Surge Mechanisms of Garmo Glacier: Integrating Multi-Source Data for Insights into Acceleration and Hydrological Control

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study applied the offset-tracking technique to a stack of Sentinel-1 images and derived its high temporal resolution velocity fields. Considering its very short-lived surge period, and other factors, it should be a hydrological controlled type of surge event, which is against the previous study (Wang Z, 2023). The study is well-designed and the results of data analysis support the conclusion. However, before publication, I still have some minor suggestions.

 

1, the 1^st paragraph in introduction is the instruction of the template, which should be removed when submitting. Possible it is a bug of the submitting system. Anyway, check the proof reading version (if there is one) when submitting.

 

2, Line 114, ‘A recent study examined the surge processes of five glaciers in western Pamir and 114 concluded that the surges were trigging by thermal mechanisms [33].’ Name these five glaciers here. In reality, I believe Gando is also a hydrological controlled type.

 

3, Line 187, the ‘Glacier flow velocity’ here is surface velocity, which combines the basal sliding and glacier internal deformation. I suggest to emphasize the glacier flow velocity as glacier SURFACE flow velocity. Please also check other place where may leads to misunderstanding.

 

4, The results part is a bit too short. The figure should appear just following the paragraph mentioning it. Eg. Figure 2 should be at Line 213. Besides, section 3.3 mentioned velocity decomposition, but the ice (internal) deformation and/or basal sliding map were not presented in the results part. Please add a figure to present them either in the result part or in the discussion part (1^st paragraph in 5.1, where the results of basal deformation rates are mentioned).

 

5, Again, figure 5 should be presented in Line 274 (after the paragraph first mentioned figure 5).

 

 

6, latitude tick should be more than one in figure 6. I am not sure if the south edge reaches 38.5 deg. If not, the latitude interval should be set to 20 minute. 

Author Response

Response to Reviewer #1,

 

We thank reviewer for the detailed reviews, and we made all suggested corrections. In this response, the reviewer’s comments are in black standard font. Our response is in standard blue font and the modifications to the manuscript are in blue bold font.

 

 

General Comments:

This study applied the offset-tracking technique to a stack of Sentinel-1 images and derived its high temporal resolution velocity fields. Considering its very short-lived surge period, and other factors, it should be a hydrological controlled type of surge event, which is against the previous study (Wang Z, 2023). The study is well-designed and the results of data analysis support the conclusion. However, before publication, I still have some minor suggestions.

 

1, the 1st paragraph in introduction is the instruction of the template, which should be removed when submitting. Possible it is a bug of the submitting system. Anyway, check the proof reading version (if there is one) when submitting.

Thank you for your review, and we deleted the first paragraph, i.e., the instruction of the template.

 

 

2, Line 114, ‘A recent study examined the surge processes of five glaciers in western Pamir and 114 concluded that the surges were trigging by thermal mechanisms [33].’ Name these five glaciers here. In reality, I believe Gando is also a hydrological controlled type.

Thank you for your review, and we added the names of these five glaciers here.

“A recent study examined the surge processes of five glaciers in western Pamir, including Sugran, Gando, Vanchdara, Shocalscogo and Garmo glaciers, and concluded that the surges were trigging by thermal mechanisms [33].”

 

 

3, Line 187, the ‘Glacier flow velocity’ here is surface velocity, which combines the basal sliding and glacier internal deformation. I suggest to emphasize the glacier flow velocity as glacier SURFACE flow velocity. Please also check other place where may leads to misunderstanding.

Thank you for your review, and we revised “glacier flow velocity” to “glacier surface velocity” and checked the full text.

L51-54: “In the active phase, a significant ice mass is rapidly transferred from the reservoir to the receiving zone, dramatically increasing surface flow velocity and redistributing glacier mass without changing the total mass [4-6].”

L55-56: “During this phase, glacier surface flow velocity can surge to 10–1000 times its standard order of magnitude.”

L189-190: “Glacier surface flow velocity is fundamentally governed by the combined effects of ice deformation and basal sliding.”

L238-239: “Figure 4. Temporal evolution of glacier surface flow velocity at Garmo Glacier from February 2022 to September 2022.”

 

4, The results part is a bit too short. The figure should appear just following the paragraph mentioning it. Eg. Figure 2 should be at Line 213. Besides, section 3.3 mentioned velocity decomposition, but the ice (internal) deformation and/or basal sliding map were not presented in the results part. Please add a figure to present them either in the result part or in the discussion part (1st paragraph in 5.1, where the results of basal deformation rates are mentioned).

Thank you for your review, and we adjusted the position of Figure 2 so that it immediately follows the paragraph where it is mentioned and checked the full text.

And we added a paragraph and the ice deformation and basal sliding map to describe the velocity components in the Result section.

