the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Mar 2025
Revealing firn structure at Dome A region in East Antarctica using cultural seismic noise
Abstract. Antarctica is mostly covered by snow, firn, and glacier ice, and the transformation from snow to firn and glacier ice influences energy transfer and material transport in polar regions. In this paper, we deployed three linear seismic arrays near Dome A in East Antarctica during China's 39th Antarctic scientific expedition and used seismic ambient noise to reconstruct the firn structure nearby. The result shows that the ambient noise mainly comes from the Kunlun Station and is related to human activities. We resolved empirical Green's function that contains abundant multi-modal surface waves from 3 to 35 Hz, and reconstructed the shallow S-wave velocity, density, and radial anisotropy structures by inverting them. The reliability of the structure was validated by the ice-core data, which demonstrates the effectiveness of using cultural seismic noise for the reconstruction of shallow structures in Antarctica. The result shows that the S-wave velocity increases rapidly with a negative radial anisotropy (SH wave travels slower than SV wave) in the top 28 m, which corresponds to the transformation from snow to firn. The firn layer shows a fairly strong positive radial anisotropy (SH wave travels faster than SV wave) between 40 m and 70 m in depth, which corresponds to the recrystallization of firn. The radial anisotropy vanishes to zero at around 84 m in depth, denoting the transformation from firn to glacier ice. Overall, the multi-parameter results clearly show the transformation from snow to ice, and the internal evolution of firn at Dome A region. Furthermore, we compared several existing S-wave velocity profiles of firn structures in West and East Antarctica and found a relatively shallower transformation depth from firn to ice in West Antarctica, which indicates a faster accumulation rate of snow in West than in East Antarctica.
- Preprint
(1073 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-1274', Yan Yang, 10 Apr 2025
- AC1: 'Reply on RC1', Yudi Pan, 09 Jun 2025
-
RC2: 'Comment on egusphere-2025-1274', Anonymous Referee #2, 12 May 2025
This paper presents a shear wave velocity model of the firn at Dome A, using established methods previously applied in other parts of the cryosphere. The results—particularly the identification of a relationship between firn compaction and radial anisotropy—are consistent with findings from studies conducted in other regions (e.g., Pearce et al., 2024; Diez et al., 2016). The value of this work lies in demonstrating that these results can be reproduced at Dome A, thereby contributing to our understanding of the consistency and geographic variability of firn anisotropy.
However, I find that the manuscript does not sufficiently establish the glaciological significance of these findings. The broader comparisons made between East and West Antarctica oversimplify the complex and highly localised nature of firn properties. Firn structure and compaction vary significantly across sites, and the paper's attempt to draw sweeping conclusions at the continental scale lacks the nuance required to be scientifically robust.
Moreover, the discussion does not adequately engage with existing literature on firn modelling and radial anisotropy. Key issues, such as intrinsic versus extrinsic causes of anisotropy, have been addressed more thoroughly in other studies yet are largely overlooked here. This gives the impression that the authors have not fully considered or integrated the existing body of research, weakening the paper’s scientific foundation. The authors should narrow their focus to the specific implications of their results for Dome A. Any comparisons should be limited and carefully contextualized. A more detailed engagement with existing firn literature and a clearer articulation of the glaciological relevance of their findings would significantly improve the manuscript.
As it stands, I believe the paper would benefit from Major Reviews. While the Dome A data, and the proof that anthropogenic noise can be used to produce a model of firn, is a valuable contribution to the glaciological community, the current framing, discussion, and treatment of the literature do present itself in an appropriate way.
 Below, I go into further comments about the paper:
Introduction:
The introduction would benefit from a clearer and more cohesive structure that better aligns with the specific focus of the study—imaging firn at Dome A. Currently, it opens with a broad discussion on Antarctica’s vulnerability to climate change but does not make a direct connection to the relevance of firn studies within that context. For a journal like The Cryosphere, such general framing may not be necessary and could be replaced with a more targeted explanation of why firn structure and compaction at Dome A are scientifically important.Additionally, the paragraph on seismic methods in Antarctica includes discussion of active-source techniques, which are not directly relevant to a study using ambient noise. A more focused discussion on ambient noise methods would strengthen the context. It would also be helpful to reference relevant studies using ambient noise to investigate firn structure, even if conducted outside Antarctica, since the methodology and findings are directly comparable to the present study.
