Shallow Shear-Wave Velocity Beneath Jakarta, Indonesia Revealed by Body-Wave Polarization Analysis

Round 1

Reviewer 1 Report

The reviewer acknowledges the load of work performed in this article. However, the method, results and analysis are not very well presented. The reviewer completely disagrees with the conclusions drawn in the ‘Abstract’ and ‘Conclusions’ segments. First, any conclusion regarding Vp is unacceptable as the authors already admitted that their Vp structure is unresolved. There is not enough justification about the accuracy of the results obtained for the Vs. Therefore, based on the present status of the article, this reviewer is under the impression that the authors can only mention that they could obtain the Vs values that may seem comparable with those from the other studies within some order of uncertainty. It is highly encouraged though to resubmit the article with better explanation of the issues addresses by the reviewer.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 1 Comments

 

 

 

Point 1: Editorial Comments : L= Line

L 41-43. Break it into two sentences as - "earthquakes pose. It requires not only .... " L 56-57. Incoherent sentences. Need restructuring
L 65-66. “near-surface vs. and Vs” – Something is wrong here
L 83. “over the shallow most 100 meters” – Something is wrong here

L 94. “et al. [17],).” – Remove the comma
L 99. “ frequencies and but interfere” – Something is wrong here
L101. “amplitude of seismic which” – Seismic wave?
L 172. “VS,app..,” - VS,app?
L 204. “estimates of VSahs ” – is it VPahs ?
L 207. VSahs is repeated twice
L 214. What is λ in equation 10?
L 217. “The toolbox TTBox [23] can be utilized”. Has it been utilized in this study?
L 225. “does not always work well low SNR” - does not always work well forlow SNR? L 275. VSahs is repeated twice




Response 1: We would like to acknowledge all of the suggestions. It significantly improves the manuscript.

 

The updates based on the suggestions:

All minor technical problem has been revised. All figures have been updated to improve their quality. However, if the file sent to the reviewer is compressed, the figures may look poor. I will attach the original figures in a separate file then. Aim and objective have been emphasized. More references are added as suggested. We add more explanation We add the geological map of Jakarta.We reprocess all data started from the beginning. We only used P-waves polarization this time since only Vs can be resolved in this study. By not including S-waves, the Vs estimates are determined mainly from P-waves which signal-to-noise ratios are higher.

 

Point 2:Figure 3. Please mention the epicentral distance and azimuth of the event; the signal of the radial component does not seem to resemble with the one displayed on top of the figure 2(a). Why is it different? If it is because of the time window taken into consideration, then please mark the reference begin times of the signals with a straight vertical line on the plots of both Vertical and Radial components in Figure 2.

 

 

Response 2:Regarding figure 2, the vertical component showed only P-waves and the radial component showed only S-waves, so there are different compared to figure 3. After the revision, we only use P-waves for data processing, therefore we update the figures to prevent misunderstanding.

 

 

 

Point 3:Does this mean that finally, for a given earthquake, different time windows (ranging from one to several cycles) have been used for different stations? If so, this is very questionable. The authors need to shed light on this issue. In Figure 4(b), it is clear that the polarization obtained in first 2s of the

window is completely different than the later part. Yet, the authors have chosen to consider the results of the later part of the signal as the best polarization trend. This is in contrast with the hypothesis that the later part of the signal contains more mixed waves (the authors have also mentioned it). If most of the polarization parameters have been obtained from results that involve such high level of uncertainty around the chosen ‘best trend’, then it questions the efficiency of this method.

 

 

Response 3:Regarding the length of time windows; it is the biggest issue of this technique. Before the revision, we used different length between regional and teleseismic events so we can include as much of the respective waveform as possible. However, more complication arises in the area of sedimentary basin due to the possible contamination from arrivals of phases converted at the basement with different horizontal slowness. We perform synthetic tests and find that the converted waves occur after 2 seconds. A shorter window achieves a purely direct P-wave signal that is uncontaminated by phases of different wave type or incidence angle, but is not necessarily stable owing to its small number of data points. A longer window may contain phases of different wave type or incidence angle, which could bias the result. The synthetic tests (Appendix A) shows that window lengths greater than 1-2 seconds are unreliable due to the contamination of converted waves from the basin basement. Therefore, we analyzed each record to judge whether a mean estimate of P-wave incidence angle could be made, while removing outliers at the beginning and end of the window that might represent instability due to limited data or contamination by converted phases, respectively.

