The authors present an updated inventory of glacier, rock glaciers and perennial snowfields. This is a much appreciated and needed update since the currently available inventory is based on topographic maps from the 1980s or earlier. This inventory is available through the GLIMS data base also part of the current version of the Randolph Glacier Inventory (vers. 6.0).

The authors identify and digitise the glaciers based on very-high resolution aerial imagery (spatial resolution 1m or better) and use imagery available through Google Earth as additional information. The inventory is nicely illustrated and details of the different mountain ranges presented.

Overall, the work seems to be performed with care and outlines seem to be of high quality. However, the paper is written like a report and not a scientific paper, the description of the methodology is a bit thin and the accuracy assessment could be improved. Moreover, some additional analyses, like the comparison to the old inventory and the inclusion of typical inventory parameters would be very beneficial. Overall the manuscript needs some improvements and revisions but should ultimately be published.

More details are presented below.

L16/17: All numbers of glacier areas should be presented along with their uncertainty ranges.

L17: What are buried-ice features? Do these include rock glaciers? As I understand you have a separate class “rock glaciers”. However as these are not glaciers per se, these should also be treated separately (see more details below)

1. 32: A nice paper which showed this (although for Asia not for the US) is Pritchard (2019). A consideration of Huss and Hock (2018) would also nicely fit here.

L39ff: This is a bit confusing: You may want to mention already here that an inventory is existing (and refer to the reference), but that the inventory is outdated.

L 62: This should be rather a scientific paper and not a report.

L 69ff: The info how the utilised images were orthorectified is missing but should be included along with some info about the accuracy of the orthoimages. How were the google earth images used? Were they co-registered to the aerial imagery? How well does the geolocation fit to the old maps resp. the former inventory?

L 82: Provide a rational for the selected threshold.

L86: It would be beneficial to attempt to distinguish between perennial snow and ice. I think it should be possible, e.g. if at one scene only bare ice is visible than it is an ice patch. But I do not know the images and it might be difficult or introduce to much subjectivity.

Paragraph L94ff: I ask the authors to be more specific with the challenges and present some example figures in the main text and not only in the regions. It is e.g. well known that debris-covered ice is partially difficult to identify and the delineation subject to the operator (and even the operation is not consistent) as nicely illustrated by Paul et al. (2013). How did you deal with the problem?

Paragraph L103ff: This section needs more depth and more detailed information. As correctly mentioned, it is not trivial to separate glaciers from rock glaciers. I recommend moving some of the text from the different regions to the main text (e.g. L271f, l309ff) and be more specific. Moreover, rock glaciers are commonly associated with permafrost. I know that opinions diverge, but this should at least be mentioned, and you need to consider the guidelines by the IPA Action Group Rock glacier inventories and kinematics (RGIK, 2022). You may also think about overlaying existing permafrost modelling results with the rock glacier inventory. Snow fields above rock glaciers are usually not considered as a part of rock glaciers. In general, the literature should be better considered here (e.g. Ostrem, 1970, Janke 2007, Mölg et al. 2018, Charbonneau and Smith, 2018, Janke and Bolch 2021 but there are several other relevant papers).

L113ff: I appreciate that you include buried ice adjacent to glaciers. However, the terminology is a bit unclear. Are these ice-cored moraines (see e.g. Lukas 2011)? Or ice identified in the forefield of glaciers? Interesting papers in this regard might be Lukas et al. (2005), Bolch et al. (2019).

L127f: More details are needed. What were the difficulties? How where they overcome (e.g. in case of debris cover)? How many of the outlines had to be adjusted?

L132ff (Uncertainty). A more rigorous uncertainty assessment is needed. The digitisation error itself might be small if the glaciers are well identifiable, but many are not. The image resolution had also an influence (cf. Paul et al. 2013 as cited earlier). You may think to use a buffer (e.g. Granshaw and Fountain, 2006, Bolch et al. 2010) as additional measure. It makes sense that a higher uncertainty is assigned to perennial snow fields. However, a justification for the 30% used would be beneficial. What about the uncertainty of rock glaciers and buried ice? There should at least be some information.

Paragraphs L148ff and L161ff should be moved before the uncertainty paragraph.

L172ff (Results section): The results section is too thin also for a data paper. More information and analysis (e.g. about topographic variables: min, max, median elevation, hypsometry, slope, aspect, mean elevation vs. distance from the sea, e.g. a figure similar to Fig. 1 but with the mean elevation colour coded would be very valuable) can easily be included and would strongly increase the value of the paper. The other inventory papers in ESSD typically provide information beyond the pure description if the inventory. Moreover, all numbers should be presented along with the uncertainty ranges.