Data & Methods:
This section could be improved by restructuring for clarity and cohesion. At present, it reads somewhat like a list of loosely connected points. For example, Line 95 begins with a description of data processing, followed immediately by a sentence about station A’s location—two pieces of information that could be better integrated into a more logically flowing narrative.Please clarify the process used to forward model the dispersion curves, specifically how different modes were identified and associated. Additionally, it is stated that density is derived from Vs, although VSH is used—this should be corrected or clarified.
Several statements such as "fairly well" appear throughout the manuscript; these should be quantified where possible to improve scientific rigor. For instance, when discussing the fit between the derived density profile and core data, numerical metrics or visual comparisons would be helpful.
On Line 165, the interpretation that faster VSV than VSH in the shallow firn is due to vertically aligned snow grains needs further explanation. Please elaborate on the physical mechanism behind this interpretation, and consider whether snow grain settling alone can produce this effect. It would also be beneficial to include a sensitivity analysis to demonstrate which depth ranges are constrained by the frequencies used in the inversion. Providing separate plots of VSH and VSV, in addition to their ratio, in the supplementary material would also help readers interpret the results more fully.
Discussion:
The discussion would benefit from a sharper focus on the specific insights gained from studying firn at Dome A. While comparing firn conditions across East and West Antarctica may seem appealing, such comparisons must be made cautiously due to the highly localized nature of firn compaction processes. The current manuscript draws broad conclusions about regional differences that are not strongly supported by the data and may oversimplify the complexity of firn behaviour.To strengthen the paper, the discussion should centre on the local significance of the Dome A firn profile, and what new insights are gained from applying ambient noise methods in this specific setting.
Citation: https://doi.org/10.5194/egusphere-2025-1274-RC2 - AC2: 'Reply on RC2', Yudi Pan, 09 Jun 2025
-
RC3: 'Comment on egusphere-2025-1274', Anonymous Referee #3, 13 May 2025
This paper presents a welcome iteration in ambient noise methods applied to imaging firn media, with a focus here on Dome A in East Antarctica. The authors use data collected from linear nodal arrays and leverage human camp noise to recover multi-modal surface waves to perform a near-surface velocity inversion. The paper is methodologically simple, in that the methods are well established, and the results are well presented.
As do the other reviewers, I have a few issues with the interpretation of results and the relative simplicity of the analysis in the global context of firn formation and structure:
1). Fig 3c: I agree with reviewer 1 on the velocity/density relationship you used here from Diez 2014, and the probable necessity for an updated form. The accumulation and strain environment of West Antarctica significantly differs from East Antarctica, and that could easily account for your disparities.
2). Firn anisotropy in East Antarctica very likely can be explained almost entirely from radial anisotropy given the minimal ice flow, but when comparing to West Antarctica, you should be cautious. For instance, flowing ice and firn has been shown to have multiple azimuthal anisotropy mechanisms related to fracturing and firn plasticity, and these can impact estimates of the magnitude of radial anisotropy estimates. Have a look at:
Chaput J, Aster R, Karplus M,  Nakata N, Gerstoft P, Bromirski PD, Nyblade A, Stephen RA, Wiens DA (2023). Near-surface seismic anisotropy in Antarctic glacial snow and ice revealed by high-frequency ambient noise. Journal of Glaciology 69(276), 773–789. https://doi.org/10.1017/jog.2022.98
Advected fractures and other embedded features can also affect the local velocity model in West Antarctica firn, so it's certainly worth mentioning the ultra-local side of firn profiles.Expanding on the discussion in terms of firn formation differences would be an improvement here, though given that edits will be largely constrained to discussion (with the exception of perhaps a test related to point 1 above), I think minor reviews are appropriate.