 

 

 

Point 4:Figure 5. Why not showing the results from the stations JK035 or JK093 (of Figures 2 and 3). Then we could better understand the entire sequence of the analysis. Why in Figure 5(a) and (b) the results of JKA12 and in Figure 5(c) the results of JK012 are shown? Do these inconsistencies (not showing results from one consistent station) mean that the authors are selectively showing the less questionable parts of the analyses?

 

 

Response 4:Regarding figure 5, the main reason we showed the results of JKA12 instead of JK035 or JK093 was to present more range of our observation. JKA12 was deployed in 2013, while JK035 and JK039 were deployed in 2018. We wanted to show both observations. In addition, when comparing to the HVSR study, JKA12 is more appropriate. However, we take the suggestion and remake the figures so the entire sequence of the analysis could be easily understood.

 

 

Point 5:However, the reviewer want to insist that the ‘Method’ segment needs to be better explained. The relation between the true and apparent incidence angles is not clear. It is not clear how the observed apparent incidence angles led to the inversion and calculation of velocities. The working principles of the toolbox TTBox and its application (it is not clear if it has been applied) require more explanation. The authors also jump into the result concerning the apparent velocities without showing any summary of the polarization parameters.

 

 

Response 5:Regarding the Method section, the complete derivation of the equation can be learned from References, while this paper does not focus on that. The derivation leads to Eqs. 1 – 3. We have arranged the sequence so the steps of the processing could be easily understood. Regarding TTBox; we require the ray parameters of P-waves for a certain station and earthquake. The programming of travel time toolbox (TTBox) [24] is utilized to computes seismic ray paths and travel time upon a 1-D spherical velocity model. The λ_ is eigenvalues as already defined in Eq. 5.

 

 

Point 6:Figure 6. Why showing the P-wave velocities if they are unresolved and if no further explanation of the results are provided? The authors mention that the stations lacking data are removed from the Figure 6. However when we compare Figure 1(b) with Figure 6, it is hard to see which stations have really been eliminated. It seems that the stations where the authors calculated higher shear wave velocities have fewer data (circled in red in the Figures below). The reviewer is wondering if the calculated high Vs is due to the selective choice of the windows and uncertainty associated with the polarization calculation. Hence, the reviewer thinks that the conclusions drawn in L 304-309 are not at all justified by the results.

 

 

Response 6:We agree that P-wave velocity is unresolved and do not need any further discussion. Therefore, we reprocess all data started from the beginning. We only use P-waves polarization this time. By not including S-waves, the Vs estimates are determined mainly from P-waves which signal-to-noise ratios are higher.

We use the stations which observed at least 4 events. It could be argued that the numbers of data at several stations are insufficient. However, getting comparable results with other studies (see below) make us decide that the numbers of data are sufficient.

 

 

 

Point 7:The reviewer is not sure why the authors have drawn a comparison with the Receiver Function method. The dataset required for Receiver function method is completely different – it is based on teleseismic P-waves and requires events recorded from a homogeneous distribution of azimuths – and used mainly to obtain the depth of Moho, which is not the case in this article.

 

 

Response 7:Regarding our mentioning about Receiver Function, we just want to acknowledge the reference (Svenningsen & Jacobsen, 2007) which also introduce this technique. It is true that receiver function is not the case in this paper, but the reference applies Eqs. 1 – 3 to some extent in their study. We argue that it is difficult to know what depth range these apparent half-space values correspond to. While we acknowledge the reference (Svenningsen & Jacobsen, 2007), we prefer to adopt the simpler approach of Park and Ishii (2018).

 

 

 

Point 8:The authors compare their results with those from Ridwan et al. only for one station. They have not mentioned how far the JK046 station is from the KMAL station. According to Figure 7, the Vs is 350 m/s at the KML station for a depth generally larger than 100 m. If Park and Ishi method is representative of the top 100 m depth and the calculated Vs is already 400 m/s, the results obtained by the authors are not convincingly comparable with the previous study. Moreover, as the Vs values are spatially variable, comparison with only one station is not very credible.

 

 

Response 8:We agree that one borehole of KMAL is insufficient to indicate an overall trend as we addressed in the discussion. However, that is the only available comparison which we have and we think their match in not a coincidence. It is unlikely possible to compare our results with the complete Vs map from Ridwan et al., that unfortunately, we do not have their digital results. If you see their paper, you can see a compatible result to some extent between our study.

