Detection of Earthquake-Induced Landslides during the 2018 Kumamoto Earthquake Using Multitemporal Airborne Lidar Data

Round 1

Reviewer 1 Report

Dear authors,

Here to you my comments:

 - lines 53-54, is there a specific reason why 'These data have not been effectively utilized to make a damage assessment.'? I would suggest to either specify this reason or remove the sentence.

 - Figure 1a, i suggest to add a location of the area within the Japan map in a small inset. Additionally, I could not find (as far as i remember) any reference to the type of motion associated to the 2018 Kumamoto Earthquake and I would suggest to only draw the section trace of the seismogenic fault in Fig.1a.

- Section 2, probably better 'Landslides associated to the 2016 Kumamoto Earthquake in the study area' as title.

- line 90, 'by the author', despite I do not like this specification at least mention who of the three authors?

- lines 145-146, the displacements values of the main shocks are much higher than the ones given in the introduction, please double-check.

- line 169, please use Figure 5b instead of 'b' only.

- lines 193-194, those DTM thresholding values do not look much smaller than the ones used for the DSM, please double-check.

Author Response

- lines 53-54, is there a specific reason why 'These data have not been effectively utilized to make a damage assessment.'? I would suggest to either specify this reason or remove the sentence.

Answer: The sentence has been removed accordingly.

 

 - Figure 1a, i suggest to add a location of the area within the Japan map in a small inset. Additionally, I could not find (as far as i remember) any reference to the type of motion associated to the 2018 Kumamoto Earthquake and I would suggest to only draw the section trace of the seismogenic fault in Fig.1a.

Answer: A Japan map has been added accordingly. The background of Figure 1a is an elevation map. We added a legend of color and the expansion in the caption as “The background is an elevation map in the rainbow color.”

 

- Section 2, probably better 'Landslides associated to the 2016 Kumamoto Earthquake in the study area' as title.

Answer: It has been revised accordingly.

 

- line 90, 'by the author', despite I do not like this specification at least mention who of the three authors?

Answer: We added the name in the caption of Figure and revised the sentence in Line 91.

 

- lines 145-146, the displacements values of the main shocks are much higher than the ones given in the introduction, please double-check.

Answer: Since the GNSS stations mentioned in the introduction were away from the fault, the observed displacements were smaller than the study area. We added the reference [17] to support the displacements in this area.

 

- line 169, please use Figure 5b instead of 'b' only.

Answer: It has been revised accordingly.

 

- lines 193-194, those DTM thresholding values do not look much smaller than the ones used for the DSM, please double-check.

Answer: The threshold values for the DSMs were from -3 m to -6 m, whereas those for the DTMs were from -2 m to -4 m. Limited by the low spatial resolution and vertical accuracy of the pre-event DTM, we set the threshold in a 1-m interval and the smallest absolute threshold value larger than 1 m. Thus, the threshold for the DTMs were set as those values.

Reviewer 2 Report

Thank you for this very interesting and well presented contribution. I have only minor comments on the presentation of the content. Also the pictures and tables are clear and informative. 

The readers would appreciate a summary of all used abbreviations.

Also a timeline with the events and the dates of the measurement campaigns would improve the manuscript. 

Author Response

Thank you for your kind review.

-The readers would appreciate a summary of all used abbreviations.

Answer: We added a list in Appendix accordingly.

 

-Also a timeline with the events and the dates of the measurement campaigns would improve the manuscript. 

Answer: We added a timeline for the events and the Lidar dates as Figure 3.

Reviewer 3 Report

Well done work

Author Response

Thank you for your kind review.

Reviewer 4 Report

The Authors presented the results of two sets of Lidar data acquired before and after the 2016 Kumamoto Earthquake. The performance of the two different sets of Lidar data are also evaluated. Despite the work is an updating of previous works published on Proceedings of SPIE and Proceedings of 39th Asian Conference on Remote Sensing, Kuala Lumpur, the results are interesting and useful for scientific community. In order to make the manuscript readable to readers the following revisions are needed.
1. Introduction section: I suggest to improve the references about the use of remote sensing technologies to understand the extent and the degree of damages after natural disasters. Following some suggested references:
- Menderesa et al. (2015). Automatic Detection of Damaged Buildings after Earthquake Hazard by Using Remote Sensing and Information Technologies. Procedia Earth and Planetary Science. Volume 15, Pages 257-262.
- Yida Fan et al. (2017). Quantifying Disaster Physical Damage Using Remote Sensing Data—A Technical Work Flow and Case Study of the 2014 Ludian Earthquake in China. International Journal of Disaster Risk Science. Volume 8, Issue 4, pp 471–488.

2. Thresholding method section: As stated by the Authors "...the landslides occurred on slopes covered by high trees". In these conditions it is recommended the use of extensive ground control points to assess the accuracy of DTMs, in order to account for biases caused by vegetation cover. This aspect has to be discussed in the text body. Following, a recent paper which discusses the DTM accuracy on forest environments.
- Simpson et al. (2017). Assessment of Errors Caused by Forest Vegetation Structure in Airborne LiDAR-Derived DTMs. Remote Sens. 2017, 9, 1101; doi:10.3390/rs9111101.

Author Response

Thank you for your kind review.

Introduction section: I suggest to improve the references about the use of remote sensing technologies to understand the extent and the degree of damages after natural disasters. Following some suggested references:

- Menderesa et al. (2015). Automatic Detection of Damaged Buildings after Earthquake Hazard by Using Remote Sensing and Information Technologies. Procedia Earth and Planetary Science. Volume 15, Pages 257-262.

- Yida Fan et al. (2017). Quantifying Disaster Physical Damage Using Remote Sensing Data—A Technical Work Flow and Case Study of the 2014 Ludian Earthquake in China. International Journal of Disaster Risk Science. Volume 8, Issue 4, pp 471–488.

Answer: According to the comment, we added four more references including the applications in different earthquakes and typhoon disasters.

 

Thresholding method section: As stated by the Authors "...the landslides occurred on slopes covered by high trees". In these conditions it is recommended the use of extensive ground control points to assess the accuracy of DTMs, in order to account for biases caused by vegetation cover. This aspect has to be discussed in the text body. Following, a recent paper which discusses the DTM accuracy on forest environments.

- Simpson et al. (2017). Assessment of Errors Caused by Forest Vegetation Structure in Airborne LiDAR-Derived DTMs. Remote Sens. 2017, 9, 1101; doi:10.3390/rs9111101.

Answer: According to the document of Geospatial Information Authority of Japan (https://fgd.gsi.go.jp/otherdata/spec/DEMgaiyo.pdf in Japanese), the horizontal error of the DTMs is less than 1 m and the vertical error is less than 0.3m. Even there is no ground point in 5-m grids, the vertical error is less than 2 m. Thus, we did not use ground control points. According to the comment, we added a short discussion as “According to the study of Simpson et al. [46], the forest conditions affect the accuracy of DTMs. The DTMs used in this study were created from the leaf-on Lidar data (both acquired in May), the vertical error would be approximately 1 m. However, the average of the DTM differences between 2006 and 2016 was 5 cm with the standard deviation 0.6 m, showing good vertical accuracy.”

Round 2

Reviewer 4 Report

The Authors answered to all the comments.