lines 170-178: Raw measurement of pulling load is not a turning (bending) moment. A turning moment depends also on the height of anchoring point and the angle of pulling line (rope) and is expressed in kNm. In your test, these pulling heights differed among plots (1 to 1.5 m), which implies biased results if raw measurements of pulling load are attributed as a critical turning moment. This is not correct as lower anchoring height will provide higher loading, but this would not be an issue if turning moment is calculated considering the anchoring height. Please provide a formula how the critical turning moment was calculated using measured load if it has been done at all. Otherwise, your results require recalculation.

31.    Figure S1: Caption needed.

32.    Figure 3: This is redundant and tells nothing important. This information can be placed in the text.

33.    Lines 196-203: The description how did you scaled the measured displacement of stem is needed more detailed.

34.    Lines 213-215: Please provide the system (units, scale) of coordinates used to calculate misalignment of Cp (msaCp).

35.    Equation 1: Please provide reference or explain steps by which this was derived.

36.    Line 220: “The stem was assumed to be a rigid body.” – For what reason this was done? What does that mean?

37.    Lines 239-241: The sample size of 10 trees is too small for reasonable evaluation of normal distribution.

38.    Lines 241-243: Why there is a comparison between plots? Are they separated stands with different growing conditions, management history, proveniences? Or those plots are just an artificial man-made division where sampling was done? If so, such comparison is does not make this study relevant.

39.    Lines 243-244; Lines 278-295; Table 2: Such small sample size (10 trees) might cause biased correlations. Obtained p-values must be multiplied with number of correlation pairs.

40.    Figure 5 (a): What does the indication of time of reaching the critical turning moment tell?

41.    Lines 307-309: Please provide concentrated results in one sentence to remind the reader how exactly.

42.    Lines 310-311: And how do you think why? Please justify why your results on Dcp are better compared with other studies.

43.    Lines 315-319: Such generalizing statement derived from sample size of 10 trees is contradictory.

44.    Lines 328-331: There is insufficient empirical basis for such statement, as compression of soil bellow the root plate has been reported particularly on saturated soils. To claim that, your sample size must be enlarged including different soil types, as well as data on soil density and moisture content.

45.    Lines 409-411: Exactly, this in combination with larger sample size is what would make this study relevant.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Author Response

Responses to comments by Reviewer 2's report

 

*Italic letters show your comments.

*Normal letters (red) show our responses.

 

The topic of the manuscript is relevant and with high scientific interest, as authors have focused on the age-long problem of facilitating root evaluation with minimized effort and investments. The study offers an approach to assess the rooting distribution without time consuming excavations or tree uprooting by winching. In general, trees just need to be pulled to reach the maximum resistive turning moment. In this regard, the necessity of applying such approach stayed unclear, as tree damage is not prevented anyway. I would suggest authors to turn this insight into the manner of non-destructive testing.

(Response)

First of all, thank you very much for your careful several comments and fruitful suggestions for revising our manuscript.

We agree that the method described here is still needed to have destructive damages for trees, eve without uprooting. As you mentioned, however, the root system evaluation is the age-long problem. In Japan, tree-pulling experiments without uprooting have been done intensively since the 2011 earthquake to estimate the tree resistance to tsunami. Generally, we could not get the information from this experiment, but the method proposed here can give an efficient indicator of belowground, particularly the maximum root depth which is a key for critical tuning moment. We added this explanation to the Introduction in the revised manuscript.

                 Moreover, we totally agree with you that “turn this insight into the manner of non-destructive testing”. Nondestructive estimates of the maximum root depth or root system structure are on the way but there is a possibility to use ground penetrating radar. We also mentioned this content in Conclusion as a future study.

 

The most important problem is unconvincing data statistical analysis, which is based on insufficient sample size and repetition – 10 trees in one soil type in same forest stand, as well as soil conditions (mechanical properties, moisture) are not considered. Sampling in three different plots of same forest was not justified as a variation with growing conditions. This study needs larger sample size, which includes different soil and site conditions, to exclude uncertainties for the statistical analysis and thus comparability. The manuscript has poor wording, which hinders the readability, thus serious work is needed for re-writing. This would help to better understand the intentions of authors, as introduction lacks a clearly stated justification of study aim and objectives.

Considering the insufficiency in sample size and necessity for improvement in writing, the manuscript is not acceptable for publication in such a state. Nevertheless, please see minor comments bellow, which might give some practical hints for improving this manuscript for eventual resubmission.

(Response)

Thank you pointing out. We agree that unconvincing data for statistical analysis. However, the coastal forest area in Japan is strictly limited to harvest trees nor to uprooting to as protected forests such as “Shifting sand prevention forest”, “Windbreak forest”, “Flood damage prevention forest”, and “Tidal wave and salty wind prevention forest”. Moreover, most coastal areas in Japan are sandy soils as present study. Thus even small number of sample size, the data on this area is quite valuable, particularly for understanding the tree resistance to tsunami. We added these contents to the Introduction to make clear objectives in this study.

 

1. Lines 29-30: Please introduce yourself with newest literature in this field.

(Response)

We added our newest literatures which have focused on resistance forces of Pinus thunbergii in coastal forest (Todo et al. 2019) and on the different root system structure such as tap or plate roots under different ground water table (Hirano et al. 2018).

 

Introduction: Lines37-40 in the revised manuscript

In previous studies, we found that the contrasting systems of the plate or tap roots of P. thunbergii under different groundwater depths in a coastal forest have different levels of resistance against tsunami winds using tree-pulling experiments and the subsequent harvesting and measurement of root systems [7,8].

 

7: Todo, C.; Tokoro, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Miyatani, K.; Doi, R.; Hirano, Y. Stability of Pinus thunbergii Between Two Contrasting Stands at Differing Distances From the Coastline. For. Ecol. Manag. 2019, 431, 44–53. DOI: 10.1016/j.foreco.2018.05.040.

8: Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.

 

 

2. Lines 30-34: With above-ground traits I would expect you to talk about tree parts, such as trunk, canopy etc. Is the formation of DBH and stem volume directly affected thus dependent on the wind or the resistance against wind loading of a tree is dependent on DBH and stem volume? Convoluted sentence – verbiage. No need to emphasize winds and tsunamis again. The cause of effect is clear from the first sentence. Please make clear statement. This sentence can be split in two. Please reference each fact of the sentence separately, instead of placing all references at the end. This hinders reading. Please do this for all sentences in the text from now on.

(Response)

As you suggested, the relevant sentence was divided into two sentences. We tried to add reference to each fact, instead of placing all references at the end.

 

Introduction: Lines45-48 in the revised manuscript

The above-ground traits of trees that are directly affected by wind and tsunamis, namely stem diameter at breast height (DBH) and stem volume related parameter (height (H) × DBH²), have a strong relationship with the critical turning moment and are thus suitable predictors of tree resistance [7,13,14,20].

 

3. Lines 34-38: Maybe with the root system structure you meant all those characteristics mentioned further? Convoluted sentence that is difficult to perceive. Please improve the wording.