Table 1: Rock glaciers should be an own class. I recommend to show the total uncertainty are ranges along with the area(e.g. 10.63 ± 0.61 km²). The meaning of Max area and Mean area is unclear.

A discussion section is completely missing but would be highly beneficial. Here the authors can discuss methodological challenges and how they overcome along with the literature and in particular compare their inventory with the existing former one(s).

The summary section should be turned into a conclusion section.

Do not hesitate to contact me in case you have any question or a comment need clarifications.

Best regards,

Tobias Bolch

References:

Bolch, T., Rohrbach, N., Kutuzov, S., Robson, B.A., Osmonov, A., 2019. Occurrence, evolution and ice content of ice-debris complexes in the Ak-Shiirak, Central Tien Shan revealed by geophysical and remotely-sensed investigations. Earth Surf. Process. Landforms 44, 129–143. https://doi.org/10.1002/esp.4487.

Charbonneau, A.A., Smith, D.J., 2018. An inventory of rock glaciers in the central British Columbia Coast Mountains, Canada, from high resolution Google Earth imagery. Arct. Antarct. Alp. Res. 50, e1489026. https://doi.org/10.1080/15230430.2018.1489026.

Granshaw, F.D., Fountain, A.G., 2006. Glacier change (1958-1998) in the North Cascades National Park Complex, Washington, USA. J. Glaciol. 52, 251–256.

Huss, M., Hock, R., 2018. Global-scale hydrological response to future glacier mass loss. Nature Clim. Change 8, 135–140. https://doi.org/10.1038/s41558-017-0049-x.

RGIK, 2022. Towards standard guidelines for inventorying rock glaciers: baseline concepts (version 4.2.2). IPA Action Group Rock glacier inventories and kinematics, 13 pp. https://bigweb.unifr.ch/Science/Geosciences/Geomorphology/Pub/Website/IPA/CurrentVersion/Current_Baseline_Concepts_Inventorying_Rock_Glaciers.pdf

Janke, J.R., 2007. Colorado Front Range rock glaciers: Distribution and Topographic Characteristics. Arctic, Antarctic, and Alpine Research 39, 74–83.

Janke, J.R., Bolch, T., 2021. Rock glaciers, in: Haritashya, U.K. (Ed.), Treatise on Geomorphology, 2nd Ed. Elsevier. Doi: 10.1016/B978-0-12-818234-5.00187-5

Lukas, S., 2011. Ice-cored moraines, in: Singh, V.P., Singh, P., Haritashya, U.K. (Eds.), Encyclopedia of Snow, Ice and Glaciers. Springer Science+Business Media B.V, Dordrecht, pp. 616–619.

Lukas, S., Nicholson, L. I., Ross, F. H., & Humlum, O. (2005). Formation, meltout processes and landscape alteration of high-Arctic ice-cored moraines. Examples from Nordenskiold Land, Central Spitsbergen. Polar Geography, 29, 157–187.

Mölg, N., Bolch, T., Rastner, P., Strozzi, T., Paul, F., 2018. A consistent glacier inventory for the Karakoram and Pamir regions derived from Landsat data: distribution of debris cover and mapping challenges. Earth Syst. Sci. Data Discuss., 1–44. https://doi.org/10.5194/essd-2018-35.

Østrem, G., Arnold, K., 1970. Ice-cored moraines in southern British Columbia and Alberta, Canada. Geogr. Ann. A 52 A, 120–128.

Paul F. et al. 2013. On the accuracy of glacier outlines derived from remote sensing data. Ann. Glaciol. 54, 171–182.

Pritchard, H.D., 2019. Asia’s shrinking glaciers protect large populations from drought stress. Nature 569, 649–654. https://doi.org/10.1038/s41586-019-1240-1.

Dear editor, dear authors,

I greatly apologize for the delay in this review, and I hope that its generally positive tone can compensate for the long waiting time. The delay is related to my recent move to another hemisphere and a heavy proposal deadline. Since it is the first time I am reviewing a scientific article in a data format, I was unsure what contents and structure are expected in such publications. Thus, my recommendations focus on the necessary improvements in the clarity and interpretation of the results rather than on requirements for the manuscript structure. Nevertheless, I do agree with Tobias that the format of a technical report is hardly appropriate for a scientific publication in a high-impact journal like the ESSD, with a discussion section painfully missing and the summary section being too brief.

