the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Jul 2025
Extreme Glacier Melt in the Central Tibetan Plateau during the Summer of 2022: Detection and Mechanisms
Abstract. Extreme glacier melt events accelerate mass loss, increase glacier instability, and temporarily mitigate downstream drought. However, the glacier energy-mass balance in the Geladandong region—the headwater of both the Yangtze River (China’s longest river) and Siling Co (Tibet’s largest lake)—and its connection to the unprecedented 2022 summer melt remain insufficiently quantified. To address this gap, we integrated in situ mass balance observations (October 2019–July 2022) and glacio- meteorological data from a 5,700 m monitoring site (October 2020–July 2022) with a surface energy-mass balance model, and applied a novel dual-threshold framework (based on mean and standard deviation) to identify extreme mass loss/melt events in short-term glacier mass balance records. Our novel methodology identified 2021–2022 as a period of extreme mass loss and melt intensity, with melt during the 2022 summer heatwave reaching unprecedented extremes. Over 52 days, the heatwave generated 1,135 mm w.e. of melt—accounting for 65.2 % of the total 2021–2022 melt, equivalent to 1.8 and 2.3 times the melt recorded in 2019–2020 and 2020–2021, respectively. This extreme melt was driven by energy balance anomalies over glacier surface, including reduced albedo, increased incoming longwave radiation, and enhanced sensible heat fluxes, with these processes strongly linked to persistent high temperatures and diminished precipitation. Such anomalies in the energy balance were driven by large-scale atmospheric circulation anomalies—specifically, the concurrent intensification and westward expansion of the Western Pacific Subtropical High (WPSH) and the eastward extension of the South Asian High (SAH). This study establishes a novel framework for identifying extreme mass loss events and substantially advances understanding of glacier mass balance responses to extreme weather and climate events.
- Preprint
(2671 KB) - Metadata XML
-
Supplement
(468 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2025-2579', Baijun Shang, 09 Oct 2025
-
AC1: 'Reply on CC1', Zhu. Meilin, 14 Oct 2025
Thank you for your comment.
Citation: https://doi.org/10.5194/egusphere-2025-2579-AC1
-
AC1: 'Reply on CC1', Zhu. Meilin, 14 Oct 2025
-
RC1: 'Comment on egusphere-2025-2579', Anonymous Referee #1, 27 Oct 2025
This manuscript by Zhu et al. investigates a critical topic. However, in my opinion, there are fundamental issues with model validation, methodological justification, and presentation. If the author can address these issues, I would recommend publication with major revisions.
Weaknesses in Model Validation, Methodology, and Presentation
1. Inadequate Temporal Validation of Core Model Outputs (Figs. 2, 3, 4 & 5):
Figs. 2 & 3 (Scatter Plots): Presenting albedo, radiation, and mass balance as scatter plots obscures the model's performance in simulating the temporal evolution of these variables. Time-series plots are essential to reveal whether the model correctly captures the timing and duration of key events, such as the rapid albedo drop during heatwaves. A good correlation (R-value) can mask significant temporal lags or failures to replicate specific events.
Fig. 3 (Mass Balance): The scatter plots validate only the annual net result. To truly validate the melt process, the authors must present observed vs. modeled cumulative mass balance over time for each year. This would indicate whether the model accurately simulates the rate and timing of ablation. Furthermore, disaggregating into observed vs. modeled ablation and accumulation separately is critical, as a correct net balance can result from compensating errors in melt and snowfall.
Unclear Representativeness: The manuscript does not clearly state how the point measurements from a limited number of stakes (Fig 1d) are representative of the glacier-wide means presented in Figs. 4 and 5.
Redundancy (Figs. 4 & 5): Fig. 4 shows only modeled monthly fluxes. If sub-annual observed data exist, this figure should be updated to include the observed cumulative mass balance overlaid on the modeled one, making it the primary validation figure. This would render the interannual bar chart in Fig. 5 redundant, but it would also provide valuable insights into the model's temporal performance.
2. Contradictory Evidence for Heatwave Identification (Fig. 7):
The central premise of the study is the impact of "extreme heatwaves." However, Fig. 7b shows that air temperature during the identified heatwave periods (grey shading) does not appear anomalously high compared to other times in the same melt season. Some peaks outside the grey areas appear just as high or higher. This directly undermines the classification of these periods as thermally extreme events at the glacier surface and calls into question the applicability of the standardized heatwave index (S_HI) derived from a distant valley station (Tuotuhe). The authors should provide a rigorous justification for why these periods are classified as heatwaves despite the apparent lack of a strong temperature signal at the AWS.