 

 

 

 

Point 9:Figure 8. The Figure 8 is of very poor quality. In this kind of comparison, generally it is expected that the authors should provide a measure of differences among the maps. It is not very scientific to do this kind of visual comparison. It also seems that the calculated Vs map (Figure 8a) could partly resemble to either of Figure 8(b), (c) or (d), which correspond to three different depths (50, 100 and 200m). This is very confusing and any conclusion drawn from such comparison cannot be accepted as concrete scientific evidence.

 

 

Response 9:Regarding the misunderstanding about Vs representative of the top 100m, we would like to emphasize that the results represent average Vs over the top of certain depths (~100m). Therefore, the average of Vs over ~100m from KMAL benchmark is comparable to our result of Vsahs.

We agree with the suggestion regarding the comparison between our Vsahs estimates and average Vs layers obtained from HVSR studies. Therefore, we update the comparison using the graphs. We find the best match at depth of 150 m. We update our conclusion noting the mapping of Vsahs appears to be correlated to average Vs profile at depths of about top 150 meters.

Reviewer 2 Report

Article presents shallow shear wave velocity beneath Jakarta, Indonesia revealed by body-wave polarization analysis. Manuscript is well written with interesting topic and after some minor corrections I recommend it for publication.

Minor comments:

- In title (line 2) and line 25, should "Vs." has dot? Spelling mistake? Please check through entire manuscript

- I'm not a native English, but some sentences should be corrected - moderate English changes are required

- Please state main aim and objective in this study (lines 79-84) rather than give final conclusion and remarks - point this in Conclusion

- line 94, after [17] there is ",".

- lines 95-103 - any references regarding what you stated?

- Figure 1 is too small - use better quality

- in general equations are with small font - could you written them in better visual font? Maybe enlarge them? 

- again, through whole manuscript you have "Vs." - please check this.

- I would suggest putting a space before and after equations, e.g., eq. 6, between equations, and between text and equations

- what is TTBox toolbox - please state in text and then refer to reference

- Also, for some cases of method description (e.g., line 235-236), I would suggest to put references regarding what you write.

- Figure 2 - better quality is needed - enlarge also + larger font

- Figure 3, 4, 5 - the same as Figure 2

- For the Figure 6 - is there possibilty that your provide interpolation map as a base on colored circles - maybe just in a response to reviewer? I wonder how would look if you put together both interpolated map and colored circles?

- line 324 - strange font - please check?

- line 327 - please define SPAC and reference?

- Figure 7 - better quality, larger - could you present some HVSR spectral curve for typical Vs profiles? Any borehole data? Geological profile?

Figure 8 - larger - too small - this is the most important figure and I suggest authors to enlarge it

In general manuscript is very good. I put some suggestions and questions. They are not mandatory, but if possible if authors could gave answers and solutions to them, it would significantly improve some parts of manuscript. But also this can be included in future studies as authors pointed in Conclusion.

Author Response

We would like to acknowledge all of the suggestions. It significantly improves the manuscript.

 

The updates based on the suggestions:

All minor technical problem has been revised. All figures have been updated to improve their quality. However, if the file sent to the reviewer is compressed, the figures may look poor. I will attach the original figures in a separate file then. Aim and objective have been emphasized. More references are added as suggested. We add more explanation We add the geological map of Jakarta. We reprocess all data started from the beginning. We only used P-waves polarization this time since only Vs can be resolved in this study. By not including S-waves, the Vs estimates are determined mainly from P-waves which signal-to-noise ratios are higher.

 

Our replies according to the questions:

Regarding TTBox; we require the ray parameters of P-waves for a certain station and earthquake geometry. The programming of travel time toolbox (TTBox) [24]is utilized to computes seismic ray paths and travel time upon a 1-D spherical velocity model. We do not provide the interpolation map for Vs^ahsin purpose. We are afraid that the interpolation map may bias the results since the depth of Vs^ahsmap that we got are still open for discussion. In addition, we would like to emphasize the estimates at each station which represent average Vs over shallow depths, which is over the top 150 m in our comparison.

Round 2

Reviewer 2 Report

Authors answer to all my questions and revised manuscript properly.