(Response)

We agreed with you and revised to use one word “root system structure”.

 

Introduction: Lines49-51 in the revised manuscript

The below-ground traits, such as the root system structure, also affect the critical turning moment [9–14,16,20,22–28].

 

4. Lines 39-41: Please clarify the connection of this sentence with the previous one. Leap into the stream of thought.

(Response)

We added the sentence of the connection which explains “while the effects of aboveground traits but now move to the effects of belowground traits.”

 

Introduction: Lines51-56 in the revised manuscript

In a case study on the effect of below-ground traits, in the 17 years after thinning management, Cryptomeria japonica trees exhibited significantly higher critical turning moments than unthinned trees when trees with the same stem volumes were compared [17]. The reason for the higher critical turning moment in the thinned trees was the increased horizontal radius of the root-soil plate (RSP), which promoted horizontal root growth [17].

 

5. Lines 42-44: What exactly indicates that aboveground parts are not always predictors of tree resistance. Such explanation (connection) is missing.

(Response)

Thank you for your suggestion. There was evidence that differences in root growth at the same aboveground biomass can contribute tree resistance to uprooting. We added the following sentences with references.

 

Introduction: Lines 51-58 of the revised MS

In a case study on the effect of below-ground traits, in the 17 years after thinning management, Cryptomeria japonica trees exhibited significantly higher critical turning moments than unthinned trees when trees with the same stem volumes were compared [17]. The reason for the higher critical turning moment in the thinned trees was the increased horizontal radius of the root-soil plate (RSP), which promoted horizontal root growth [17]. Cucchi et al. [13] also reported that the critical turning moments of trees at the border of a stand were higher than those of trees from inside the stand, probably because of their larger RSP volume.

 

6. Lines 46-47: Citation needed.

(Response)

We added the relevant references.

 

Introduction: Lines 65-66 of the revised MS

Such measurements of below-ground traits are time-consuming, labor-intensive, and impractical for repeated measurements [8,31,32].

 

8: Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.

 

31: Yamase, K; Tanikawa, T; Dannoura, M; Ohashi, M; Todo, C; Ikeno, H; Aono, K; Hirano, Y. Ground-penetrating radar estimates of tree root diameter and distribution under field conditions. Trees. 2018, 32, 1657–1668. DOI: 10.1007/s00468-018-1741-9

 

32: Danjon, F.; Stokes, A.; Bakker, M.R. Root Systems of Woody Plants. In Plant Roots; The Hidden Half; CRC Press: Boca Raton, FL, USA, 2013; p. 848.

 

 

7. Lines 49-53: Please consider to use another preposition instead of “in” for starting the sentences.

(Response)

We revised them not to starting “in”.

 

Introduction: Lines 65-67 of the revised MS

Such measurements of below-ground traits are time-consuming, labor-intensive, and impractical for repeated measurements [8,31,32].

 

8: Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.

 

31: Yamase, K; Tanikawa, T; Dannoura, M; Ohashi, M; Todo, C; Ikeno, H; Aono, K; Hirano, Y. Ground-penetrating radar estimates of tree root diameter and distribution under field conditions. Trees. 2018, 32, 1657–1668. DOI: 10.1007/s00468-018-1741-9

 

32: Danjon, F.; Stokes, A.; Bakker, M.R. Root Systems of Woody Plants. In Plant Roots; The Hidden Half; CRC Press: Boca Raton, FL, USA, 2013; p. 848.

 

8. Lines 52-57: I did not understand how this derives from the previous part of the paragraph. Please provide the connection. Citations are needed.
9. Lines 55-56: “Although destructive measurements of maximum root depth are important...” on what exactly this statement is based?
10. Lines 55-58: Citation needed.

(Response)

On your comment 5, we replied to the importance of the maximum root depth for tree resistance. Tap root systems have generally deeper root depth than plate and herringbone root systems. Thus, evaluates of the maximum root depth are important for understanding the critical turning moments and tree uprooting mechanisms [16, 18, 24, 25]. We added these contents with references.

 

Introduction: Lines 71-78 of the revised MS

In a tree anchorage simulation, a tap root system exhibited a larger critical turning moment than the plate or herringbone root systems [35]. These data indicated that maximum root depth is closely related to resistance to uprooting. Thus, evaluating the maximum root depth is important for understanding the critical turning moments and tree uprooting mechanisms [16,28,34,35]. Therefore, if the maximum root depth can be estimated using simple measurements during tree-pulling experiments, the contribution of below-ground traits to uprooting resistance in various tree species can be further understood.

 

16: Nicoll, B.C.; Gardiner, B.A.; Rayner, B.; Peace, A.J. Anchorage of Coniferous Trees in Relation to Species, Soil Type, and Rooting Depth. Can. J. For. Res. 2006, 36, 1871–1883. DOI: 10.1139/x06-072.

 

28: Yang, M.; Défossez, P.; Danjon, F.; Dupont, S.; Fourcaud, T. Which Root Architectural Elements Contribute the Best to Anchorage of Pinus species? Insights From In Silico Experiments. Plant Soil. 2017, 411, 275–291. DOI: 10.1007/s11104-016-2992-0.

 

34: Yang, M.; Défossez, P.; Danjon, F.; Fourcaud, T. Analyzing Key Factors of Roots and Soil Contributing to Tree Anchorage of Pinus species. Trees. 2018, 32, 703–712. DOI: 10.1007/s00468-018-1665-4.

 

35: Dupuy, L.; Fourcaud, T.; Stokes, A. A Numerical Investigation into Factors Affecting the Anchorage of Roots in Tension. Eur. J. Soil Science 2005, 56, 319–327. DOI: 10.1111/j.1365-2389.2004.00666.x

 

11. Line 60: Pleas introduce readers what exactly is the centre point of rotation (Dcp) and how it can be determined. Also, what does Dcp indicate?
12. Lines 66-68: This thought must be placed somewhere at the beginning for introducing reader with the Dcp.

(Response)

We moved the definition and explanation of the depth at the center of rotation (Dcp) and the center of rotation (Cp) from section of Methods to Introduction to introduce readers what exactly means.

 

Introduction: Lines 79-85 of the revised MS

As a candidate method for simple measurements during tree-pulling experiments, the depth at the center point of rotation (Dcp) has been proposed as an indicator of the maximum depth of the RSP [7,17–19,36–38]. Dcp is the depth from the ground surface to the position at which the displacement of the center point of rotation (Cp) converges (Figure 1) [7,17,19,38]. Dcp was originally devised to estimate root system stability in a standing tree that was used as an anchor for cable yarding in timber logging, and was calculated against the lateral loads of wire ropes [36].

 

12. Lines 61-62: Citation needed.

(Response)

We deleted the sentences of shapes of root-soil plate.

 

14. Lines 69-72: Convoluted sentence that is difficult to perceive. Please improve the wording.

(Response)

We divided it into two sentences. The description is as follows.