3. Unsupported Extrapolation from Point to Glacier-Wide Data:
The study relies on a single Automatic Weather Station (AWS) at 5,700 m a.s.l. (Fig. 1). Yet, the energy and mass balance results are presented as "glacier-wide" (e.g., Figs. 4, 8).
The method for spatially interpolating meteorological data across the entire glacier's elevation range (5,400–6,104 m a.s.l.) from this single point needs to be sufficiently justified. Using a constant lapse rate and assuming wind speed and relative humidity are uniform across the glacier is a major simplification that likely introduces significant errors, especially during extreme and stable atmospheric conditions. The manuscript should explicitly discuss the uncertainties introduced by this extrapolation and provide evidence (e.g., from other studies or model sensitivity tests) that the chosen method is valid for deriving glacier-wide energy balance.
4. Unjustified Application of the Framework to Other Glaciers (Fig. 6):
The application of the dual-threshold method to other glaciers is presented without a clear rationale. If intended as validation, it is circular, as it applies the method to other short records without an independent benchmark. If intended to show prevalence, it dilutes the focus from the detailed process-based analysis of Sangqu Glacier. The purpose needs to be clearly stated and justified.
Major Revisions Required:
1) Fundamental Validation: Replace scatter plots in Figs. 2 and 3 with time-series comparisons. Crucially, include time series of observed vs. modeled cumulative mass balance to validate the ablation process.
2) Justify Heatwave Classification: Reconcile the apparent discrepancy in Fig. 7. Provide a compelling explanation for why the grey-shaded periods are classified as extreme heatwaves based on the in-situ AWS data, or re-evaluate the analysis periods.
3) Address Spatial Representation: Justify the extrapolation from a single AWS to a glacier-wide scale. Discuss the associated uncertainties and limitations explicitly. A sensitivity analysis of the model to the spatial interpolation scheme would significantly strengthen the study.
4) Streamline Narrative and Clarify Rationale:
Update Fig. 4 to include observed mass balance time series.
Remove or modify redundant figures (e.g., Fig. 5).
Provide a clear and convincing rationale for the analysis presented in Fig. 6.
Conclusion
In my opinion, the manuscript requires major revisions. The current validation is insufficient to prove the model accurately simulates ice melt dynamics. The identification of the key driver, the heatwaves, is not convincingly supported by the presented in-situ temperature data, and the extrapolation of point measurements to the entire glacier is a major, unquantified source of uncertainty. Until these fundamental issues are addressed, the analysis of energy balance mechanisms remains speculative and is built on unverified premises.
Citation: https://doi.org/10.5194/egusphere-2025-2579-RC1 -
RC2: 'Comment on egusphere-2025-2579', Anonymous Referee #2, 23 Dec 2025
The manuscript ‘Extreme Galcier Melt in the Central Tibetan Plateau during the Summer of 2022: Detection and Mechanisms’ by Zhu and co-authors presents a detailed analysis of the energy-balance of Sangqu Glacier using the EBFM model. The authors further apply a statistical approach to determine whether or not the 2022 summer was extreme in terms of glacier melt, and discuss the underlying atmospheric mechanisms leading to the high recorded melt rates.
This is a good study presenting a detailed process-oriented approach to better understand the links between glacier and atmosphere. The model used was applied with care and extensively validated against in situ measurements. Despite this, I still had a number of points to raise that I feel would need to be addressed before this manuscript can be considered for publication.
General comments:
Statistical analysis with very short time series: I found it hard to follow the description of the new framework used to identify these extreme events from time-series of just a few years. This methodological section goes straight into technicalities, without explaining the general principle, making it difficult to understand what the authors are actually after. Furthermore, given the extremely short focus period, I have serious doubts about the value of such an approach. Along these lines, I fail to understand why the authors followed what seems to be a purely statistical approach without leveraging the energy-balance model that they validated at this site and could be expanded in time.
Parallel with extreme glacier mass losses in other regions of the world: This study is very much focused on the Tibetan Plateau, and I find it a bit disappointing that it does not draw more parallels with recent glacier extreme events that have been reported in other regions of the world, such as Canada, Switzerland, the Andes or Svalbard. Some mechanisms described here are also described for these other regions and it would be interesting to insist on how these events relate or differ.