Â
Â
Â
Â
Â
Â
Citation: https://doi.org/10.5194/egusphere-2025-1274-RC3 - AC3: 'Reply on RC3', Yudi Pan, 09 Jun 2025
Status: closed
-
RC1: 'Comment on egusphere-2025-1274', Yan Yang, 10 Apr 2025
General Comments
This manuscript presents results using high-frequency cultural seismic noise from the Kunlun Station to image the shallow firn structure in the Dome A region of East Antarctica. The work resolves S-wave velocity and radial anisotropy down to ~100 m in the firn and validates the results with nearby ice-core data and results from other sites in Antarctica. The study offers an application of passive seismic methods in a remote polar region with limited prior coverage. The paper is well organized. The results are well illustrated. The implications for regional differences in firn compaction and accumulation rates are relevant. Overall, I believe this manuscript is well suited for publication in The Cryosphere after minor revisions.Specific Comments
1. In Figure 3c, the density model generally agrees with borehole studies, but some misfit is still present—specifically, overestimation below ~50 m and underestimation above ~30 m. A similar misfit pattern is reported in the cited study by Yang, Zhan et al. (2024), which motivated the development of an East Antarctica–specific empirical velocity–density relationship to better match observed firn density profiles. I see that you use Equation (2) from Diez et al. (2014), which is based on SH-wave velocity. Since you also resolve Vsv and your site is in the East Antarctic Plateau, I am curious how the results would compare if you applied the Yang, Zhan et al. (2024) relationship using your Vsv model. Additionally, Equation (2) assumes an ice density of 900 kg/m³—would using a more conventional value such as 920 kg/m³ change your results significantly? I understand the need for site-specific relations, but a brief comparison or discussion would strengthen this section.2. You have cited studies reporting radial anisotropy in firn at levels of 10–15% for several West Antarctic sites. I suggest also citing Schlegel et al. (2019), which examines radial anisotropy at the Kohnen site in East Antarctica. Additionally, I am curious about the robustness of the radial anisotropy inferred above 20 m depth. The cited study Pearce et al. (2024), using similar frequency bands, noted a lack of sensitivity to the top ~20 m in surface-wave inversions and therefore did not interpret their observed shallow radial anisotropy. Could you show the Rayleigh wave sensitivity kernel and comment on whether your inversion results are similarly limited in sensitivity in the uppermost firn?
3. Lines 1 and 2 are oriented in different azimuths, providing an excellent opportunity to investigate azimuthal anisotropy. Applying the same dispersion analysis workflow to Line 2 could help evaluate directional dependence of seismic velocities, which may relate to ice flow direction or crystal fabric. Is there a reason why dispersion analysis was not performed on Line 2—perhaps due to the absence of a short-spacing array needed for resolving higher modes? Regardless, I suggest including a discussion on the potential for azimuthal anisotropy and how it might be constrained by the existing dataset.
4. The observed difference in firn density profiles between East and West Antarctica is interpreted as a result of differences in snow accumulation rates. Temperature is another factor that significantly affects firn densification rates. Could you provide information or discussion on the differences in mean annual temperature between your site and the West Antarctic sites included in your comparison?Â
References:
Schlegel R, Diez A, Löwe H, et al. Comparison of elastic moduli from seismic diving-wave and ice-core microstructure analysis in Antarctic polar firn. Annals of Glaciology. 2019;60(79):220-230. doi:10.1017/aog.2019.10Â
Citation: https://doi.org/10.5194/egusphere-2025-1274-RC1 - AC1: 'Reply on RC1', Yudi Pan, 09 Jun 2025
-
RC2: 'Comment on egusphere-2025-1274', Anonymous Referee #2, 12 May 2025
This paper presents a shear wave velocity model of the firn at Dome A, using established methods previously applied in other parts of the cryosphere. The results—particularly the identification of a relationship between firn compaction and radial anisotropy—are consistent with findings from studies conducted in other regions (e.g., Pearce et al., 2024; Diez et al., 2016). The value of this work lies in demonstrating that these results can be reproduced at Dome A, thereby contributing to our understanding of the consistency and geographic variability of firn anisotropy.
However, I find that the manuscript does not sufficiently establish the glaciological significance of these findings. The broader comparisons made between East and West Antarctica oversimplify the complex and highly localised nature of firn properties. Firn structure and compaction vary significantly across sites, and the paper's attempt to draw sweeping conclusions at the continental scale lacks the nuance required to be scientifically robust.