 

Introduction: Lines 85-88 of the revised MS

The advantages of this indicator are twofold: first, it can be relatively quickly calculated based on video images of the test tree from the beginning of the tree-pulling experiment until immediately after recording the maximum critical turning moment [7,17,36]; second, a test tree does not need to be uprooted [7,17,36].

 

15. Lines 73-77: All this can be written shorter, avoiding verbosity.

(Response)

As you suggested, we made the description shorter as follows.

 

Introduction: Lines 111-114 of the revised MS

The results of tree-pulling experiments have shown that Dcp is positively related to the maximum critical turning moment [7,17,18,38]. The Dcp in P. thunbergii trees was deeper in a land-side plot, having a relatively higher maximum critical turning moment, than in a sea-side plot [7].

 

16. Lines 78-80: What exactly this is important. Please justify the need for such comparison and explain the contribution of that in the field of tree stability research or any practical application.

(Response)

Thank you for your suggestion. The vertical growth pattern of roots such as tap or plate root systems can be estimated by the maximum root depth. This is important for determining the degree of resistance to uprooting as shown in several references [16, 28, 34, 35].  In Japan, where tree-pulling experiments are often conducted without allowing the roots to uprooting, our proposed method is practical for further clarifying the relationship between resistance and maximum root depth. This may contribute to the formulation of guidelines for management measures such as embankment.

 

Introduction: Lines 69-78, 114-117 of the revised MS

The maximum critical turning moment is strongly affected by the maximum depth of the tap roots in Pinus species [28,34]. Deep rooting at the same stem mass as Picea sitchensis increases the critical turning moment by 10–15% compared with shallow rooting [16]. In a tree anchorage simulation, a tap root system exhibited a larger critical turning moment than the plate or herringbone root systems [35]. These data indicated that maximum root depth is closely related to resistance to uprooting. Thus, evaluating the maximum root depth is important for understanding the critical turning moments and tree uprooting mechanisms [16,28,34,35]. Therefore, if the maximum root depth can be estimated using simple measurements during tree-pulling experiments, the contribution of below-ground traits to uprooting resistance in various tree species can be further understood.

 

However, researchers have not yet examined the relationship between the Dcp and maximum root depth of the target tree in a field. To understand the uprooting mechanisms related to the maximum root depth, the accuracy of the relationship must be determined.

 

17. Line 81: Please use previously defined abbreviation if Dcp is meant by “depth at the center point of rotation” here.

(Response)

Thank you for your suggestion. We revised to use Dcp.

 

18. Lines 81-82: The explanation of the concepts of both terms of center point of rotation (Dcp) and the center of rotation (Cp) is crucially necessary somewhere at the begging. The difference between them remains unclear. Otherwise this text is highly confusing considering poor wording.

(Response)

Thank you for your suggestion. The concepts of depth at center point of rotation (Dcp) and center point of rotation (Cp) are important in this paper. Therefore, we have tried to define them properly in the Introduction. The location and the description is as follows.

 

Introduction: Lines 79-88 of the revised MS

As a candidate method for simple measurements during tree-pulling experiments, the depth at the center point of rotation (Dcp) has been proposed as an indicator of the maximum depth of the RSP [7,17–19,36–38]. Dcp is the depth from the ground surface to the position at which the displacement of the center point of rotation (Cp) converges (Figure 1) [7,17,19,38]. Dcp was originally devised to estimate root system stability in a standing tree that was used as an anchor for cable yarding in timber logging, and was calculated against the lateral loads of wire ropes [36]. The advantages of this indicator are twofold: first, it can be relatively quickly calculated based on video images of the test tree from the beginning of the tree-pulling experiment until immediately after recording the maximum critical turning moment [7,17,36]; second, a test tree does not need to be uprooted [7,17,36].

 

19. Lines 87: At which particular edge? Upper, lower, on the radii?

(Response)

Cp is the upper edge of the RSP, so we have stated it as such.

 

20. Lines 88-90: What exact improvements tree-pulling experiments will gain by monitoring the displacement of Cp? This has to be clarified.

(Response)

The displacement of Cp is measured to identify the measurement position of Dcp. We confirmed here that knowing the displacement of Cp, the measurement of Dcp should be at the depth of Cp when the maximum moment is recorded.

 

Introduction: Lines 118-127 of the revised MS

The Cp position can change with increasing load during tree-pulling experiments, because the respective positions of the rotation axis depend on the soil properties and root structure [40]. In tree-pulling experiments on C. japonica and C. obtusa, Cp was concentrated at a single point just below the stem [18, 36]. In contrast, Coutts [9] re-ported that the Cp of P. sitchensis during uprooting is located at the upper edge of the RSP on the leeward side. The displacement of Cp must be determined to indicate the measured position of Dcp. Therefore, the displacement of Cp should be monitored during tree-pulling experiments to clarify the discrepancy in the position of Cp at the time of the maximum critical turning moment, that is, Dcp. This will support the use of Dcp at the right position as an indicator of the maximum root depth in tree-pulling experiments.

 

21. Lines 100-104 and Table 1: This belongs to Materials and Methods.

(Response)

In the revised manuscript, we deleted the lines 100-104. Table 1 has been moved to Materials and Methods.

 

22. Lines 118-136: This belongs to Introduction.

(Response)

Thank you for your suggestion. As suggested, we have moved the theory of Dcp measurement to the Introduction.

 

Introduction: Lines 79-98, 104-110 of the revised MS

As a candidate method for simple measurements during tree-pulling experiments, the depth at the center point of rotation (Dcp) has been proposed as an indicator of the maximum depth of the RSP [7,17–19,36–38]. Dcp is the depth from the ground surface to the position at which the displacement of the center point of rotation (Cp) converges (Figure 1) [7,17,19,38]. Dcp was originally devised to estimate root system stability in a standing tree that was used as an anchor for cable yarding in timber logging, and was calculated against the lateral loads of wire ropes [36]. The advantages of this indicator are twofold: first, it can be relatively quickly calculated based on video images of the test tree from the beginning of the tree-pulling experiment until immediately after recording the maximum critical turning moment [7,17,36]; second, a test tree does not need to be uprooted [7,17,36]. As the tree starts to uproot along the boundary between the outside soil and the RSP in response to lateral forces [6], the assumptions are that (i) the tree rotates at a point on the vertical line through the stem center within a relatively small range of inclination (Figure 1a) and (ii) both the lateral movement of the tree toward the load and the bending range of the stem around the root trunk are negligible. In this study, we set measuring points P1 and P2 on a vertical straight line through the stem center of a test tree at different heights from the ground surface without any lateral forces. After loading the lateral forces, we measured the new positions of P1 (P1') and P2 (P2'). We measured the lines passing through P1 and P2 and through P1' and P2' intersect as the Cp, and the depth from the ground surface to Cp as the Dcp (Figure 1a, [36]).

