the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Jun 2025
Analysis of Long-Term Dynamic Changes of Subglacial Lakes in the Recovery Ice Stream, Antarctica
Abstract. The dynamic activity of subglacial lakes plays a crucial role in modulating glacial dynamic processes and influencing the mass balance of the Antarctic ice sheet. The Recovery Ice Stream (RIS), projected to experience significant mass loss in East Antarctica during coming centuries, requires continued investigation of its subglacial hydrological activity. Prior to 2012, nine active subglacial lakes and their dynamics in the RIS were identified through combined analysis of ICESat satellite altimetry and IceBridge airborne altimetry data. However, recent subglacial hydrological dynamics require further investigation. In this study, we investigate recent activity of nine subglacial lakes in the inventory of the RIS region using newly available ICESat-2 altimetry data, updating their outlines based on observed activity during the ICESat-2 mission period. Furthermore, we identify 14 newly active subglacial lakes designated as RecN1-RecN14. By synthesizing multi-source altimetry data (ICESat, IceBridge, ICESat-2), we establish a 21-year (2003–2023) elevation change time series for subglacial lakes in the RIS region. Through crossover analysis, we precisely identify elevation changes within subglacial lakes and elucidate spatiotemporal patterns across distinct lake sectors, revealing significant disparities between lake centers and their peripheries with a maximum difference of 4 m (e.g., lake Rec6-1). Finally, quantitative analysis of volume changes within primary drainage networks, based on constructed subglacial hydrological networks, confirms effective hydraulic connectivity between subglacial lake clusters in the RIS region. This study elucidates the hydrological dynamics of subglacial lakes in the RIS region, providing scientific basis for a deeper understanding of the evolutionary processes of Antarctic subglacial hydrological systems.
- Preprint
(37562 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-1632', Shuai Yan, 10 Jul 2025
-
RC2: 'Comment on egusphere-2025-1632', Whyjay Zheng, 28 Aug 2025
Thank you for the work on the subglacial lakes in the Recovery Ice Stream (RIS), Antarctica (https://doi.org/10.5194/egusphere-2025-1632). This work updates the existing inventory of subglacial lakes in the RIS by remapping the lake outlines and identifying more active lakes within the RIS, making a valuable contribution to the community. The writing is clear and easy to understand. I only have one major comment and a few minor ones listed below.
Major comment:
The Results and Discussion section is primarily focused on data description, and very little, if any, discussion is present. I would love to read more interpretations from the authors about their data for justification and curiosity. Maybe consider addressing some of the following questions:
- What is the core reason that some lake outlines need to be updated? Is it because the existing records are inaccurate, or the lake extent naturally changed (due to different water volume, perhaps) over time? L432 might already provide a short answer from the authors, but I would really love to see a more complete elaboration.
- Why did the draining-refilling pattern change significantly after January 2022 (the end of Period III according to Figure 9)? Why did the water stop migrating from the upstream lakes to the downstream lakes?
- Is there any drainage-recharge pattern associated with the geographical distribution of the lakes? In other words, do the lakes have similar patterns within a cluster (lower trough, upper trough, and upstream area) and distinct patterns across clusters? If yes, what does it imply?
Minor comments:
- The authors use the terms “drainage-filling cycle/pattern” and “storage-drainage cycle/pattern” interchangeably. They may need to be unified for clarity. Additionally, I think it is awkward to say “storage event.” A rechange event may better describe this dynamic behavior.
- How did you determine the gap location and width between Rec1-1 and Rec1-2? I ask this because the boundary does not always seem clear to me. For example, data from RGT 0505 (Figure 4c) seems to indicate a wider Rec1-1 section than currently depicted.
- Please ensure the submitted work adheres to the TC data policy (https://www.the-cryosphere.net/policies/data_policy.html). The data availability section does not mention the data generated from the work (e.g., new and updated subglacial lake outlines and elevation change maps for each lake). Please deposit “data that correspond to journal articles in reliable (public) data repositories, assigning digital object identifiers, and properly citing data sets as individual contributions.” Additionally, I suggest including visualization(s) (in the main text or as supporting material) that are similar to Figure 3 but encompass all reported lakes. This will enable readers to more easily explore and assess the results.
- L44: “DEM differencing” is more common than “differencing DEM.”
- L138: What does this slope constraint actually do?
- Section 3.2.1: I am not sure whether I understand this correctly, but it appears that the reference DEM is created by interpolating only ICESat-2 data. What is the nominal date of this DEM? If it represents elevations from the ICESat-2 era, why does every time series of elevation changes start from 0 in 2004 (Figures 5-7)?
