Source Model for Sabancaya Volcano Constrained by DInSAR and GNSS Surface Deformation Observation

Round 1

Reviewer 1 Report

I carefully read the manuscript remotesensing-796989 untitled “Source model for Sabancaya volcano constrained by InSAR and GNSS surface deformation observation”. This is a well-written manuscript highlighting an important research topic worth of further investigation and can be useful among the remote sensing community.

I find the manuscript well structure, interesting to read. The abstract is explicative, keywords are appropriate and introduction is well written. The research method and data analysis techniques follow well-established routines and complying with state-of-the-art methodologies. I felt the figures, result sections need further improvement for better clarification.

I recommend for publication after addressing the following issues:

• Authors mentioned ‘InSAR’ term on the title but use both ‘DInSAR’ and ‘InSAR’ terms throughout the manuscript. This will create confusion for readers without having interferometry background. So suggestion is stick to one term either.
• I noticed some inconsistency in measurement units and typo errors in the manuscript that need to be corrected. Use either cm/yr or mm/yr for the velocity measurement throughout the manuscript. Need correction on lines 181-182 where R1 point is missing and replaced by R3, line 235 “1,5 km” should be “1.5 km”.
• In Figure 1(b) information missing for satellite direction, path and frame number are not highlighted. Incidence and heading angle are not mentioned in the caption.
• Section 2.1 mentioned Sentinel-1 data for processing but didn’t describe about the sensor mode, number of bursts used and how many sub-swaths considered for the processing. No clear explanation on why few scenes available between 2014 and 2016 for descending datasets, whether it is due to co-registration issue or data unavailability need to be clarified.
• Line 97: authors didn’t describe how many interferogram used for ascending and descending datasets separately. This is not clear if 486 is for total interferograms or for each pass. Authors would explain why large temporal baseline of 1374 days used for interferogram generation using Sentinel data. Better to provide SAR acquisition data list for both ascending and descending data including perpendicular baseline information.
• Line 99: no discussion on resolution of SRTM DEM used in the processing.
• Lines 100-107: authors mentioned model based GACOS approach applied for phase correction. Could authors explain if that sufficient to mitigate orbital correction, coastline effect, residual topography and atmospheric effect after phase unwrapping.
• Figure 2 (a & b): points (R1, R2 and R3) are not clearly visible. Need to increase font size. Authors would specify if the Figure 2 (a & b) are representing vertical velocity or along line of sight. Better to highlight on the colorbar or in caption which help easy understanding for the readers.
• Lines 203-210: making confusion with respect to Figure 3 and Table-2 that explaining components considered for the comparison of LOS velocity of ascending DInSAR results. How about comparing directly vertical and horizontal component of GNSS with respect to DInSAR components? In that case DInSAR displacement components can be derived combining ascending and descending together.
• On Section 3.1 authors mentioned spherical geometry as a best-fit with the DnSAR deformation based on Table-3 parameters. Authors didn’t describe much about the uncertainty measure for volume change rate and other parameters for other geometries.
• In Table-3 authors didn’t include sill geometry for GNSS data during 2014-2016 and spheroid during 2016-2018, any explanation for that? Why uncertainty measure only for sphere geometry?

Author Response

Dear reviewer,

first, we want to thank you for helping us in improving this work with your constructive comments.

Please see the attachment for our reply to your comments

Best regards, Maurizio Battaglia

Author Response File: Author Response.docx

Reviewer 2 Report

The article demonstrate new possibilities of using microwaves for volcanic research. The authors developed original methods of modeling, interpretation, and processed a huge array of data obtained by InSAR and GNSS devices in the period from 2014 to 2019. The results of the work next possible a better understanding of the tectonic processes occurring in the volcanic system Sabancaya. The article shows the possibility of obtaining data on the location of deformation sources, the velocity and shape of ground deformation, on the volume of magma movement. As a recommendation, it would be useful to give a more detailed comparative analysis of the errors in the determination of the surface deformation characteristics by the InSAR and GNSS devices. The article is of undoubted interest and will be useful for specialists in the field of geophysics and remote sensing.

Author Response

Dear reviewer,  we thank you for helping us in improving this work with you constructive comments.

Best regards, Maurizio Battaglia

Reply to Reviewer

[…] As a recommendation, it would be useful to give a more detailed comparative analysis of the errors in the determination of the surface deformation characteristics by the InSAR and GNSS devices. The article is of undoubted interest and will be useful for specialists in the field of geophysics and remote sensing. […]

Following the Reviewer recommendation, we added the following two paragraphs

[line 110 – 115]: Because of the contribution from the atmosphere, delay anomalies, spatial averaging used to down sample the InSAR data, and the structure of the noise in the InSAR data estimate, it is difficult to specify in a rigorous way the uncertainty in displacement measurements made with InSAR. On the basis of our own experience and consultations with colleagues who specialize in InSAR, we assign an uncertainty to the InSAR displacement measurements reported here of ±10 mm.

[line 136 – 140] The uncertainties of the GPS velocities were derived by scaling the formal error by the square root of the residual chi-square per degrees of freedom of the solution. We also apply additional corrections considering monument instability (random walk), instrumental error (white noise), and flicker noise [18].