“Based on the 2000 glacier thickness data[46] and surface elevation changes from 2000 to 2017, the average ice thickness in the ablation area of Garmo Glacier was 228 m in 2017. The TanDEM date in 2017 indicated that the surface slope of the ablation area was 4.3°. According to Glenn's law [44], the ice deformation velocity in February 2017 was 0.17 m d⁻¹. This accounted for approximately 90% of the observed surface velocity of 0.19 m d⁻¹ during the same period. In 2020, the average ice thickness and surface slope in the ablation area of Garmo Glacier were 230 m and 4.3°, respectively. Consequently, the ice deformation velocity in April 2020 increased to 0.19 m d⁻¹, accounting for approximately 80% of the observed surface velocity of 0.23 m d⁻¹ (Figure 5).”

Figure 5. The components of surface velocity in 2017 and 2020.

5, Again, figure 5 should be presented in Line 274 (after the paragraph first mentioned figure 5).

Thank you for your review, and we adjusted the position of Figure 5 so that it immediately follows the paragraph where it is mentioned.

 

 

6, latitude tick should be more than one in figure 6. I am not sure if the south edge reaches 38.5 deg. If not, the latitude interval should be set to 20 minute. 

Thank you for your review, and we added the latitude tick of Figure 6 to two.

Figure 7. The tendencies of temperature and precipitation around the Gramo Glacier from 2000 to 2023. (a) The tendency of temperature. (b) The tendency of summer temperature (from May to October). (c) The tendency of precipitation. (d) The tendency of winter precipitation (from November to April)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In present study, authors used various remote sensing data to investigate the surge of Garmo Glacier in the western Pamir. Along with this, authors also analyzed the influence of hydrological control and climate change on the surging mechanism of Garmo glacier.

 

Concept of paper is good and also written in well manner.

 

There some points are need to address.

 

(1) The length of the whole paper is too short, only 13 pages, and the content of the paper needs to be further added.

(2) Please delete the first paragraph of the introduction. This is a template for the MDPI journal paper.

(3) The D-InSAR in the keywords can be modified to Differential InSAR or differential phase. D-InSAR is usually used to describe algorithm that use differential phase to generate ground deformation with millimeter accuracy, however, this paper uses differential phase to generate ground surface elevation changes with submeter accuracy. The above two algorithms are completely different, and the writing of D-InSAR can cause confusion among readers.

(4) In Figure 2.(e), there is almost no change in glacier surface elevation change along the center line during 2014-2017 and 2017-2020. However, there are sharp fluctuations in glacier surface elevation changes during 2020-2022. Is this due to the use of different radar and optical data? The TSX/TDX data obtained in 2022 can be used for verification.

Author Response

Response to Reviewer #2,

 

We thank reviewer for the detailed reviews, and we made all suggested corrections. In this response, the reviewer’s comments are in black standard font. Our response is in standard blue font and the modifications to the manuscript are in blue bold font.

 

 

General Comments:

In present study, authors used various remote sensing data to investigate the surge of Garmo Glacier in the western Pamir. Along with this, authors also analyzed the influence of hydrological control and climate change on the surging mechanism of Garmo glacier.

Concept of paper is good and also written in well manner.

 

There some points are need to address.

 

(1) The length of the whole paper is too short, only 13 pages, and the content of the paper needs to be further added.

Thank you for your review, and we added the detailed description about the methods of glacier surface velocity and elevation change, and the result of velocity decomposition.

In the section of Data and Methods, the following content has been added:

“3.2 Glacier surface velocity and elevation change

Glacier surface velocities were estimated using normalized cross-correlation offset tracking from sentinel-1A data. We applied intensity offset tracking to a set of Sentinel SAR image pairs to generate offset fields, primarily using the GAMMA interferometry software. To address initial offset errors in the raw data, orbit correction was per-formed for each image pair using precise orbit parameters. Co-registration of the im-age pairs was achieved with the aid of the SRTM-C DEM. Subsequently, the offset tracking module in the GAMMA software was executed with a patch window of 256 × 64 pixels (range × azimuth). A signal-to-noise ratio (SNR) threshold was set during this process to filter out unreliable results, discarding pixels with an SNR below 4.0. The calculated offsets between the two SAR intensity images reflected the combined ef-fects of satellite orbit errors, glacial motion, terrain undulation, and ionospheric dis-turbances [33]. Satellite orbit errors were corrected during preprocessing using precise orbit files. Since our study area is located in mid-latitudes, the influence of ionospheric disturbances—more significant in polar regions—was considered negligible. Addition-ally, the use of the SRTM-C DEM in offset tracking accounted for terrain undulations, significantly enhancing accuracy. Finally, geocoding of the glacier was performed us-ing the SRTM-C DEM, and a 5 × 5 pixel median low-pass filter was applied to remove outliers, resulting in the final velocity map. We calculated the mean velocity and standard deviation for a non-glacier flat area, resulting in an average uncertainty of 0.03 m d⁻¹. As this uncertainty is significantly smaller than the velocities observed during the surge phase, we consider the surface velocity derived from our calculations to be reliable.

