Review of nhess-2022-161

Summary and general comments

The paper “Temporal evolution of crack propagation characteristics in a weak snowpack layer: conditions of crack arrest and sustained propagation” deals with the dynamic crack propagation within weak snowpack layers. The dynamic crack propagation occurs within the release process of dry-snow slab avalanches and is therefore of interest for the (scientific) avalanche community. The authors performed 24 propagation saw tests (PSTs) over a period of three months, recorded those tests with a high-speed video camera, and manually measured snowpack properties. The camera data was analysed with digital image correlation techniques. From the PST experiments the authors derived the elastic moduli of the slab and the weak layer and the specific fracture energy of the weak layer with a homogeneous and a layered slab model. Moreover, they measured crack speed, touchdown distance, crack length, and energy dissipation.

The data presented within the paper is certainly new and of value to the scientific community. The writing of the paper sometimes seems unnecessarily complicated to me. For readers familiar with the topic, the paper should be good to read. I suggest performing the modifications to the paper listed below. All in all, I feel that the paper is a valuable contribution to the scientific community.

Specific comments

First sentence of abstract: This sentence might seem a bit confusing and complicated for readers not familiar with slab avalanche release. I suggest making it simple: “For a slab avalanche to release, we need crack propagation in a weak snow layer beneath a cohesive snow slab.” Then continue describing crack propagation, etc.

Abstract again: a PST is not common knowledge, I would think. Maybe write “we performed crack propagation experiments… ”. You can then describe the PST in more detail in the intro and/or methods section.  

Figure 2: What about the crack speed curves for the other experiments? It would be interesting to see the others as well. Are they very similar or completely different? If there are differences, how can those differences be explained?

Line 225: How valid is the assumption that the slab and substratum are in the same stress state before and after crack propagation? Will there be no plastic deformation within the slab due to the collapse?

Line 275: Here you say that the slab was shallow and soft and it broke while cutting the weak layer. How does this fit with the assumption in line 225?

Discussion: In general, the first sentence of a paragraph should summarize the paragraph and tell the reader what the paragraph is about. I have the impression that this “first-sentence-summary” concept was not used in the discussion, which makes it a bit tedious to read. I suggest adding “first-sentence-summaries” at the beginning of the paragraphs.

Technical corrections

Line 52 and at many further places throughout the document: Variable identifiers (in your case the f and the dyn) are not italicized.

https://blog.apastyle.org/apastyle/2011/08/the-grammar-of-mathematics-writing-about-variables.html#:~:text=Identifiers%20(which%20can%20be%20superscript,boys%E2%80%9D)%20are%20not%20italicized.

Line 78: omit the “initial”

Lines 84-85: firstly and secondly (adverbs referring to the verb “is”)

Figure 1b: using a second y-axis (on the right) for the temperature would be more elegant.

In "Temporal evolution of crack propagation characteristics in a weak snowpack layer: conditions of crack arrest and sustained propagation" long Propagation Saw Tests (PSTs) were performed over 10 weeks in a well-controlled outdoor setting and analyzed using digital image correlation. Material and dynamic parameters such as crack speed were measured. Single and multi-layer analytical models were compared.

I found this research compelling and recommend it be published subject to technical corrections. The analytical models applied have quite a few new features and the PSTs were substantially longer than in previous studies. The measured touchdown distances of over 5 m and the conclusion that beams must be longer than this distance to study self-sustained crack propagation is important, suggesting that it is nearly impossible for a practitioner, who does not have all day to perform a PST nor such an ideal location, to witness self-sustained propagation. The importance of layered model is well demonstrated.

I have a few minor critiques.

• The authors note that in some cases the "snowpack tended to arrest cracks without slab fractures" but no information is provided about how the absence of slab fracture was measured.
• The choice of column length in all the experiments (Table 1) is not justified. I presume other tests were done before the authors decided to excavate a 9 m long PST? The abstract says that tests up to 10 m long were performed, but the longest column length in Table 1 is 9 m. Please explain.
• Edge effects from the near and far end of the PSTs are discussed, but the edge effect of the width of the PST is only briefly mentioned (as an experimental error on Feb 22 2019). I assume the width of most of the PSTs was 30 cm? This needs to be stated.
• There are at least 2 studies, e.g., Bobillier (2022) and Trottet et al. (2022), that are not publicly available, as they are in review and in press, respectively. By Copernicus standards (https://publications.copernicus.org/for_authors/manuscript_preparation.html), these articles can only be cited if they are available to reviewers, and must be publicly available at the time of final submission.

Additional comments are attached as an annotated PDF.

NB 7/19/22