Reviewer 3 Report

This paper presented source inversion results for the Sabancaya volcano from DInSar and GNSS data.  The results verify the presence of a source 12-15 km below the ground as speculated by other papers.  My main concern with this paper was that there were few technical details, and instead a high-level description of the methods used was given.  Some acronyms were also not defined, e.g., DInSar or differential interferometric SAR.  Also many of the conclusions were speculative, and it would be better if more evidence would be given other than opinion of the authors.  However, the results are quite significant and worth publishing, once some of these other issues have been corrected.

Author Response

Dear Reviewer

First, we want to thank you for helping us in improving this work with your constructive comments.

Best regards, Maurizio Battaglia

Reply to Reviewer (3)

[…] My main concern with this paper

(1) there were few technical details, and instead a high-level description of the methods used was given.

We tried our best to provide an exhaustive but concise description of the methods employed. We changed several paragraphs of the manuscript following this recommendation. Many of the technical details of the software tools employed to process and model the data are well known in the existing literature. We decided to refer the reader to existing literature for technical details and give more emphasis to the original results of our work.

(2) Some acronyms were also not defined, e.g., DInSar or differential interferometric SAR.

We now define the following acronyms

[line 56]: DInSAR (Differential Interferometry of Synthetic Aperture Radar) 

[line 83] GNSS (Global Navigation Satellite System)

[line 94] Small Baseline Subsets (SBAS)

[line 108] Zenith Time Delay (ZTD)

[line 119] the Volcanologic Observatory of INGEMMET (OVI)

[line 132] International GNSS Service (IGS).

[line 133] South America (SOAM) reference frame

[line 169] Line-of-Sight (LOS)

(3) Also, many of the conclusions were speculative, and it would be better if more evidence would be given other than opinion of the authors.

We agree with the reviewer. Following his/her recommendation, we deleted from the text any conclusion that sounded speculative. For example, we deleted several speculative paragraphs (the list below is not exhaustive)

“The deformation observed by [9], attributed by the authors to a non-magmatic source, indicates that the analyzed period covers the present unrest from since the beginning. “

“Finally, explosive eruptions are triggered by due a more differentiated residual fraction of magma.”  

 “These deep magmatic source could possibly have fed the other ASVC volcanoes as well. “

  and we also edited other paragraphs. For example,

"The best-fit source, located between 12 to 15 km depth beneath Hualca Hualca, reproduces the stress field and volume change of a magma reservoir that we believe fed Sabancaya’s past and present eruptions."

"The best-fit source, located between 12 to 15 km depth beneath Hualca Hualca, reproduces the stress field and volume change of a magma reservoir that might have fed Sabancaya’s eruptions since the mid-90s"

Reviewer 4 Report

This is an interesting and well-presented study. Please see the attached file for minor remarks.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer

First, we want to thank you for helping us in improving this work with your constructive comments.

Best regards, Maurizio Battaglia

[line 40] [OVI-INGEMMET]

deleted

[line 91-92] State the software you used to process the SAR/SBAS data

Done. 

"They were processed using our own implementation of the DInSAR Small Baseline Subsets (SBAS) time-series approach [12], which allows a proper spatial and temporal characterization of the deformation patterns occurring within the studied area."

Round 2

Reviewer 1 Report

I appreciate author's comprehend response to my comments. I recommend for publication with minor spell check required.

Author Response

First, we would like to thank the reviewer for his/her help in improving our manuscript. 

Reply to reviewer

Formatting of symbols was inconsistent, for example ~ 7 km N of Sabancaya, versus  ±10 mm and (0 - 0.1) versus 12–14 km on page 8.  I reviewed the manuscript carefully to eliminate this type of inconsistency.

Last sentence section 1.  Replaced have with has as the subject, source is singular.

Section 2.2, paragraph 2.  Should be per degree of freedom (not degrees). Last sentence of first paragraph in 2.2 Modelling.  Replaced χ²v here with χ²_v.

Section 2.1 heading is 12 font, section 2.2 was 10 font.  There were 2 sections headed as 2.2.  Modelling is now2.3.  Modelling is now always spelled as modelling in the manuscript.  Down-sampled is now always spelled down-sampled.

Page 5 first line.  Replaced “around” with “about”.

In the text, I replaced the format Fig 2.a, 2.b with Figure 2a and 2c.

Table 2.  November is now capitalized.

Page 8. (0 - 0.1) I eliminated the extra space.

Page 8. 26 x10⁶ m³/yr and 43 x10⁶ m³/yr here versus 22 x10⁶ m³/yr and 46 x10⁶ m³/yr in the abstract and 22–51 x10⁶ m³/yr, in the discussion. These relate to different time spans and modeled sources (Table 3).  We now always present the correct estimate of 26 x10⁶ m³/yr and 43 x10⁶ m³/yr in the text.

Table 3. Edited so that DInSar fits in one line.

In the discussion, figures are now referred Figure X, whereas earlier they were (Fig. X).

Discussion paragraph 2.  I deleted “regional”.

Figure 5. km on the right vertical axis was partially cut off. 

Conclusions.  The last phrase of the first sentence was awkward.  Modified.

Author contributions.. the name of R. Miranda Cruz is now the same as in the  author list.

Conflict of interest.  Interest was misspelled.

Resolved several inconsistencies in the reference format and abbreviations and spelling of journals.

Thanks again for your help and time