Review 10.5194/tc-2021-303: A regionally resolved inventory of High Mountain Asia surge-type glaciers, derived from a multi-factor remote sensing approach

Guillet et al.

The authors present a comprehensive inventory of surge-type glaciers for HMA from 2000-2018, using readily accessible glacier velocity, mass loss and morphology datasets. They find a larger number of surging glaciers for all regions than previously identified in regional inventories. With knowledge of surge duration and velocity they show that power laws explain the occurrence of certain durations/velocities.

The paper is very clearly structured, the science is sound, well explained and clearly followed through and the creation of this inventory and the questions that are addressed through it are of great interest to the community. It definitely fits the scope of the Journal very well. I apologize for a very short review – it’s not for lack of interest or dedication, there is, in my view simply not much to address before it can be recommended for publication. There are a few general concerns I have below I would like to see addressed, followed by very minor comments.

General:

• There is one essential study that deals with controls of surges in HMA I am surprised you have left out of your Introduction but more importantly of your Discussion, namely (Barrand and Murray 2006). I am no co-author on that, so no bias, but they specifically tried to find controls, for the same variables that you investigated found similar results but also looked at other potential drivers. Added to that I am wondering, whether or why not you have considered to look at bed topography (simply via (Farinotti et al. 2019))? For some surging glaciers a particular shape of the valley/bed coincides with the transition from reservoir to receiving zone. That is not a request to do it, I am just wondering why you haven’t expanded to other variables, like (Barrand and Murray 2006) also have done.
• The only really strong relation you find between variables is between glacier length and surge area (L215ff, and then Discussion). I am wondering however whether you did these comparisons also for the baseline ‘non-surging’ glaciers? Because I would assume that relation just holding true for any glacier, the longer it is the bigger it is and hence also the larger potential “surge area” it has. And then that wouldn’t really be a signal from the surging glaciers per se. Or am I missing something here?
• L230ff/L240ff: Since you do not address that further in the Discussion, I am curious why you think (a) there seems to be general smaller mass loss for surging glaciers but then later you show that they substantially loose mass at the end of the surge or just after it and (b) why you think that rapid onset of mass loss is? Just because ice mass was transported to lower elevations and hence melts faster? Or are there dynamic reasons at play that disguise some redistribution of mass as mass loss?
• L309ff: I think a major shortcoming of your study – that is naturally just stemming from the data we have access to – is that the period you investigate is shorter than some surge cycles. Khurdopin cycles have been around 20 years, Muchchuhar Glacier next to Shisper even much longer. So in a way you may miss some within that period. So when you compare to inventories that go further back I would have expected them to catch some surges which you may have failed to catch. But your numbers are consistently higher. That would suggest that your numbers are still too low (because you missed some of the currently quiescent ones). Considering that for example (Bhambri et al. 2017)’s data and maybe others as well are accessible, wouldn’t it be prudent to compare your inventories and see which glaciers you agree on and where they find some you didn’t and vice versa? (Bhambri et al. 2017) has a very similar number but you still say yours is ‘more accurate’. From the evidence you provide I find that difficult to be sure of.
• As you are definitely aware, more recently a number of glaciers have been found to ‘detach’ rapidly (Kääb et al. 2021; Leinss et al. 2020). Have you made sure that these detachments are not classified as a surge by you? Even maybe for yet undetected detachments?

Minor comments:

L54: A study that actually looks at economic and infrastructure impacts of a surge, and that same surge, would be (Muhammad et al. 2021). I am a co-author on that study and I also do not think it is at all essential to cite here. I leave it up to you in case you find it helpful to make your point.

L146: “Consortium et al. 2017” reads funny – I know citing the updated RGI is a bit strange but the official format is below, so I would at least go with ‘RGI Consortium 2017’:

RGI Consortium (2017). Randolph Glacier Inventory – A Dataset of Global Glacier Outlines: Version 6.0: Technical Report, Global Land Ice Measurements from Space, Colorado, USA. Digital Media. DOI: https://doi.org/10.7265/N5-RGI-60

L155 and throughout: capitalize ‘Glacier’ when associated to a specific glacier. You do that sometimes, sometimes not. Same for ‘Tibetan Plateau’ on L181

L304: You write ‘Data not shown’ but wouldn’t you be able to state here in the text by how much it reduces if you set k to 3 or 2?

Literature:

Barrand, Nicholas E., and Tavi Murray. 2006. “Multivariate Controls on the Incidence of Glacier Surging in the Karakoram Himalaya.” Arctic, Antarctic, and Alpine Research 38 (4): 489–98. https://doi.org/10.1657/1523-0430(2006)38[489:MCOTIO]2.0.CO;2.

