Study on the Moisture Content Diagnosis Method of Living Trees Based on WASN and CTWGAN-GP-L
Round 1
Reviewer 1 Report
Comments related to the article by Yin Wu, Nengfei Yang and Yanyi Liu “Study on the moisture content diagnosis method of living wood based on WASN and CTWGAN_GP-L” submitted in Forests:
Measurement of wood moisture content is of fundamental importance to forestry and wood sciences. The limitation of commonly used wood moisture measurement methods is limited accuracy and the inability to monitoring of moisture changes in real time. Obtaining accurate moisture content measurements requires the use of the oven-dry method, i.e. the destructive method. It is also impossible to take into account the variability of wood properties, which reduces the reliability of the measurement. The reviewed paper verified the possibility of using the low-power consumption, wireless acoustic sensor network system for diagnosing moisture of wood tissue of living trees. The authors showed that the developed real-time long-term measurement system is characterized by a very high MC diagnostic accuracy (above 95%). The scientific and practical value of the manuscript is high. I am convinced that the presented paper delivers useful knowledge in the area of application of wireless measures system to diagnostics of MC. The manuscript was written clearly and logically. The presentation of the results on the figures and the data compilation in the tables is clear. However, in the opinion of the reviewer, the chapter on “Materials and Methods” is too extensive. I propose to consider transferring the details of the algorithms used to the “Appendix” chapter. I also believe that in relation to the methods of measuring wood moisture, the term “reference oven-dry method” should be used instead of the “traditional drying method” (Page 1). In my opinion, it is more appropriate to use the term "living trees" instead of "living wood" (in the title of the manuscript). In the caption of Figure 11, the term "synthetic data" is used. I suggest the term "generated data".
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors used a wireless acoustic emission sensor-based system to measure the water content of living trees, extended the dataset using GAN, and used the LightGBM algorithm to identify the water content of the data features. In order for readers to better understand the methods and intentions of the paper, I hope that the authors will discuss the following aspects or add them at the corresponding locations in the paper.
(1) Figure 4 is the collection curve, not the interface
(2) The range of moisture content of living trees is extensive, and the moisture content measured by the pinning method is part of the local point measurement, which should be measured at multiple points to select the appropriate average moisture content.
(3) the problem of repeatability, whether the data are stable after multiple measurements.
(4) whether there is consistency in the average value of the same method measured at different locations and different time of the tree trunk.
(5) whether the variability of test data is due to the different internal structures of different tree species, such as loose pore and ring pore wood, which have different forms of internal moisture organization, and the variability of moisture content with seasonal changes, and whether this variability of tree species may have some influence on the acoustic emission signal transmission.
(6) The article needs to be well organized, highlighting the relevant key points, such as the focus on specific experimental acquisition, data analysis, and data validation to illustrate the comparative nature of this method.
(7) The conclusion section is inadequate for the analysis of the data and comparison with other methods to develop the discussion and analysis. It needs further analysis and explanation, as mentioned before for the repeatability of a measurement instrument, consistency, and adaptability to different tree species. The analysis of the advanced Ness and inadequacy with other existing testing methods are developed.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Thank you for the author's reply and revision, I have no other comments.