Currently, the manuscript reads almost like a mini article (i.e., a correspondence type), except for its enormous appendix section. Maybe it fits the expected format of ESSD manuscripts, but there are some additions that are necessary to make it easier to follow its arguments and understand the interpretation of the dataset. I hope the recommendations that Tobias and I have provided are useful to achieve such a transformation. Given the above thoughts, I recommend the publication of this manuscript after moderate revisions.

Specific comments:

This article presents an important dataset - a new inventory of glaciers and perennial snowfields for the entire western US that has been long overdue. Most glaciated regions in the world, especially in the “collective West”, undergo repeated country- and region-scale inventories of cryospheric landforms on a regular basis, and it is somewhat surprising that such inventories are missing in the western US. I think it would be great to include a paragraph in the introduction section explaining the reasons for why systematic mapping of glacier and snowfield changes has not been done on a regular basis.

I am wondering about the statement in the introduction section (lines 59-61) that “automated schemes are known to misclassify debris-covered ice”. I do not agree with this generalization, since most recent studies have shown that machine learning algorithms can assist the mapping of complicated landforms such as debris-covered and rock glaciers when surface movements (associated with periglacial/glacial processes) are mapped in time and using a combination of satellite products, resulting in a relatively high precision of mapping compared to manual (e.g., Lu et al., 2022; Robson et al., 2020). I think this recent development deserves some space in the introduction and also in the currently missing outlook (i.e., future technological improvements). It is also relevant to mention here the statement from lines 84-85 – if a long-term surface movement could be automatically tracked, one would not need to rely on such superficial criteria as the presence of crevasses. For example, in Norway there are several major ice caps that are nearly completely void of crevasses and thus do not qualify as glaciers according to the established criteria. In contrast, in the Andes, there are glaciers that are surrounded by snow/ice penitents that might appear as crevasses from above but are not part of the glaciers. Such exceptions exist in every glaciated part of the world, albeit with regional differences. Hence, possible future upgrades of the mapping methods need to be discussed in the “future perspectives” to avoid such limiting assumptions.

When reading through the text, there are several references to the appendix section that discusses individual cryospheric landforms causing various complications when being mapped or showcasing common unusual behavior. I think it is counterintuitive that in these instances, the article sends its reader to the appendix and its different sections. I believe that many of these belong to the discussion and should be visualized by two additional figures in the main text, including multiple insets corresponding to different issues encountered during mapping and different phenomena observed. These can both feed the non-existing discussion and provide support to the claims in the article that are poorly grounded right now, since the reader must read additional sections from the appendix to visualize and validate them. Furthermore, references to the interesting cases presented in the appendix are not providing a complete picture. I believe they could be strengthened in the main text by adding more examples from the appendix and crafting good visualizations.

The methods section related to the “uncertainty” has two main issues. First, the strategy for estimating uncertainties seems random and non-mathematical (at least in part). Could the authors explain why they decided to adapt such strategies and how they relate to other existing studies with similar objectives (especially, those that are more mathematically grounded)? Ideally, this should be done in the new discussion section. Also, large parts of the uncertainty section read like a discussion to me. I think this is where they should be relocated.

Finally, the results section has two paragraphs of text and is not attached to any discussion. I understand it is a data article but… two paragraphs? Is there nothing more to write about these data in the main text?      

Technical edits:

Lines 23 – 24: This needs to be expanded. Also, in lines 21 – 34 and in the article, the hazardous nature of glaciers can be included as a motivation (glacial lake outburst floods, glacier collapses and avalanches, etc.) in addition to their function as freshwater resources.

Line 39: A reference to Andreassen et al. (2022) could be useful here.

Lines 107 – 112: This text needs further clarification. It is a bit hard to follow.

Correct grammar/errors in lines 49, 57, 64, 142, and 181.

References:

Andreassen, L., Nagy, T., Kjøllmoen, B., and Leigh, J. (2022). An inventory of Norway’s glaciers and ice-marginal lakes from 2018-19 Sentinel-2 data. Journal of Glaciology, 68(272), 1085-1106, doi: 10.1017/jog.2022.20.

Robson, B.A., Bolch, T., MacDonell, S., Hölbling, D., Rastner, P., and Schaffer, N. (2020). Automated detection of rock glaciers using deep learning and object-based image analysis, Remote Sensing of Environment, 250, 112033, doi: 10.1016/j.rse.2020.112033.

Lu, Y., Zhang, Z., Kong, Y., and Hu, K. (2022). Integration of optical, SAR and DEM data for automated detection of debris-covered glaciers over the western Nyainqentanglha using a random forest classifier, Cold Regions Science and Technology, 193, 103421, doi: 10.1016/j.coldregions.2021.103421.

See attached PDF.