Bhambri, R., K. Hewitt, P. Kawishwar, and B. Pratap. 2017. “Surge-Type and Surge-Modified Glaciers in the Karakoram.” Scientific Reports 7 (1): 15391. https://doi.org/10.1038/s41598-017-15473-8.

Farinotti, D., M. Huss, J. J. Fürst, J. Landmann, H. Machguth, F. Maussion, and A. Pandit. 2019. “A Consensus Estimate for the Ice Thickness Distribution of All Glaciers on Earth.” Nature Geoscience 12 (3): 168–73. https://doi.org/10.1038/s41561-019-0300-3.

Kääb, Andreas, Myléne Jacquemart, Adrien Gilbert, Silvan Leinss, Luc Girod, Christian Huggel, Daniel Falaschi, et al. 2021. “Sudden Large-Volume Detachments of Low-Angle Mountain Glaciers - More Frequent than Thought.” The Cryosphere 15: 1751–85. https://doi.org/10.5194/tc-2020-243.

Leinss, Silvan, Enrico Bernardini, Mylène Jacquemart, and Mikhail Dokukin. 2020. “Glacier Detachments and Rock-Ice Avalanches in the Petra Pervogo Range, Tajikistan (1973–2019).” Natural Hazards and Earth System Sciences Discussions, 1–31. https://doi.org/10.5194/nhess-2020-285.

Muhammad, Sher, Jia Li, Jakob F. Steiner, Finu Shrestha, Ghulam M. Shah, Etienne Berthier, Lei Guo, Li-xin Wu, and Lide Tian. 2021. “A Holistic View of Shisper Glacier Surge and Outburst Floods: From Physical Processes to Downstream Impacts.” Geomatics, Natural Hazards and Risk 12 (1): 2755–75. https://doi.org/10.1080/19475705.2021.1975833.

Overview

Overall, this paper provides a useful inventory of surge-type glaciers across all of High Mountain Asia based on a comprehensive analysis of changes in glacier velocity, surface elevation and surface crevassing. It is the most comprehensive such inventory of its type for this region, and the topic is of direct relevance for The Cryosphere. Overall the paper is well written and well illustrated, and I believe that it is worthy of publication. However, there are many smaller items that need to be addressed, as detailed below. The most significant of these is the use of inconsistent regions in different parts of the text, tables and figures; for example, Table 1 lists 22 regions, Fig. 7 uses 6 regions, Fig. 9 uses 8 regions, and Table 4 uses 4 regions. Insufficient information is provided to understand how these relate to each other, which makes it difficult to follow the arguments and statements in the paper. I also don’t believe the statement in the abstract that this study newly identified 491 surge-type glaciers, particularly when this seems to directly contradict what is shown in Table 4.

Detailed comments

L30: it would be useful to describe more explicitly as to which ‘combinations of regional… and local… factors encourage instability’, and in particular whether these are found in HMA

L43: change to ‘distribution of surge-type glaciers…’

L45: it’s a bit debateable to state that ‘no HMA-wide inventory of surge-type glacier exists’, as one is already encompassed within the global study of Sevestre and Benn (2015). So I would refine the wording to say something like no ‘dedicated’ HMA-wide inventory currently exists.

L46: state which hazards are being referred to here – e.g., ice-dammed lakes? Perhaps also ice avalanches?

L60: change to ‘studies have documented…’

L75: you say here that you identify surge-type glaciers from distinct widely used criteria, but the studies that you quote all use the presence of looped surface moraines as a major criteria, whereas it seems that you don’t. It would be useful to add a sentence to make this clear, and to explain why you don’t use this criteria.

L78-80 (and elsewhere, e.g., L85): specify the time period that you’re referring to for ‘substantial and spatially concentrated surface elevation changes’ and ‘substantial variations in a glacier’s velocity field’ to be classified as indicative of surging

L97: I believe that the measurement periods for Hugonnet et al. (2021) are 2000-2004 and 2005-2009 (also check caption for Fig. 1, which should be 2010-2019, and elsewhere throughout your paper). See: http://maps.theia-land.fr/theia-cartographic-layers.html?year=2021&month=09&collection=glaciers

Fig. 2a: I assume that Fig. 2a shows the velocity patterns for a non-surging glacier, so make this clear in the caption. It would also help to highlight the positive heavy tail in Fig. 2b if you used a symmetrical x-axis scale for Fig. 2b (e.g., -150 to +150), to match the symmetrical scale already used for Fig. 2a

L139: provide the resolution in m that defines VHR

L140: capitalize Bing Maps (and elsewhere, such as L161)

L140: I don’t know what ‘infirm’ means in this context; do you mean ‘infer’?

L163: you provide the resolution of the velocity data here (240 m), but this should also be mentioned in Section 2.2

L173: to avoid any potential ambiguity, I would suggest modifying this sentence to say something like ‘at least 2 of 3 proposed identification criteria of rapid changes in surface elevation, surface velocity and surface crevassing’ (assuming that these are the criteria that you’re referring to here!)