 

Morioka [36] showed that the Dcp of C. japonica trees with a DBH of 6–21 cm was 13–52 cm, and concluded that the outer range reflects the depth of the RSP. During a tree-pulling experiment, Nonoda et al. [18] demonstrated that Chamaecyparis obtusa rotated at the center point just below the stem. Kamimura et al. [39] reported that the horizontal displacement of C. obtusa stems after loading with lateral forces was only a few millimeters and was thus negligible. Because the maximum root depth increases with increasing RSP depth [10], the maximum root depth can be assumed to be related to the Dcp.

 

23. Figure 1: This should be integrated in the description of methods.

(Response)

We decided to keep Figure 1 in the Introduction because the reader can understand the theory of Dcp easily with Figure 1.

 

24. Figure 2: This is redundant and tells nothing important. This information can be placed in the text.

(Response)

We would like to show plots in our study site where is closed to the sea in this Figure. Thus we moved to the Supplemental section as Figure S1.

　

25. Line 150: Are there other plots as well? What are these plots – long-term monitoring e.g.?

(Response)

There are no other plots. The plots in this study are the areas where we obtained information on tree-pulling experiments and trees. The continuous P. thunbergii forest in this coastal area was divided into three sections based on the depth of groundwater table and the growth of P. thunbergii trees. Thus the plots are representative of these areas. Within the plots, there are areas where trees were dug up the root system. We also summarized the conditions of plots to Table S1. We added the description as follows.

 

Materials and Methods: Lines 158-169 of the revised MS

Within the continuous P. thunbergii forest, the stands had different tree growths, which was due to the different depths of the groundwater table (Tables 1 and S1) [7,8,41,42]. Therefore, we divided the stands into three sections, and surveyed representative sites as plots (Figure S1 [7,8,41,42]). Plot A was located 188–199 m from the shoreline [41,42]; plot B was also located close to the shoreline (150–270 m). Plot C was further from the shoreline (620–740 m [7]; Figure S1). The mean age of the P. thunbergii in the three plots was 45 years. The P. thunbergii stand density was lowest in plot A (100–200 individuals ha^–1), and mid-range in plot B (200–400 individuals ha^–1), and highest in plot C (400–1000 individuals ha^–1). Table S1 summarizes the other characteristics, such as soil water content and understory species among the plot. We established the locations where we dug the test trees within the plots. The areas where the P. thunbergii trees that we dug out were 400 × 15 m for plot A, 120 × 180 m for plot B, and 170 × 120 m for plot C.

 

 

26. Lines 161-164; On what tree selection was based?
27. Lines 163-164: What exactly limited the sample size?

(Response)

As we mentioned above, the coastal forest area in Japan is strictly limited to harvest trees nor to uprooting to as protected forests such as “Shifting sand prevention forest”, “Windbreak forest”, “Flood damage prevention forest”, and “Tidal wave and salty wind prevention forest”. Moreover, most coastal areas in Japan are sandy soils as present study. Thus even small number of sample size, the data on this area is quite valuable, particularly for understanding the tree resistance to tsunami. The test trees are representative, i.e., the mean tree aboveground growth, from within those three plots. Japanese coastal forests have legal restrictions of logging that prohibit digging out the entire root system. Hence we only did minimal digging.

 

Materials and Methods: Lines 172-175 of the revised MS

We selected ten P. thunbergii trees from the three plots to examine the relationships between the maximum critical turning moment, Dcp, and maximum root depth (Table 1). We selected only 10 average trees because complete excavation was legally limited in Japan for forest protection.

 

28. Lines 164-165: Every abbreviation must be explained first.

(Response)

In the introduction, we already explained the abbreviation of DBH.

 

Introduction: Lines 45-48 of the revised MS

The above-ground traits of trees that are directly affected by wind and tsunamis, namely stem diameter at breast height (DBH) and stem volume related parameter (height (H) × DBH²), have a strong relationship with the critical turning moment and are thus suitable predictors of tree resistance [7,13,14,20].

 

29. Lines 166-169: Detailed description of different parameters among A, B, and C plots is required if this notable differences in, e.g. growing conditions, tree dimensions are present. Otherwise, this is verbosity, which inflates text hindering to follow it.

(Response)

Thank you for your suggestion. As mentioned above, we divided the P. thunbergii trees with different growth into three sections, each of which was designated as a plot. The characteristics of the plots are clearly described in Table 1 and Table S1. Descriptions in the manuscriptions are as follows.

 

Materials and Methods: Lines 147-148, 158-169 of the revised MS

The groundwater level was almost constant from the shoreline inland [41]. However, the depth of the groundwater level from the ground surface deepened inland as elevation increased (Table S1).

 

Within the continuous P. thunbergii forest, the stands had different tree growths, which was due to the different depths of the groundwater table (Tables 1 and S1) [7,8,41,42]. Therefore, we divided the stands into three sections, and surveyed representative sites as plots (Figure S1 [7,8,41,42]). Plot A was located 188–199 m from the shoreline [41,42]; plot B was also located close to the shoreline (150–270 m). Plot C was further from the shoreline (620–740 m [7]; Figure S1). The mean age of the P. thunbergii in the three plots was 45 years. The P. thunbergii stand density was lowest in plot A (100–200 individuals ha^–1), and mid-range in plot B (200–400 individuals ha^–1), and highest in plot C (400–1000 individuals ha^–1). Table S1 summarizes the other characteristics, such as soil water content and understory species among the plot. We established the locations where we dug the test trees within the plots. The areas where the P. thunbergii trees that we dug out were 400 × 15 m for plot A, 120 × 180 m for plot B, and 170 × 120 m for plot C.

 

30. Lines 170-178: Raw measurement of pulling load is not a turning (bending) moment. A turning moment depends also on the height of anchoring point and the angle of pulling line (rope) and is expressed in kNm. In your test, these pulling heights differed among plots (1 to 1.5 m), which implies biased results if raw measurements of pulling load are attributed as a critical turning moment. This is not correct as lower anchoring height will provide higher loading, but this would not be an issue if turning moment is calculated considering the anchoring height. Please provide a formula how the critical turning moment was calculated using measured load if it has been done at all. Otherwise, your results require recalculation.

(Response)

Thank you for your suggestion. We used the definition of “critical turning moment by Nicol et al. (2006); ”The anchorage of trees can be expressed as the critical (maximum) resistive turning moment at the base of the stem during overturning. Turning moment is defined simply as force × length of a lever arm. As you pointed out, the anchor height and rope angle are important for the critical turning moment. Since the explanation was missing, we have added the equation. The height of the anchor is taken into account when calculating the critical turning moment. Also, the rope is pulled parallel to the ground surface, but the tree is slightly tilted, so the equation takes that into account. The following statement was added to the introduction, materials and methods section.

 

Introduction: Lines 41-45 of the revised MS

The resistance of trees to uprooting caused by strong winds or tsunamis can be estimated as the critical turning moment [7,9–15], which is defined as the force × length of the lever arm [16]. To evaluate critical turning moments in Japan, tree-pulling experiments have often been conducted without uprooting because of legal regulations regarding forest preservation [7,17–19].