Differences in mass balance measurement periods: The mass balance records are reported to end in July in 2022, compared to October in the previous years, which bodes the question of how comparable these measurements are, especially considering that 2022 is pointed out as the extreme year of the time series.
Calibration and validation of the energy-balance model over the same time period: if I understand correctly the model was calibrated (for the precipitation gradient) and validated against the same data. This makes me confused as to why there seems to be a bias between the simulated and measured point mass balances in Fig. 3a? And if this statement is correct, this is a limitation that would need to be raised in the discussion.
Limited details on the heat transfer in the ice: my expectation is that this glacier is cold-ice , which would have potentially a strong influence on the energy budget. However there is very limited description of how the heat transfer into the ice was calculated. What did you take for the heat transfer coefficients? Was a sensitivity analysis performed? Was there any validation data available?
Line-by-line comments:
L54: This feels a bit like an oversell. While the study itself is interesting, I didn’t find anything absolutely groundbreaking there.
L61: I’m not convinced that wildfires fit in the category of ‘weather and climate events’. References would be welcome here.
L62: Could you be a bit more specific here? Give some numbers?
L62, l95: Should be ‘the Tibetan Plateau’
L101-103: See general comment, this does not sound realistic.
L110: remind here the period to which this anomaly applies.
L153: It would be useful to indicate the mean value of annual precipitation at this site. I expect this to be a very dry site, and it would be useful to have these numbers in mind to put the energy budget into context.
L157: Are there any available geodetic mass balance estimates for this glacier?
L161: Photo should be singular
Figure 1: Panel d is hard to make sense of. It would make sense to show a DEM hillshade or satellite image of the glacier, or at the very least use a different colorscale.
L163: Please indicate the location of these precipitation gauges.
L171: greater -> more
L174: More details would be welcome here, as well as a picture appended to Fig. 1. Was this precipitation gauge heated? Shielded?
L210: ‘where’ does not fit here
L241-244: I don’t get this. Why do these time series need to be reconstructed if they already exist?
L247: What is a ‘three-fold cross validation’? And a 10-fold?
L251-252: Did you consider looking into seasonal bias corrections?
L270: How was this lapse-rate determines. It has been shown that the lapse rate on glaciers can strongly vary from the environmental lapse rate.
L270: ‘the parameter’ -> ‘a calibrated parameter’
L271: Where does the albedo come from? Can the albedo at the AWS location be considered representative of the whole glacier.
L281: What emissivity value was used?
L281: What was used for the surface roughness and why?
L286-288: I fail to see the relevance here, and in this entire paragraph – everything should be calculated relative to the AWS elevation?
Figure 2: What does the color scale correspond to?
L326: How were the glacier-wide values obtained? Since your model does not seem to account for any type of snow redistribution process, I wonder if a validation against a glacier-wide mass balance extrapolated from stake data is useful here.
L368-369: missing references here.
L388: How many are there?
L432: I don’t think I’ve seen any mention of the spatial resolution at which the model was run? Were snow redistribution processes (from avalanches and wind) accounted for in any way?
Figure 4: It would be helpful to plot the net energy balance on top of this bar plot. Also, were there any altitudinal variations?
L522: why were these glaciers chosen specifically?
L541-542: references missing.
L738: this whole paragraph sounds like a repetition of what was stated above.
L755: I am missing a discussion of the limitations from the energy-balance model and its application.
Citation: https://doi.org/10.5194/egusphere-2025-2579-RC2
Model code and software
The Dual-Threshold Framework Fei Zhu https://github.com/zhuzhufeifei/detect_extreme_events
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,198 | 146 | 34 | 1,378 | 64 | 40 | 48 |
- HTML: 1,198
- PDF: 146
- XML: 34
- Total: 1,378
- Supplement: 64
- BibTeX: 40
- EndNote: 48
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study is a standout contribution to glaciology and climate science. It skillfully addresses a critical knowledge gap in the hydrologically vital Geladandong region by integrating in-situ observations, high-altitude data, and energy-mass balance modeling. Its novel dual-threshold framework offers a replicable tool for identifying extreme glacier melt events, while its clear quantification of the 2022 summer extreme melt (1,135 mm w.e. over 52 days, 65.2% of annual melt) and linkages to large-scale atmospheric circulation (WPSH, SAH) deepen understanding of glacier-climate interactions. It advances both methodology and practical insights for climate projections and water resource management.