Crack Propagation and Fracture Process Zone (FPZ) of Wood in the Longitudinal Direction Determined Using Digital Image Correlation (DIC) Technique
Round 1
Reviewer 1 Report
Dear Authors,
You have made a good effort in addressing the cohesive fracture mechanics and its characterization through Digital Image Correlation. The methodology and motivation behind the quasi-brittle fracture in heterogeneous layered structure as wood is explained well, however a few details are not supported with appropriate citations. The underlying physics behind development of three different stages during three point bend test should be explained well. For an example, a softening region in P1-P4 has some load oscillations. These need to be explained in detail to characterize the behavior of wood. Relevant literature is suggested for citations to make the paper more meaningful. Please see the comments in attached document.
The three point bend set up is used here. The same can be performed with four point bend, so explain the choice. Suggestions are given in the attached document. The DIC results look convincing, but a few details on facet size and its deformation in subsequent loading stages should be addressed.
The figure 1 needs a detailed caption to show dimensions in mm. The figure 2 should point out the longitudinal direction f fibers explicitly. In figure 11, explain the details of solid and dotted lines.
Many other points are suggested in the document.
Comments for author File: 
Comments.pdf
Author Response
You have made a good effort in addressing the cohesive fracture mechanics and its characterization through Digital Image Correlation. The methodology and motivation behind the quasi-brittle fracture in heterogeneous layered structure as wood is explained well, however a few details are not supported with appropriate citations.
Point 1: The underlying physics behind development of three different stages during three point bend test should be explained well. For an example, a softening region in P1-P4 has some load oscillations. These need to be explained in detail to characterize the behavior of wood.
Response 1: The underlying physics behind the development of three different stages during 3-p-b test is re-written. The behaviours at point P1 to P4 of the load-displacement curve are explained in detail. Specific descriptions are added to characterize the oscillation of the curve. Please see the manuscript for details.
Point 2: Relevant literature is suggested for citations to make the paper more meaningful.
Response 2: Several relevant literatures are added into the manuscript, according to Reviewer 1’s comments, to make the paper more meaningful.
Point 3: Please see the comments in attached document.
Response 3: Reviewer 1’s comments in the attached document are all considered one by one. Corresponding modification are made and marked in the present manuscript. Please see the manuscript for details.
Point 4: The three point bend set up is used here. The same can be performed with four point bend, so explain the choice. Suggestions are given in the attached document.
Response 4: According to Lin et al. (2009), Morel et al. (2010) and Watanabe et al. (2011), both three point bend and four point bend test can be used to investigate Model I fracture. Because the huge difference of the Young’s modulus of wood in the transverse and longitudinal direction, the relative vertical displacement between the two sides of the wood fracture is very limited. And from Fig. 9 in the manuscript, it can be seen that the sliding displacement along the crack is much smaller than the transverse displacement. So, the effect due to the shearing in a 3-p-b test can be neglected in this test. Since the 3-p-b test is simple that 4-p-b test, it is used here.
[1] Lin Q, Fakhimi A, Haggerty M, Labuz JF. Initiation of tensile and mixed-mode fracture in sandstone. International Journal of Rock Mechanics and Mining Sciences, 2009, 46(3): 489–97.
[2] Morel S, Lespine C, Coureau J-L, Planas J, Dourado N. Bilinear softening parameters and equivalent LEFM R-curve in quasibrittle failure. International Journal of Solids and Structures, 2010, 47(6): 837-850.
[3] Watanabe K, Shida S, Ohta M. Evaluation of end-check propagation based on mode I fracture toughness of sugi (Crytomeria japonica). Journal of Wood Science, 2011, 57: 371-376.
Point 5:
The DIC results look convincing, but a few details on facet size and its deformation in subsequent loading stages should be addressed.
Response 5: The facet size is 70mm*30mm. The images at different loading points are added to the manuscript. Corresponding description are added to the paper.
Point 6: The figure 1 needs a detailed caption to show dimensions in mm. The figure 2 should point out the longitudinal direction fibers explicitly. In figure 11, explain the details of solid and dotted lines.
Response 6: A detailed caption is added to Fig. 1 to show dimensions in mm. The longitudinal direction is indicated in Fig.2. The legend in Fig.11 is modified to show the details of solid and dotted lines.
Point 7: Many other points are suggested in the document.
Response 7: Reviewer 1’s comments in the attached document are all considered one by one. Corresponding modification are made and marked in the present manuscript. Please see the manuscript for details.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The work presented investigates the fracture and fracture process zone (FPZ) of Douglas fir suing Digital Image Correlation (DIC). I have my reservations about the work and suggest a significant review. My two major concerns are:
I'm not sure what the aim of the work is. The concept of an FPZ in wood is well established, so is measuring FPZ using DIC. The work does not provide any additional insight into the fracture of Douglas fir. There is no new knowledge of the work presented. Please clarify this.
The concept of a notch length dependant FPZ, fracture toughness or an equivalent fracture toughness by adjusting the crack length is fundamentally flawed. Both are material properties and as such should not depend on a sample geometry and/or notch geometry.