 

Materials and Methods: Lines 190-193 of the revised MS

We calculated the critical turning moment (M) using Equation (1) [7,42].

M = F × h × cosθ₀                                                                                       (1)

where F (kN) is the maximum applied force, h (m) is the attachment height of the pulling sling, and θ₀ (°) is the angle of the vertically inclined trees.

 

31. Figure S1: Caption needed.

(Response)

A caption has been added to the supplemental figure.

 

32. Figure 3: This is redundant and tells nothing important. This information can be placed in the text.

(Response)

The "Time Variation of Moments" graph in Figure 3 has been moved to the supplemental section, as we believe it aids in understanding the movement of the Cp position.

 

33. Lines 196-203: The description how did you scaled the measured displacement of stem is needed more detailed.
34. Lines 213-215: Please provide the system (units, scale) of coordinates used to calculate misalignment of Cp (msaCp).

(Response)

The image is taken together with the scale and the actual distance per pixel (cm) is converted. After measuring the vertical and horizontal displacement as pixels, the scale converted to real distance (cm) using the previous conversion.

 

Materials and Methods: Lines 224-226 of the revised MS

We measured the length per pixel on a scale. We converted the horizontal and vertical displacements measured in pixels to displacements in centimeters using the length per pixel.

 

35. Equation 1: Please provide reference or explain steps by which this was derived.

(Response)

This formula means the displacement of both vertically and horizontally from one time to the final time in Figure 5(a). The coordinate system is two-dimensional. The distance was just　calculated using the most common formula of Euclidean distance. We described it as follows.

 

Materials and Methods: Lines 226-232 of the revised MS

We evaluated the misalignment of Cp (msaCp) toward zero from the start of pulling by the commonly used Euclidean distance using the following formula:

formula   (2)

where x and y are the horizontal and vertical coordinates of Cp, respectively, at each time point during the pulling experiments; x_m and y_m are the horizontal and vertical coordinates of Cp, respectively, when the critical turning moment is the maximum.

36. Line 220: “The stem was assumed to be a rigid body.” – For what reason this was done? What does that mean?

(Response)

It was important that there be no significant deflection in the stem in order to measure Dcp by connecting the marks on the stem with a line [36]. In this case, no large deflection of the test tree was observed when the video was taken, and thus, it was assumed to be rigid. We revised as the following sentences;

 

Materials and Methods: Lines 233-238 of the revised MS

It was important that there be no significant deflection in the stem in order to measure Dcp by connecting the marks on the stem with a line [36]. In this case, we observed no large deflection of the test tree when a video was captured. The wire rope attached during the pulling experiment was approximately 1 m above the ground, and we assumed the stem to be a rigid body. We used open-source software ImageJ version 1.51 [48] for image processing and measurements.

 

37. Lines 239-241: The sample size of 10 trees is too small for reasonable evaluation of normal distribution.
38. Lines 243-244; Lines 278-295; Table 2: Such small sample size (10 trees) might cause biased correlations. Obtained p-values must be multiplied with number of correlation pairs.

(Response)

The limited number of individuals harvested by legal restrictions made it difficult to increase the sample size, in other words, but thus it is valuable even the smaller sample size. The samples were checked for normality with the Kolmogorov-Smirnov test, but just to be sure, a nonparametric method was used to analyze the samples.

 

38. Lines 241-243: Why there is a comparison between plots? Are they separated stands with different growing conditions, management history, proveniences? Or those plots are just an artificial man-made division where sampling was done? If so, such comparison is does not make this study relevant.

(Response)

Thank you for your suggestion. As mentioned above about setting plots with different pine growth for reply to the major comment, we decided here the statistics among plots were removed because of size limitation.

 

40. Figure 5 (a): What does the indication of time of reaching the critical turning moment tell?

(Response)

The time notation is necessary to show that Cp does not move at times close to the maximum critical turning moment. The depth of Cp at the time that exhibits the maximum critical turning moment is Dcp.

The following additions were made to the manuscript.

 

Results: Lines 266-268 of the revised MS

The results showed that the position of Cp did not move with increasing tensile time when the critical turning moment was close to its maximum value compared with just after the start time of the tree-pulling experiment (Figure 3b).

 

41. Lines 307-309: Please provide concentrated results in one sentence to remind the reader how exactly.

(Response)

We indicated at the beginning of the paragraph that Dcp is an indicator of the maximum root depth in the actual field measurement.

 

Discussion: Lines 321-324 of the revised MS

We found that Dcp is a suitable indicator of the maximum root depth of P. thunbergii grown in sandy soil in a tree-pulling experiment. Several researchers have measured Dcp [7,17–19,36,38,42]; however, evidence demonstrating its value as a below-ground indicator is lacking.

 

42. Lines 310-311: And how do you think why? Please justify why your results on Dcp are better compared with other studies.

(Response)

This is the only study that directly measures the maximum root depth and shows its relationship to Dcp.

The manuscript states as follows:

 

Discussion: Lines 321-325 of the revised MS

We found that Dcp is a suitable indicator of the maximum root depth of P. thunbergii grown in sandy soil in a tree-pulling experiment. Several researchers have measured Dcp [7,17–19,36,38,42]; however, evidence demonstrating its value as a below-ground indicator is lacking. This is the first study in which root depth was measured using actual measurements and compared with the Dcp values.

 

43. Lines 315-319: Such generalizing statement derived from sample size of 10 trees is contradictory.
44. Lines 409-411: Exactly, this in combination with larger sample size is what would make this study relevant.

(Response)

Thank you for your remarks on the sample size. Due to logging restrictions as we mentioned above, it is difficult to increase the sample size in this forest. However, we have analyzed the available data to the best of our ability using the valuable harvesting data in this forest. In the future, we would like to obtain data on root systems while also considering non-destructive surveys as you suggested.

 

44. Lines 328-331: There is insufficient empirical basis for such statement, as compression of soil bellow the root plate has been reported particularly on saturated soils. To claim that, your sample size must be enlarged including different soil types, as well as data on soil density and moisture content.

(Response)

Thank you for your suggestion. The results from this survey are only data on Pinus thunbergii   grown in coastal sandy soils. We do not have enough data to generalize to various locations yet. However, we believe that they are worthy of evidence if they are limited to the location of sandy coastal forests of a single tree species.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors

A lot of corrections have been made in the revised version of the article.

Thanks

Author Response

Thank you very much for providing important comments. We are thankful for the time and energy you expended. 

Reviewer 2 Report

Comments for author File: Comments.pdf

Author Response

Responses to comments by Reviewer 2's report

 

*Italic letters show your comments.

*Normal letters (red) show our responses.

 

To be honest, I did not find precise answers to few comments I gave in Round 1. It is not enough just to write that some things are implemented in the text. I would like to see particular changes as quotes with corresponding line numbers right after the answer to my comments here.

