Dear authors,
Given your responses, I look forward to receive the revised version of your "cast shadows" article and further evaluate it, likely with the help of the original referees,
Best wishes,
Etienne

Dear Authors,

Both reviewers recognized the large amount of work made to improve the manuscript. One is now fully satisfied with the revised text whereas the other reviewer still challenge the use of View Finder Panorama as the best DEM for the Baltoro case study. A sensitivity analysis, describing the difference of dh on Baltoro when using SRTM1 instead of VFP (as nicely down for Aletsch) should help to reconcile both views and strengthen further your study. (I wondered whether you and the reviewer used the same version of SRTM1. If you want me to ask his version of SRTM1 do not hesitate to contact me directly by email.)

I also agree with his point that the comparison to Hugonnet et al. rate of elevation change should be performed only on the common pixels, hence at the edge of the cast shadow and not over the entire shadowed glacier area.

L235. It is unconventional to acknowledge a colleague in the main text. This belongs to the acknowledgement section (where you need to correct “help”). Hence you could reformulate L235 and rather cite a personal communication for a peculiar data extraction.

Figure 5. The legend needs to explain which results are from DEMs and which are from cast shadows.

Looking forward to receive your revised text.

Best regards,

Etienne Berthier

Dear authors,

Thanks a lot for revising your manuscript and for taking into account the final comments made by the referees. I warmly thank here the reviewers for their involvement. The review process has led to some significant improvements and clarifications and, also, to a vibrant discussion with Anonymous Referee #1 about the “best” DEM over the steep peak and ridges surrounding Baltoro Glacier (Karakoram). I appreciated the way you considered every comment very seriously and performed some further sensitivity analysis. Fortunately, all these exchanges will be available to our readers. Your study and the related discussions will thus provide an excellent starting point to those who would like to explore the same research direction in the future and, who knows?, maybe apply it globally.

I am pleased to accept your study for publication in TC.

I made myself a few final suggestions that you will find below and I trust you will take my comments into account before uploading the final files for copy-editing.

Thanks for choosing our journal to publish your work,

Best regards,

Etienne Berthier

L51-52. I checked these two papers and did not find elevation changes derived from outlines. Rankl & Braun obtain elevation changes from repeat DEMs. Zhang et al. 2016 is a “classical” multi-temporal inventory. Be more specific or remove this sentence. [Maybe I missed something?]

L276. “Sperry and South Cascade Glacier lost on average” This gives the impression that the entire glaciers are observed. Rephrase to indicate that this for your shadow areas.

L293. Clarify for what period the ref DEMs are considered. Landsat period only?

L300. Make it clear in the title of the figure that these results are only from Ref DEM (and not from cast shadows).

L368. Could be misunderstood as an absence of voids! Rephrase to avoid ambiguity

L392. I do not think you used SRTMGL1 elsewhere in the article (SRTM-1).

L411. Do you want to be more specific in this sentence and indicate "Cast shadows from satellite imagery obtained for the date..." ?

L464. You cannot use “melt rate” in this context. Elevation changes are the results of SMB (melt and accumulation) and also change in ice dynamics (emergence velocity).

Appendix. Decide if you want to have this material as an appendix (I think you will need to pay some page fees for that) or in a supplement. Both are fine for me.