Author Response
You have made a good effort in addressing the cohesive fracture mechanics and its characterization through Digital Image Correlation. The methodology and motivation behind The work presented investigates the fracture and fracture process zone (FPZ) of Douglas fir suing Digital Image Correlation (DIC). I have my reservations about the work and suggest a significant review. My two major concerns are:
Response 1: The underlying physics behind the development of three different stages during the 3-p-b test is re-written. The behaviors at point P1 to P4 of the load-displacement curve are explained in detail. Specific descriptions are added to characterize the oscillation of the curve. Please see the manuscript for details.
Point 1: I'm not sure what the aim of the work is. The concept of an FPZ in wood is well established, so is measuring FPZ using DIC. The work does not provide any additional insight into the fracture of Douglas fir. There is no new knowledge of the work presented. Please clarify this.
Response 1: The aim of this research work is to apply the DIC method to determine the critical crack length ac and the FPZ size. Then ac is used to calculate the failure fracture toughness by the linear elastic fracture mechanics, while FPZ is used to explain quasi-brittle fracture property by the crack and fiber bridging effect of wood. By comparing the ac results by DIC data and CMOD curve, it is that DIC is an easy and accurate method study the fracture property of wood. Also, on the base of the DIC method in this work, the mode-I fracture parameters in the transverse direction can be researched. Also, the boundary effect mode (BEM) fracture theory will be investigated in the future study, in which the FPZ size and ac length are very important.
Point 2: The concept of a notch length dependant FPZ, fracture toughness or an equivalent fracture toughness by adjusting the crack length is fundamentally flawed. Both are material properties and as such should not depend on a sample geometry and/or notch geometry.
Response 2: The fracture toughness, as well as the FPZ, are material properties, and they are determined by the microstructure material. When the specimen is big enough, for example, 1 meter high (more than 100 times of the fiber length), the fracture parameters are stable to measure. But for small specimens in the lab, the tested fracture parameters are influenced by the specimen geometry and size. The next work of us is to determine the material fracture parameter independent of small specimen size.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper topic is well in the scope of the journal. Due to its scientific content, I would advise the editor to accept the paper but only after some modifications have been made.
1. The authors should explain more in detail the optical methods used in the present study. For example the influence of the experimental noises on the measurements.The DIC principle should be added in the paper.2. A more detailed discussion of the fracture process parameters is necessary (crack length, crack orientation, …).3. How the crack opening displacement is measured. The authors should elaborate this aspect.4. In the literature review the recent work may be added. Please refer to the relevant references.
Author Response
Point 1:
The authors should explain more in detail the optical methods used in the present study. For example the influence of the experimental noises on the measurements.The DIC principle should be added in the paper.
Response 1: The Corresponding description about the basic principle of DIC is added to section 1. In terms of the experimental noise, the DIC technique is not sensitive of the experimental noise. However, light intensity and speckle quality are the main impact factors, which have been addressed in the manuscript. Limited by the space, some works of literature are also added referring to the fundamental of DIC technique. Please see the manuscript for the details.
Point 2:
A more detailed discussion of the fracture process parameters is necessary (crack length, crack orientation, …).
Response 2: Description of the physical properties of the crack is added in section 3.1. The speckle images of the failed specimen (area of 70mm30mm) at these loading points are shown in Fig. 7. Although 267 images are obtained in every second, it is difficult to see the difference between the first five images by human vision, before the crack propagates unsteadily.
Point 3:
How the crack opening displacement is measured. The authors should elaborate this aspect.
Response 3: In the test, one facet of the specimen is painted with speckles. On the other facet, two clip gauges with the measuring range of ±2mm were fixed by knife edges to measure the crack mouth opening displacement (CMOD) and crack tip opening displacement (CTOD). The corresponding description is listed in section 2.2.
Point 4:
In the literature review the recent work may be added. Please refer to the relevant references.
Response 4: Several recent works have been added to the literature reviews.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The revised document looks better. The changes are made as suggested, however a small comment on citations related to DIC. A couple of papers are cited, but since the work is much relevant to the suggested literature in older document, I continue to recommend specific papers outlined in the previous document. The material system is different but the loading stages show oscillations and DIC characterization is identical in this study hence the specific paper that has a similar work was recommended to cite.
Author Response
Thanks for your comments. The suggested references have been cited in the manuscript.
Reviewer 2 Report
The revised version and review letter have answered my concerns. However, the following edits are still needed:
Detailed assessment or error in the presented work. This included DIC noise error estimates and the computation of FPZ size estimates. For example, Fig 8 shows a fairly smeared FPZ. How did the subset size choice (and post-processing) affect this measurement. I expect it would.
Details of the DIC settings used, such as subset size, subset spacing, correlation technique used, etc. The authors are advised to consult other work that has been published on DIC.
Language editing.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
Thank you for addressing my comments. I would have liked to see further details in the DIC setup, but the information given should be informative enough.