Here are the points from previous round, which do not have convincing answers as well as quotes with changes in the manuscript (improvements must be pointed out here). These comments include several points that must be answered.

(Response)

Thank you for review our manuscript again, and we are sorry not to indicate “changes as quotes with corresponding line number right after the answer to your comments” As for the following section, we added the previous response to the corresponding line number in the text.

 

 

“The topic of the manuscript is relevant and with high scientific interest, as authors have focused on the age-long problem of facilitating root evaluation with minimized effort and investments. The study offers an approach to assess the rooting distribution without time consuming excavations or tree uprooting by winching. In general, trees just need to be pulled to reach the maximum resistive turning moment. In this regard, the necessity of applying such approach stayed unclear, as tree damage is not prevented anyway¹⁾. I would suggest authors to turn this insight into the manner of non-destructive testing²⁾.”

(Response)

Underline 1) We agree that the method described here is still needed to have destructive damages for trees, eve without uprooting. As you mentioned, however, the root system evaluation is the age-long problem. In Japan, tree-pulling experiments without uprooting have been done intensively since the 2011 earthquake to estimate the tree resistance to tsunami. Generally, we could not get the information of root systems from this experiment, but the method proposed here can give an efficient indicator of belowground, particularly the maximum root depth which is a key for critical tuning moment. We added this explanation to the Introduction in the revised manuscript.

Introduction: Lines 30-40 in the revised manuscript

Coastal forests play an important role in preventing damage from strong salty winds caused by the sea and tsunamis [1,2]. Most coastal areas in Japan have been planted with Pinus thunbergii, which has been resistant to salt and wind for the past several hundred years [3,4]. In 2011, the Great East Japan Earthquake triggered a tsunami that severely damaged the coastal P. thunbergii forests, which caused uprooting and stem breakage in P. thunbergii in several coastal forests [5]. Although P. thunbergii originally had deep tap roots [6], the uprooted trees in the damaged coastal forests had plate root systems because of the shallower groundwater table [5]. In previous studies, we found that the contrasting systems of the plate or tap roots of P. thunbergii under different groundwater depths in a coastal forest have different levels of resistance against tsunami winds using tree-pulling experiments and the subsequent harvesting and measurement of root systems [7,8].

1. Murai, H.; Ishikawa, M.; Endo, J.; Tadaki, Y. The Coastal Forest in Japan. Soft Science Inc., Tokyo, 1992, p. 513 (in Japanese).
2. Zhu, J.; Matsuzaki, T.; Sakioka, K. Windspeeds within a single crown of Japanese black pine (Pinus thunbergii). For. Ecol. Manag. 2000, 135, 19–31. DOI: 10.1016/S0378-1127(00)00295-4.
3. Konta, F. The present conditions and functions of the coastal forests in Japan. Jpn. Soc. Coast. For. 2001,1, 1–4 (in Japanese with English summary).
4. Ogawa, M. Microbial flora in Pinus thunbergii forest of coastal sand dune. For. For. Prod. Res. Inst. 1979, 305, 107–124 (in Japanese with English summary).
5. Forestry and Forest Products Research Institute, Japan. Regeneration of coastal forests affected by tsunami. 2012, pp. 1–24.
6. Karizumi, N. The Illustrations of Tree Roots; Seibundoshinko-sya: Tokyo, 1979; p. 1121. (in Japanese)
7. Todo, C.; Tokoro, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Miyatani, K.; Doi, R.; Hirano, Y. Stability of Pinus thunbergii Between Two Contrasting Stands at Differing Distances From the Coastline. Ecol. Manag. 2019, 431, 44–53. DOI: 10.1016/j.foreco.2018.05.040.
8. Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 1007/s11104-018-3630-9.

 

Introduction: Lines 64-78 in the revised manuscript

To measure below-ground traits, uprooting and digging of root systems have been adopted in tree-pulling experiments [4,29,30]. Such measurements of below-ground traits are time-consuming, labor-intensive, and impractical for repeated measurements [8,31,32]. The maximum root depth is particularly difficult to directly measure [8] and depends on tree species, soil conditions, presence of bedrock, and groundwater level [25,29,30,33]. The maximum critical turning moment is strongly affected by the maximum depth of the tap roots in Pinus species [28,34]. Deep rooting at the same stem mass as Picea sitchensis increases the critical turning moment by 10–15% compared with shallow rooting [16]. In a tree anchorage simulation, a tap root system exhibited a larger critical turning moment than the plate or herringbone root systems [35]. These data indicated that maximum root depth is closely related to resistance to uprooting. Thus, evaluating the maximum root depth is important for understanding the critical turning moments and tree uprooting mechanisms [16,28,34,35]. Therefore, if the maximum root depth can be estimated using simple measurements during tree-pulling experiments, the contribution of below-ground traits to uprooting resistance in various tree species can be further understood.

4. Ogawa, M. Microbial flora in Pinus thunbergii forest of coastal sand dune. Bull. For. For. Prod. Res. Inst. 1979, 305, 107–124 (in Japanese with English summary).
5. Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.
6. Nicoll, B.C.; Gardiner, B.A.; Rayner, B.; Peace, A.J. Anchorage of Coniferous Trees in Relation to Species, Soil Type, and Rooting Depth. Can. J. For. Res. 2006, 36, 1871–1883. DOI: 10.1139/x06-072.
7. Yang, M.; Défossez, P.; Danjon, F.; Dupont, S.; Fourcaud, T. Which Root Architectural Elements Contribute the Best to Anchorage of Pinus species? Insights From In Silico Experiments. Plant Soil. 2017, 411, 275–291. DOI: 10.1007/s11104-016-2992-0.
8. Preti, F. Forest Protection and Protection Forest: Tree Root Degradation Over Hydrological Shallow Landslides Trig-gering. Ecol. Eng. 2013, 61, 633– 645. DOI: 10.1016/j.ecoleng.2012.11.009.
9. Giadrossich, F.; Schwarz, M.; Marden, M.; Marrosu, R.; Phillips, C. Minimum Representative Root Distribution Sampling for Calculating Slope Stability in Pinus radiata D.Don Plantations in New Zealand. N. Z. J. For. Sci. 2020, 50, 5. DOI: 10.33494/nzjfs502020x68x.
10. Yamase, K; Tanikawa, T; Dannoura, M; Ohashi, M; Todo, C; Ikeno, H; Aono, K; Hirano, Y. Ground-penetrating radar estimates of tree root diameter and distribution under field conditions. Trees. 2018, 32, 1657–1668. DOI: 10.1007/s00468-018-1741-9.
11. Danjon, F.; Stokes, A.; Bakker, M.R. Root Systems of Woody Plants. In Plant Roots; The Hidden Half; CRC Press: Boca Raton, FL, USA, 2013; p. 848.
12. Kokutse, N.K.; Temgoua,A.G.T.; Kavazovi ́, Z. Slope Stability and Vegetation: Conceptual and Numerical Investigation of Mechanical Effects. Ecol. Eng. 2016, 86, 146–153. DOI: 10.1016/j.ecoleng.2015.11.005.
13. Yang, M.; Défossez, P.; Danjon, F.; Fourcaud, T. Analyzing Key Factors of Roots and Soil Contributing to Tree Anchorage of Pinus species. Trees. 2018, 32, 703–712. DOI: 10.1007/s00468-018-1665-4.
14. Dupuy, L.; Fourcaud, T.; Stokes, A. A Numerical Investigation into Factors Affecting the Anchorage of Roots in Tension. Eur. J. Soil Science 2005, 56, 319–327. DOI: 10.1111/j.1365-2389.2004.00666.x