Glacier elevation changes during the periods of 2000-2014, 2014-2017 and 2017-2020 were acquired using differential synthetic aperture radar interferometry (Differential InSARDInSAR) from SRTM DEM and TSX/TDX images. Bistatic interferograms were generated from TSX/TDX images and the SRTM DEM using the GAMMA SAR software. The terrain phase was simulated using the SRTM DEM, and the residual phase was obtained by subtracting the flat and terrain phases from the interferograms [33]. This residual phase was used to produce a differential interferogram, where the differential phase represents topographic variations between the data acquisition times. Finally, glacier surface elevation changes were derived. It is important to note that the SRTM C-band (5.6 GHz) can penetrate snow and ice, which must be consid-ered when assessing glacier elevation changes. In contrast, the SRTM X-band (9.7 GHz) has a negligible penetration depth in snow and ice compared to the C-band. Thus, the difference between the two bands is typically used to estimate the penetration depth of C-band data. The average penetration depth of SRTM C-band in this region is ap-proximately 2 meters.

Glacier elevation changes for the period 2020-2022 were calculated by DEM dif-ferencing with TSX/TDX DEM and GF-7 DEM [42, 43]. The relative horizontal and ver-tical shifts between the TSX/TDX and GF-7 DEMs were corrected based on the cosi-nusoidal relationship between elevation difference, slope, and aspect in non-glacierized areas [43]. During co-registration, terrain in non-glacierized areas with elevation differences between the 5% and 95% quantiles and slopes between 5° and 80° was used to compute co-registration parameters [42]. This adjustment reduced the mean elevation difference in non-glacierized areas from 5.27 m to 0.82 m, confirming that the co-registration and DEM differencing were suitable for estimating glacier ele-vation changes. The uncertainty of these changes was derived using the normalized median absolute deviation (NMAD) over non-glacierized areas [33], resulting in an uncertainty range for 2000-2022 elevation changes from 0.6 m to 1.4 m.”

 

In the section of Results, the following content has been added:

“Based on the 2000 glacier thickness data[46] and surface elevation changes from 2000 to 2017, the average ice thickness in the ablation area of Garmo Glacier was 228 m in 2017. The TanDEM date in 2017 indicated that the surface slope of the ablation area was 4.3°. According to Glenn's law [44], the ice deformation velocity in February 2017 was 0.17 m d⁻¹. This accounted for approximately 90% of the observed surface velocity of 0.19 m d⁻¹ during the same period. In 2020, the average ice thickness and surface slope in the ablation area of Garmo Glacier were 230 m and 4.3°, respectively. Consequently, the ice deformation velocity in April 2020 increased to 0.19 m d⁻¹, accounting for approximately 80% of the observed surface velocity of 0.23 m d⁻¹ (Figure 5).”

Figure 5. The components of surface velocity in 2017 and 2020.

 

 

(2) Please delete the first paragraph of the introduction. This is a template for the MDPI journal paper.

Thank you for your review, and we deleted the first paragraph of the introduction.

 

 

(3) The D-InSAR in the keywords can be modified to Differential InSAR or differential phase. D-InSAR is usually used to describe algorithm that use differential phase to generate ground deformation with millimeter accuracy, however, this paper uses differential phase to generate ground surface elevation changes with submeter accuracy. The above two algorithms are completely different, and the writing of D-InSAR can cause confusion among readers.

Thank you for your review, and we revised the keyword of “D-InSAR” to “Differential InSAR”, and checked the full text.

L33-34: “Keywords: glacier surge; glacier velocity; surface elevation; offset-tracking; Differential InSAR; the Western Pamir”

L191-193: “Glacier elevation changes during the periods of 2000-2014, 2014-2017 and 2017-2020 were acquired using differential synthetic aperture radar interferometry (Differential InSAR) from SRTM DEM and TSX/TDX images.”

 

(4) In Figure 2.(e), there is almost no change in glacier surface elevation change along the center line during 2014-2017 and 2017-2020. However, there are sharp fluctuations in glacier surface elevation changes during 2020-2022. Is this due to the use of different radar and optical data? The TSX/TDX data obtained in 2022 can be used for verification.

Thank you for your review. We have carefully re-examined the glacier surface elevation changes and confirm that the observed changes are accurate, not a result of inconsistencies between different data sources. While we did not obtain the 2022 TSX/TDX data, the normalized median absolute deviation (NMAD) over non-glacierized areas in the DH maps indicates an uncertainty range for the 2000–2022 elevation changes of 0.6 m to 1.4 m.

In addition, based on the 2000 glacier thickness data and surface elevation changes during 2000-2017 and 2017-2020, we calculated the average ice thickness and surface slope in the ablation area of Garmo Glacier. These values were used to estimate ice deformation for 2017 and 2020, which aligns reasonably well with the observed surface velocity.

Therefore, we believe that the elevation changes across different periods are reliable, even with minimal changes observed during 2014–2017 and 2017–2020, followed by a sharp fluctuation during 2020–2022.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Authors have incorporated all the suggestion given by me. 

Now , this paper is suitable for publication.