Distribution of Holocene Marine Mud and Its Relation to Damage from the 1923 Earthquake Disaster in the Tokyo Metropolitan Area, Japan

Round 1

Reviewer 1 Report

The manuscript "Distribution of Holocene marine mud and its relation to damage from the 1923 earthquake disaster in the Tokyo Metropolitan area, Japan" is an interesting contribution to Geosciences. It concerns the relations between the lithofacies organization of the post-LGM incised valley fills beneath the wide Tokyo metropolitan area and the geometrical and geothecnical (Vs) properties of soils. This permits a convincing comparison between the damage distribution of wooden buildings, after the strong 1923 earthquake, and the extension and thickness of a particular lithofacies of the valley infill, the N1 mud. 

I think that the article is very well written, as it is very rich in subsoil data (boreholes, Nspt data) that have been well processed by means of a 3D voxel model in order to reconstruct with good detail the subsoil of Tokyo. The resonance period of wooden houses is well compared with that of the natural ground in correspondence of the major thickness of N1.  This research can be a good contribution to assess seismic hazard at local scale in recent alluvial and coastal plain settings, being the article an interesting and well based case of study. 

Only two changes in the text:

1) at line 64 change the word "proposes" into "purposes";

2) at lines 180-182 change the phrase into : "The N-value of the estuary (delta) system decreases (increases) upward, which suggests the fining-upward (coarsenig-upward) trend of the lithological successions"   

Author Response

1. Reply to Reviewer 1’s comments

Highlighted in yellow.

1.1. English correction.
Corrected (line 79).

1.2. English correction.
Corrected (line 197-199).

Reviewer 2 Report

Distribution of Holocene marine mud and its relation to damage from the 1923 earthquake disaster in the Tokyo Metropolitan area, Japan

By Tanabe et al.

General Considerations

The manuscript deals with the near-surface geology and its relation to damage caused by the 1923 Taisho Kanto earthquake in the Tokyo metropolitan area.

In my opinion, the manuscript is quite well written, the methodology is sound, the discussion of the data is appropriate, and the conclusions are supported by the data.

However, I find that some aspects need to be improved.

Specific comments

1. Introduction. The authors will discuss the 1923 Taisho Kanto earthquake in more depth, indicating the effects it caused by the available statistical data.  Furthermore, it is suitable to frame the research in the wider literature concerning the correlation between near-surface geology and earthquake damage. In this regard, some suggestions are supplied below;
2. Materials and Methods, line 145. To explain the interpolation method used;
3. Relation between the thickness ... How important was the resonance phenomenon? Is it possible to estimate the percentage of wooden houses that potentially experienced stronger shakings and vibrations due to resonance?

Bibliography suggested (not to be considered exhaustive)

Guidoboni, E.; Mariotti, D.; Giammarinaro, M.S.; Rovelli, A. Identification of amplified damage zones in Palermo, Sicily (Italy), during the earthquakes of the last three centuries. Bull. Seismol. Soc. Am. 2003, 93, 1649–1669.

Maresca, R.; Nardone, L.; Gizzi, F.T.; Potenza, M.R. Ambient noise HVSR measurements in the Avellino historical centre and surrounding area (southern Italy). Correlation with surface geology and damage caused by the 1980 Irpinia-Basilicata earthquake. Measurement 2018, 130, 211–222. https://0-doi-org.brum.beds.ac.uk/10.1016/j.measurement.2018.08.015

Author Response

2. Reply to Reviewer 2’s comments

Highlighted in blue.

2.1. Add some effects of the 1923 Taisho Kanto earthquake.
Added (line 53-54).

2.2. Refer wider literature.
Added (line 70-72).

2.3. Explain the interpolation method of Ishihara et al. (2013).
Added (line 158-159).

2.4. How important was the resonance phenomenon?
Almost 100% of the wooden houses were collapsed by the resonance. This explanation added in line 346-347.

Reviewer 3 Report

This manuscript presents a comprehensive characterization of the distribution of soft (mud) sediments from the Tokyo and Nakagawa lowlands, areas that were impacted by the 1923 earthquake.  It was a pleasure to read the manuscript, which is well-crafted, and presents the most comprehensive investigation of soft sediments that I have ever seen.  My compliments to the authors.  I recommend the manuscript for publication once the concerns are addressed.   I will list some minor concerns below: they should all be straightforward to address. My one substantive concern is that Figure 8 does not present a convincing argument for a significant correlation between damage and sediment thickness. It would be helpful to show bin-averaged values in Figure 8, to highlight systematic trends in the data.  Perhaps an alternative approach to visualization would be useful, for example if damage ratio could be superimposed on sediment thickness maps?  Or perhaps more statistical tests would be useful to demonstrate a significant correlation?   An alternative approach might be to recraft the paper, including the title and abstract, to focus on the characterization of sediment thickness, and mention the correlation with damage as a fruitful area for future research?  I have two other concerns related to the discussion of damage: 1) I expected to find some discussion about the quality of construction.  Is it safe to assume that the distribution shown in Figure 7 is not controlled in part by differing style or quality of construction in different areas?  And 2) The “quarter-wavelength” approximation is reasonable for first-order consideration, but the authors should note that the 1-d model is simplified.  Also see my last minor comment below.

Minor comments are listed below, by line number:

13. “cyclical” is not the best word here because it implies periodicity. Maybe “struck frequently”?

19. replace “following the” with “caused by”

19 “Our results show”

29. “exceptional” is not the best word – other world cities are at risk due to earthquake rates and local geological setting. Maybe rewrite: “Seismic hazard is high in the Tokyo area due to its proximity to the active plate boundary and its underlying geological structure (REFERENCE)”

67. 13,000! Nice! It would be helpful to say a few words about how the examination in this study extends the earlier work by the cited references.

Figure 1.  At least one of the two panels needs to include latitude and longitude.  Panel (a) also needs a scale.

Figures 3 and 4. These figures are very nicely crafted, but the labels suggest that it should be rotated 90 degrees clockwise.

References are good in general but not sufficient for the ground motions/damage section. An extensive literature review is not needed, but more discussion and at least a few references are needed to appropriately introduce and motivate the discussion of site response.

I waive anonymity. 

Author Response

3. Reply to Reviewer 3’s comments

Highlighted in green.

3.1. Add bin-averaged values in Figure 8.
The thickness value in Figure 8 is a 5-m-interval averaged value. This explanation added in the caption of Figure 8 (line 588).

3.2. Overlay the maps.
Done. New figures (Figures 8 and 9) are made.

3.3 Explain the quality of construction.
Added (line 316).

3.4. Explain that the “quarter-wavelength” approximation is a simplified 1-d model.
Added (line 344-346).

3.5. English correction.
Corrected (line 19).

3.6. English correction.
Corrected (line 26).

3.7. English correction.
Corrected (line 26).

3.8. English correction.
Corrected (line 40-41).

3.9. Explain how the previous studies are extended.
Explained in line 82-85.

3.10. Figure 1. Add latitude, longitude, and scale.
Done.

3.11. Figures 3 and 4. They should be rotated.
Done.

3.12. Add wider references in the discussion.
Done (line 365-367).

Round 2

Reviewer 3 Report

• The topic of site response needs to be introduced properly, with appropriate references and some acknowledgement of issues beyond the 1/4-wavelength approximation.
• Some information needs to be provided about the quality of construction in 1923.  In virtually every area, at any time, the quality of construction varies between richer and poorer areas.
• There is no compelling evidence to support the conclusion that sediment thickness correlates with damage.  None of the last 3 figures are convincing -- there needs to be compelling visual evidence for a spatial correlation or statistical analysis.