 

(Response)

Underline 2) Moreover, we totally agree with you that “turn this insight into the manner of non-destructive testing”. Nondestructive estimates of the maximum root depth or root system structure are on the way but there is a possibility to use ground penetrating radar. We also mentioned this content in Conclusion as a future study as follows;

Conclusions: Lines 444-449 in the revised manuscript

However, the Dcp must still be measured for destructive tree-pulling experiments. Recently, nondestructive geophysical surveys on soil surfaces using ground penetrating radar have been used to estimate the position and diameter of tree roots as point datasets [31,55]. When an algorithm to reconstruct the point data of root position and size into the whole root system structure is established in a specific tree species [56], the maximum root depth can be nondestructively estimated.

• Yamase, K; Tanikawa, T; Dannoura, M; Ohashi, M; Todo, C; Ikeno, H; Aono, K; Hirano, Y. Ground-penetrating radar estimates of tree root diameter and distribution under field conditions. Trees. 2018, 32, 1657–1668. DOI: 10.1007/s00468-018-1741-9.
• Hirano, Y.; Yamamoto, R.; Dannoura, M.; Aono, K.; Igarashi, T.; Ishii, M.; Yamase, K.; Makita, N.; Kanazawa, Y. Detection Frequency of Pinus thunbergii Roots by Ground-Penetrating Radar Is Related to Root Biomass. Plant Soil 2012, 360, 363–373. DOI: 10.1007/s11104-012-1252-1
• Ohashi, M.; Ikeno, H.; Sekihara, K.; Tanikawa, T.; Dannoura, M.; Yamase, K.; Todo, C.; Tomita, T.; Hirano, Y. Reconstruction of Root Systems in Cryptomeria japonica Using Root Point Coordinates and Diameters. Planta 2019, 249, 445–455. DOI: 10.1007/s00425-018-3011-x

 

“The most important problem is unconvincing data statistical analysis, which is based on insufficient sample size and repetition – 10 trees in one soil type in same forest stand³⁾, as well as soil conditions (mechanical properties, moisture) are not considered⁴⁾. Sampling in three different plots of same forest was not justified as a variation with growing condition⁴⁾. This study needs larger sample size, which includes different soil and site conditions, to exclude uncertainties for the statistical analysis and thus comparability⁵⁾. The manuscript has poor wording, which hinders the readability, thus serious work is needed for rewriting⁶⁾. This would help to better understand the intentions of authors, as introduction lacks a clearly stated justification of study aim and objectives⁷⁾.

 

(Response)

Thank you pointing out. We agree that unconvincing data for statistical analysis. However, as we mentioned above, the coastal forest area in Japan is strictly limited to harvest trees nor to uprooting to as protected forests such as “Shifting sand prevention forest”, “Windbreak forest”, “Flood damage prevention forest”, and “Tidal wave and salty wind prevention forest”. Moreover, most coastal areas in Japan are sandy soils as present study. Thus even small number of sample size, the data on this area is quite valuable, particularly for understanding the tree resistance to tsunami. We added these contents to the Introduction and Methods to make clear objectives in this study.

 

Under line 3) Introduction: Lines 41-45 of the revised MS

The resistance of trees to uprooting caused by strong winds or tsunamis can be estimated as the critical turning moment [7,9–15], which is defined as the force × length of the lever arm [16]. To evaluate critical turning moments in Japan, tree-pulling ex-periments have often been conducted without uprooting because of legal regulations regarding forest preservation [7,17–19].

7. Todo, C.; Tokoro, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Miyatani, K.; Doi, R.; Hirano, Y. Stability of Pinus thunbergii Between Two Contrasting Stands at Differing Distances From the Coastline. For. Ecol. Manag. 2019, 431, 44–53. DOI: 10.1016/j.foreco.2018.05.040.
8. Coutts, M.P. Components of Tree Stability in Sitka Spruce on Peaty Gley Soil. Forestry. 1986, 59, 173–197. DOI: 10.1093/forestry/59.2.173.
9. Ray, D.; Nicoll, B.C. The Effect of Soil Water-Table Depth on Root-Plate Development and Stability of Sitka Spruce. Forestry. 1998, 71, 169–182. DOI: 10.1093/forestry/71.2.169.
10. Moore, J.R. Differences in Maximum Resistive Bending Moments of Pinus radiata Trees Grown on a Range of Soil Types. For. Ecol. Manag. 2000, 135, 63–71. DOI: 10.1016/S0378-1127(00)00298-X.
11. Peltola, H.; Kellomäki, S.; Hassinen, A.; Granander, M. Mechanical Stability of Scots Pine, Norway Spruce and Birch: an Analysis of Tree-Pulling Experiments in Finland. For. Ecol. Manag. 2000, 135, 143–153. DOI: 10.1016/S0378-1127(00)00306-6.
12. Cucchi, V.; Meredieu, C.; Stokes, A.; Berthier, S.; Bert, D.; Najar, M.; Denis, A.; Lastennet, R. Root Anchorage of Inner and Edge Trees in Stands of Maritime Pine (Pinus pinaster Ait.) Growing in Different Podzolic Soil Conditions. Trees. 2004, 18, 460–466. DOI: 10.1007/s00468-004-0330-2.
13. Nicoll, B.C.; Achim, A.; Mochan, S.; Gardiner, B.A. Does Steep Terrain Influence Tree Stability? A Field Investigation. Can. J. For. Res. 2005, 35, 2360–2367. DOI: 10.1139/x05-157.
14. Tanaka, N. Effectiveness and Limitations of Coastal Forest in Large Tsunami: Conditions of Japanese Pine Trees on Coastal Sand Dunes in Tsunami Caused by Great East Japan Earthquake. J. Jpn. Soc. Civ. Eng. Ser. B1. 2012, 68, II_7–II_15. DOI: 10.2208/jscejhe.68.II_7.
15. Nicoll, B.C.; Gardiner, B.A.; Rayner, B.; Peace, A.J. Anchorage of Coniferous Trees in Relation to Species, Soil Type, and Rooting Depth. Can. J. For. Res. 2006, 36, 1871–1883. DOI: 10.1139/x06-072.
16. Todo, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Hirano, Y. Effect of Thinning on the Critical Turning Moment of Sugi (Cryptomeria japonica (L. f.) D. Don). J. Jpn. Soc. Reveget Tec. 2015, 41, 308–314. (in Japanese with English summary) DOI: 10.7211/jjsrt.41.308.
17. Nonoda, T.; Hayashi, S.; Kawabe, H.; Honda, K.; Koyabu, K. The Mechanism of the Tree-Uprooting Occurred by Pulling Down a Tree. J. Jpn. For. Soc. 1996, 78, 390–397. (in Japanese with English summary) DOI: 10.11519/jjfs1953.78.4_390.
18. Okada, Y. Measuring the Critical Turning Moment of the Japanese Cedar (Cryptomeria japonica) In Situ. J. For. Res. 2019, 24, 168–177. DOI: 10.1080/13416979.2019.1617098.