Moreover, the discussion does not adequately engage with existing literature on firn modelling and radial anisotropy. Key issues, such as intrinsic versus extrinsic causes of anisotropy, have been addressed more thoroughly in other studies yet are largely overlooked here. This gives the impression that the authors have not fully considered or integrated the existing body of research, weakening the paper’s scientific foundation. The authors should narrow their focus to the specific implications of their results for Dome A. Any comparisons should be limited and carefully contextualized. A more detailed engagement with existing firn literature and a clearer articulation of the glaciological relevance of their findings would significantly improve the manuscript.
As it stands, I believe the paper would benefit from Major Reviews. While the Dome A data, and the proof that anthropogenic noise can be used to produce a model of firn, is a valuable contribution to the glaciological community, the current framing, discussion, and treatment of the literature do present itself in an appropriate way.
 Below, I go into further comments about the paper:
Introduction:
The introduction would benefit from a clearer and more cohesive structure that better aligns with the specific focus of the study—imaging firn at Dome A. Currently, it opens with a broad discussion on Antarctica’s vulnerability to climate change but does not make a direct connection to the relevance of firn studies within that context. For a journal like The Cryosphere, such general framing may not be necessary and could be replaced with a more targeted explanation of why firn structure and compaction at Dome A are scientifically important.Additionally, the paragraph on seismic methods in Antarctica includes discussion of active-source techniques, which are not directly relevant to a study using ambient noise. A more focused discussion on ambient noise methods would strengthen the context. It would also be helpful to reference relevant studies using ambient noise to investigate firn structure, even if conducted outside Antarctica, since the methodology and findings are directly comparable to the present study.
Data & Methods:
This section could be improved by restructuring for clarity and cohesion. At present, it reads somewhat like a list of loosely connected points. For example, Line 95 begins with a description of data processing, followed immediately by a sentence about station A’s location—two pieces of information that could be better integrated into a more logically flowing narrative.Please clarify the process used to forward model the dispersion curves, specifically how different modes were identified and associated. Additionally, it is stated that density is derived from Vs, although VSH is used—this should be corrected or clarified.
Several statements such as "fairly well" appear throughout the manuscript; these should be quantified where possible to improve scientific rigor. For instance, when discussing the fit between the derived density profile and core data, numerical metrics or visual comparisons would be helpful.
On Line 165, the interpretation that faster VSV than VSH in the shallow firn is due to vertically aligned snow grains needs further explanation. Please elaborate on the physical mechanism behind this interpretation, and consider whether snow grain settling alone can produce this effect. It would also be beneficial to include a sensitivity analysis to demonstrate which depth ranges are constrained by the frequencies used in the inversion. Providing separate plots of VSH and VSV, in addition to their ratio, in the supplementary material would also help readers interpret the results more fully.
Discussion:
The discussion would benefit from a sharper focus on the specific insights gained from studying firn at Dome A. While comparing firn conditions across East and West Antarctica may seem appealing, such comparisons must be made cautiously due to the highly localized nature of firn compaction processes. The current manuscript draws broad conclusions about regional differences that are not strongly supported by the data and may oversimplify the complexity of firn behaviour.To strengthen the paper, the discussion should centre on the local significance of the Dome A firn profile, and what new insights are gained from applying ambient noise methods in this specific setting.
Citation: https://doi.org/10.5194/egusphere-2025-1274-RC2 - AC2: 'Reply on RC2', Yudi Pan, 09 Jun 2025
-
RC3: 'Comment on egusphere-2025-1274', Anonymous Referee #3, 13 May 2025
This paper presents a welcome iteration in ambient noise methods applied to imaging firn media, with a focus here on Dome A in East Antarctica. The authors use data collected from linear nodal arrays and leverage human camp noise to recover multi-modal surface waves to perform a near-surface velocity inversion. The paper is methodologically simple, in that the methods are well established, and the results are well presented.
As do the other reviewers, I have a few issues with the interpretation of results and the relative simplicity of the analysis in the global context of firn formation and structure:
1). Fig 3c: I agree with reviewer 1 on the velocity/density relationship you used here from Diez 2014, and the probable necessity for an updated form. The accumulation and strain environment of West Antarctica significantly differs from East Antarctica, and that could easily account for your disparities.