- Equation 1: How did you determine the values of p? I know they are “based on the minimum distance from the elevation measurement points to the lake outline,” but I suppose you did a certain conversion or normalization for p, correct?
- Figure 3: Are all panels at the same scale?
- Figure 4 b-d: might need a clearer way to tell readers that only the cyan line (elevation change range) follows the axis to the right. I realized this after multiple reads.
- Figures 5-7: What are the exact locations of this time series over the subglacial lake area? I ask this because you mentioned that the elevation change is weaker at the edge than at the center of the lake, and in Figure 8 you showed many time series sampled from different locations within one lake.
- L267: I do not think we have sufficient evidence to say there is a “continuous rise” due to the huge data gap.
- L275: Again, it is a bit confusing to say “seamless connection” here since the IceBridge data have lengthy temporal gaps.
- Figure 8: What do the thick dashed lines show?
- L382: Elevations cannot “reach their trough.” I suggest finding an alternative to characterize the time series. Maybe something like “ice surface elevations decrease until August 2020…?” Also, “maximum elevation change of -2.8 m” seems awkward since the value is negative.
- L395-398: This should be put in the method section. To estimate the volume change, you multiply the lake area by what?
- L437-438: “Most newly discovered lakes gradually show active trends during the Operation IceBridge and ICESat-2 periods…” I am not sure if it is legitimate to say this; after all, the altimetry method requires the lake to be active to be detected.
- L442: Is it always true over time? (see the second bullet point in the major comment)
Citation: https://doi.org/10.5194/egusphere-2025-1632-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 477 | 74 | 15 | 566 | 5 | 21 |
- HTML: 477
- PDF: 74
- XML: 15
- Total: 566
- BibTeX: 5
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study uses satellite-observed changes in ice surface elevation to infer subglacial hydraulic activity in the RIS region. It addresses a topic of considerable scientific interest and importance and provides a detailed analysis based on a substantial dataset. The methods are thoroughly described, and the results are presented with substantial detail. However, the manuscript lacks a discussion of the broader implications of the findings. Please see my detailed comments below.
Major comments:
1. I appreciate the substantial dataset the authors have analyzed and presented in this manuscript. Given the richness of the data, it is especially important to present the findings in a way that is clear and accessible to readers. I recommend highlighting the key takeaway message at the beginning of each subsection or paragraph in the Results section to help guide the reader through the analysis.
2. Section 4 is titled “Results and Discussions”; however, in my view, it primarily presents the study’s findings with minimal discussion or interpretation of their broader implications. I commend the authors for the significant amount of work invested in this analysis and for the thorough presentation of the results. That said, for a scientific manuscript, it is important to move beyond simply describing findings to interpreting their meaning and situating them within the broader scientific context. For example, what are the implications of discovering a connected subglacial lakes network in a region with significant ice mass loss potential? Could this connectivity influence ice dynamics or impact sea-level rise projections? Additionally, could the methodology presented in this study be applied to other regions of Antarctica? If so, what challenges might arise in extending this approach to different glaciological or observational contexts? These are merely examples intended to encourage the authors to consider the broader scientific context of their work—there is no need to focus on these specific points.
3. The authors estimate lake volume change by multiplying ice surface elevation change by the estimated lake area. While this method can provide a first-order approximation of subglacial water movement and is acceptable for comparing drainage and filling timing, it is known to be an oversimplification (e.g., Stubblefield et al., 2021, https://doi.org/10.1029/2021GL094658). More accurate volume estimates require consideration of additional physical processes and ice–water interactions. The limitations of this approach should be explicitly acknowledged in the Discussion section, along with a justification of why it is appropriate for the specific goals and scope of this study.
Minor comments:
Line #108: “To maximize measurement accuracy” could be more precisely phrased as “To minimize uncertainty” or “To reduce noise in the data,” depending on the intended meaning.
Line #124: “with 500 m resolution” would be clearer as “with 500 m horizontal resolution” to specify the dimension being referred to. The same for line #126.
Line #128: Why not obtain bedrock elevation directly from datasets such as Bedmap3 or BedMachine?
Lines #135 and #151: The basis for selecting the “0.5 m threshold” (line #135) and the “10 km” value (line #151) is unclear.
Line #148: “elevation changes of subglacial lakes” would be more accurately phrased as “elevation changes of the ice surface above subglacial lakes”.
Line #253-254: There appears to be a repeated or redundant statement in this section.
Line #391: more elaboration is needed for “the first-order and second-order stream network”.