L11 and L184: I’m not convinced by the statement that you newly identified 491 surge-type glaciers as it seems that you’re only making comparisons with the RGI here? After working with the RGI myself, I know that their inventory is incomplete for their designation of which glaciers are surge-type. Rather, you need to make comparisons with other previously published studies to obtain an accurate number of which glaciers you’ve newly identified as surge-type, as you already do in Section 4.2 and Table 4. Indeed, Table 4 suggests that you haven’t identified many new surge-type glaciers in some regions such as the Karakoram, and may have actually missed large numbers of them in the Pamirs.

L199: to reinforce the point that there is high correlation between the geometrical parameters, it might be useful to state how longer glaciers by definition have shallower slopes if they cover the same elevation range as smaller glaciers.

Fig. 5: please include part labels (a, b, c, d) for these figures, and indicate in the caption as to which geometrical attribute each figure part shows. It’s currently a bit cryptic to try and figure out what log_range refers to, for example. Also provide units for each x-axis.

L235: I dislike sentences with clauses in brackets as it makes them difficult to follow, particularly when there are multiple such sentences back-to-back. It takes about the same amount of space to write out the sentences properly, but makes them easier to understand: e.g., ‘We however note that balance distribution for surge-type glaciers in the Karakoram is positively skewed with g = 1.1, and negatively skewed with g = −1.4 for non surge-type glaciers.’

L238: I don’t follow the comment that the Himalayas are not representative with <10 surge-type glaciers, as Fig. 7 specifies that the Himalayas has n=13 surge-type glaciers, and Table 1 suggests n=14 (unless Himalayas is defined in a different way in Fig. 10 than in other figs and tables, which relates to my comments below)

Fig. 7 and 8: please clarify how the six regions here compare to the HIMAP regions listed in Table 1 (perhaps by adding an extra column to Table 1?). For example, I can’t figure out which of the six regions the Gangdise Mountains fits into. Some numbers also seem to be inconsistent between Table 1 and Figure 7: e.g., Central + Western + Eastern Himalaya = 14 in Table 1, but n = 13 in Fig. 7. Also present the figure parts in the same order in each figure so that it’s easier to compare them (e.g., Tien Shan is shown first in Fig. 7, but Tibet is shown first in Fig. 8)

Fig. 9 and 10: these use 8 regions, compared to the 6 regions in Figs. 7 and 8, and 22 regions in Table 1. This makes it essentially impossible to make comparisons between the different figures, and makes it even more confusing as to which regions in Table 1 are included in which regions in the figures. Please be consistent throughout, and clearly define how the regions relate to each other.

L242 (and elsewhere): formal glacier names should be capitalized when referring to a single glacier: e.g., Khurdopin Glacier, Hispar Glacier

L244: add superscript -1 at the end of: ’ -0.22 – 0.3 m w.e. a^-1’

Table 2: define in the caption as to whether these values are for the entire basin of each glacier?

Fig. 11: I don’t follow the x-axis label for parts a and b: how can the units be both dimensionless and in m yr^-1?

L260: I’ve read little, if anything, about previous surges lasting for 18 years in the HMA, so it would be useful to expand on this to provide more information about the location and characteristics of these, and how you can be sure that they surged for that entire time. Presumably it would also be more accurate to say ‘at least 18 years’, since this is the maximum length of your record?

L261: Is the equation reference here correct? IPR is defined in Equation 3, while Equation 4 defines the surge index.

L271: ‘active phase’ is repeated twice here

L278: change to ‘prevents identification of active phases…’

L286-7: this statement is a bit meaningless without anything to back it up; instead, it would be useful to provide some specific numbers here to convince the reader that the patterns you measured are real. For example, provide an average value for the quoted elevation uncertainty in the Hugonnet dataset, and state what the average elevation changes were that you measured on surge-type glaciers. Same for velocity changes.

L296: remind the reader here of what the discriminatory criteria are, so that they don’t have to go searching back through the previous sections

L302: change to ‘certain level of certainty’

L315: you also need to make comparisons with the inventory of Sevestre and Benn (2015) here. You currently do this in section 4.3, but that text would be better moved to here. It would also be helpful to discuss why Sevestre and Benn (2015) identify so many more surge-type glaciers in the Pamirs than you, even after removing their duplicates.

L322: you’re missing the inventory of Copland et al. (2011), who identified 90 surge-type glaciers in the Karakoram

L350: change to ‘up to 45% of total glacier area…’ to make it clear that you’re referring to the entire glaciated region here, and not just the proportion impacted on each individual surge-type glacier

L372: seem to be missing some words here? ‘in mass balance a single…’ doesn’t make sense as written