2). Firn anisotropy in East Antarctica very likely can be explained almost entirely from radial anisotropy given the minimal ice flow, but when comparing to West Antarctica, you should be cautious. For instance, flowing ice and firn has been shown to have multiple azimuthal anisotropy mechanisms related to fracturing and firn plasticity, and these can impact estimates of the magnitude of radial anisotropy estimates. Have a look at:
Chaput J, Aster R, Karplus M,  Nakata N, Gerstoft P, Bromirski PD, Nyblade A, Stephen RA, Wiens DA (2023). Near-surface seismic anisotropy in Antarctic glacial snow and ice revealed by high-frequency ambient noise. Journal of Glaciology 69(276), 773–789. https://doi.org/10.1017/jog.2022.98
Advected fractures and other embedded features can also affect the local velocity model in West Antarctica firn, so it's certainly worth mentioning the ultra-local side of firn profiles.Expanding on the discussion in terms of firn formation differences would be an improvement here, though given that edits will be largely constrained to discussion (with the exception of perhaps a test related to point 1 above), I think minor reviews are appropriate.
Â
Â
Â
Â
Â
Â
Citation: https://doi.org/10.5194/egusphere-2025-1274-RC3 - AC3: 'Reply on RC3', Yudi Pan, 09 Jun 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 559 | 58 | 25 | 642 | 13 | 13 |
- HTML: 559
- PDF: 58
- XML: 25
- Total: 642
- BibTeX: 13
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General Comments
This manuscript presents results using high-frequency cultural seismic noise from the Kunlun Station to image the shallow firn structure in the Dome A region of East Antarctica. The work resolves S-wave velocity and radial anisotropy down to ~100 m in the firn and validates the results with nearby ice-core data and results from other sites in Antarctica. The study offers an application of passive seismic methods in a remote polar region with limited prior coverage. The paper is well organized. The results are well illustrated. The implications for regional differences in firn compaction and accumulation rates are relevant. Overall, I believe this manuscript is well suited for publication in The Cryosphere after minor revisions.
Specific Comments
1. In Figure 3c, the density model generally agrees with borehole studies, but some misfit is still present—specifically, overestimation below ~50 m and underestimation above ~30 m. A similar misfit pattern is reported in the cited study by Yang, Zhan et al. (2024), which motivated the development of an East Antarctica–specific empirical velocity–density relationship to better match observed firn density profiles. I see that you use Equation (2) from Diez et al. (2014), which is based on SH-wave velocity. Since you also resolve Vsv and your site is in the East Antarctic Plateau, I am curious how the results would compare if you applied the Yang, Zhan et al. (2024) relationship using your Vsv model. Additionally, Equation (2) assumes an ice density of 900 kg/m³—would using a more conventional value such as 920 kg/m³ change your results significantly? I understand the need for site-specific relations, but a brief comparison or discussion would strengthen this section.
2. You have cited studies reporting radial anisotropy in firn at levels of 10–15% for several West Antarctic sites. I suggest also citing Schlegel et al. (2019), which examines radial anisotropy at the Kohnen site in East Antarctica. Additionally, I am curious about the robustness of the radial anisotropy inferred above 20 m depth. The cited study Pearce et al. (2024), using similar frequency bands, noted a lack of sensitivity to the top ~20 m in surface-wave inversions and therefore did not interpret their observed shallow radial anisotropy. Could you show the Rayleigh wave sensitivity kernel and comment on whether your inversion results are similarly limited in sensitivity in the uppermost firn?
3. Lines 1 and 2 are oriented in different azimuths, providing an excellent opportunity to investigate azimuthal anisotropy. Applying the same dispersion analysis workflow to Line 2 could help evaluate directional dependence of seismic velocities, which may relate to ice flow direction or crystal fabric. Is there a reason why dispersion analysis was not performed on Line 2—perhaps due to the absence of a short-spacing array needed for resolving higher modes? Regardless, I suggest including a discussion on the potential for azimuthal anisotropy and how it might be constrained by the existing dataset.
4. The observed difference in firn density profiles between East and West Antarctica is interpreted as a result of differences in snow accumulation rates. Temperature is another factor that significantly affects firn densification rates. Could you provide information or discussion on the differences in mean annual temperature between your site and the West Antarctic sites included in your comparison?Â
References:
Schlegel R, Diez A, Löwe H, et al. Comparison of elastic moduli from seismic diving-wave and ice-core microstructure analysis in Antarctic polar firn. Annals of Glaciology. 2019;60(79):220-230. doi:10.1017/aog.2019.10
Â