 

Materials and Methods: Lines 172-175 of the revised MS

We selected ten P. thunbergii trees from the three plots to examine the relationships between the maximum critical turning moment, Dcp, and maximum root depth (Table 1). We selected only 10 average trees because complete excavation was legally limited in Japan for forest protection.

 

 

Underline 4) Materials and Methods: Lines 147-148, 158-169 of the revised MS

The groundwater level was almost constant from the shoreline inland [41]. However, the depth of the groundwater level from the ground surface deepened inland as elevation increased (Table S1).

 

Materials and Methods: Lines 158-169 of the revised MS

Within the continuous P. thunbergii forest, the stands had different tree growths, which was due to the different depths of the groundwater table (Tables 1 and S1) [7,8,41,42]. Therefore, we divided the stands into three sections, and surveyed representative sites as plots (Figure S1 [7,8,41,42]). Plot A was located 188–199 m from the shoreline [41,42]; plot B was also located close to the shoreline (150–270 m). Plot C was further from the shoreline (620–740 m [7]; Figure S1). The mean age of the P. thunbergii in the three plots was 45 years. The P. thunbergii stand density was lowest in plot A (100–200 individuals ha^–1), and mid-range in plot B (200–400 individuals ha^–1), and highest in plot C (400–1000 individuals ha^–1). Table S1 summarizes the other characteristics, such as soil water content and understory species among the plot. We established the locations where we dug the test trees within the plots. The areas where the P. thunbergii trees that we dug out were 400 × 15 m for plot A, 120 × 180 m for plot B, and 170 × 120 m for plot C.

7. Todo, C.; Tokoro, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Miyatani, K.; Doi, R.; Hirano, Y. Stability of Pinus thunbergii Between Two Contrasting Stands at Differing Distances From the Coastline. For. Ecol. Manag. 2019, 431, 44–53. DOI: 10.1016/j.foreco.2018.05.040.
8. Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.
9. Tanaka, J.; Nakazawa, H.; Satou, T. Case Study of Root Systems Growth of Pinus thunbergii Parlatore and Groundwater Level in the Nishinohama Coastal Forest, Tahara City, Aichi Prefecture. J. Jpn. Soc. Reveget Tec. 2017, 43, 298–301. (in Japanese) DOI: 10.7211/jjsrt.43.298
10. Todo, C.; Ikeno, H.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Dannoura, M.; Kimura, T.; Hirano, Y. Reconstruction of Conifer Root Systems Mapped With Point Cloud Data Obtained by 3D Laser Scanning Compared With Manual Meas-urement. Forests. 2021, 12, 1117. DOI: 10.3390/f12081117.

 

(Response)

Underline 5) The limited number of individuals harvested by legal restrictions made it difficult to increase the sample size, in other words, but thus it is valuable even the smaller sample size. The samples were checked for normality with the Kolmogorov-Smirnov test, but just to be sure, a nonparametric method was used to analyze the samples.

Materials and Methods: Lines 254-256 of the revised MS

We used the Mann–Whitney U test to detect the differences between the horizontal and vertical displacements. We tested the relationships between traits using Spearman's rank correlation coefficient.

 

 

Underline 6) The whole manuscript

We rewrite the whole manuscript. The English of revised manuscript was carefully corrected by native-English speaking, professional editor, with extensive research editing experience, “editage, https://www.editage.jp/. We added the following certification.

 

 

Underline 7) Introduction: Lines 30-40 of the revised MS

Coastal forests play an important role in preventing damage from strong salty winds caused by the sea and tsunamis [1,2]. Most coastal areas in Japan have been planted with Pinus thunbergii, which has been resistant to salt and wind for the past several hundred years [3,4]. In 2011, the Great East Japan Earthquake triggered a tsunami that severely damaged the coastal P. thunbergii forests, which caused uprooting and stem breakage in P. thunbergii in several coastal forests [5]. Although P. thunbergii originally had deep tap roots [6], the uprooted trees in the damaged coastal forests had plate root systems because of the shallower groundwater table [5]. In previous studies, we found that the contrasting systems of the plate or tap roots of P. thunbergii under different groundwater depths in a coastal forest have different levels of resistance against tsunami winds using tree-pulling experiments and the subsequent harvesting and measurement of root systems [7,8].

1. Murai, H.; Ishikawa, M.; Endo, J.; Tadaki, Y. The Coastal Forest in Japan. Soft Science Inc., Tokyo, 1992, p. 513 (in Japanese).
2. Zhu, J.; Matsuzaki, T.; Sakioka, K. Windspeeds within a single crown of Japanese black pine (Pinus thunbergii Parl.). For. Ecol. Manag. 2000, 135, 19–31. DOI: 10.1016/S0378-1127(00)00295-4.
3. Konta, F. The present conditions and functions of the coastal forests in Japan. J. Jpn. Soc. Coast. For. 2001,1, 1–4 (in Japanese with English summary).
4. Ogawa, M. Microbial flora in Pinus thunbergii forest of coastal sand dune. Bull. For. For. Prod. Res. Inst. 1979, 305, 107–124 (in Japanese with English summary).
5. Forestry and Forest Products Research Institute, Japan. Regeneration of coastal forests affected by tsunami. 2012, pp. 1–24.
6. Karizumi, N. The Illustrations of Tree Roots; Seibundoshinko-sya: Tokyo, 1979; p. 1121. (in Japanese)
7. Todo, C.; Tokoro, C.; Yamase, K.; Tanikawa, T.; Ohashi, M.; Ikeno, H.; Dannoura, M.; Miyatani, K.; Doi, R.; Hirano, Y. Stability of Pinus thunbergii Between Two Contrasting Stands at Differing Distances From the Coastline. For. Ecol. Manag. 2019, 431, 44–53. DOI: 10.1016/j.foreco.2018.05.040.
8. Hirano, Y.; Todo, C.; Yamase, K.; Tanikawa, T.; Dannoura, M.; Ohashi, M.; Doi, R.; Wada, R.; Ikeno, H. Quantification of the Contrasting Root Systems of Pinus thunbergii in Soils With Different Groundwater Levels in a Coastal Forest in Japan. Plant Soil. 2018, 426, 327–337. DOI: 10.1007/s11104-018-3630-9.

 

Author Response File: Author Response.pdf