Dear Authors:I have received the recommendations from Martin Schneebli and Matthew Sturm for your revised m/s which is to accept with minor revisions. I look forward to receiving your revised m/s and responses to their comments included below.

Best regards, Ross Brown (Ed)

1: Dr. Scheebli is satisfied with your responses to his comments but has provided some suggestions for consideration in the final version of the paper:

- Concerning subnivean life:
Actually, subnivean life is very active also in alpine snowpacks. There is a lot of activity (especially mice) under alpine snowpacks. Section 4.2 is still unexpected and in my view not relevant for this paper.

- Concerning the thermal conductivity of depth hoar:
Concerning the compactive strength of depth hoar, indeed the compilation by Shapiro et al (Fig. B5) suggests that depth hoar with a density of 300 kg m-3 easily supports the weight of the snowpack studied (about 0.6 kPa). If this is the case, then the mechanical stability contradicts the very low thermal conductivity, as these processes are in an intimate physical relationship.

I suggest that the authors also consider the parameterization by Löwe et al. 2013, eq. 5, and fig. 1. This dataset includes more data points (167) than the paper by Calonne et al, and also at ice fraction below 0.1 (i.e. a density below 90 kg m-3). I would also like to know how the authors conclude that their (single) point value is representative, but at the same time write "even though there are clearly spatial variations". I am not able to extract any uncertainties from their figs. 2 and 7, or ranges of thermal conductivities. In summary, all current parameterizations suggest that the value measured at the bottom is highly uncertain, and needs further investigation.

- Concerning water vapor fluxes:
I agree with the revised calculations of the authors. As the authors point out, an extended air gap is not visible outside of the collapsed void or burrows in Fig. 8. Convective cells would even reduce the vapor flux, as their would be more cooling of the soil, and a convective cell formation seems also improbable as the authors describe hard and consequently less permeable layers above the depth hoar layer (Sturm et al made their measurements in a taiga snowpack).

- Concerning the mass loss, see also the paper by Ebner et al 2015, in this (and other) papers it is shown that advecting air is saturated within a few millimeter distance, and can not contribute significantly to mass transport.

In this view, I suggest that the authors formulate alternate reasons for the formation of the air gaps at the bottom of the snowpack more openly. I suggest that reasons are:
- burrows by lemmings
- possibly spontaneous collapse of the depth hoar during snow pit excavation

- Concerning the conclusions, I believe that it should be mentioned that the data and the models are not completely conclusive, and additional measurements are necessary.

Löwe, H., Riche, F. & Schneebeli, M. 2013 A general treatment of snow microstructure exemplified by an improved relation for thermal conductivity. Cryosph. 7, 1473–1480. (doi:10.5194/tc-7-1473-2013)

Ebner, P. P., Andreoli, C., Schneebeli, M. & Steinfeld, A. 2015 Tomography-based characterization of ice-air interface dynamics of temperature gradient snow metamorphism under advective conditions. J. Geophys. Res. Earth Surf. 120, 2437–2451. (doi:10.1002/2015JF003648)

2. Dr. Sturm is satisfied with the revised m/s but feels it could be improved further with more concise wording and attention to typographical errors. He asked that the following items be addressed before publication:

- There are a fair number of typos in the revised text.....the authors need to carefully clean these up., not pass them on to the tc staff editors.

- Indurated depth hoar: in my last review I reminded the lead author that I introduced him to this snow texture I think in about 2006(?). He cites our paper (Derksen et al., 2009) as the first published description, and as I recall, the section in that paper he quotes is a section I wrote. But I would request that he cite where this was first described in greater detail (attached - note: copy of report e-mailed to lead author by ed.) :

Sturm, M., Derksen, C., Liston, G., Silis, A., Solie, D., Holmgren, J., & Huntington, H. (2008). A reconnaissance snow survey across Northwest Territories and Nunavut, Canada, April 2007 (No. ERDC/CRREL-TR-08-3). ENGINEERING RESEARCH AND DEVELOPMENT CENTER HANOVER NH COLD REGIONS RESEARCH AND ENGINEERING LAB.

In that easily obtained report, we describe indurated depth hoar on pages 6, 14, 18, and provide a percentage of the snow sampled in Figure 10.

- A personal pique, but values like density and thermal conductivity we generally ascribe to a layer, and therefore (in lieu of an contradictory data) the same across the whole layer. Thus, when drawing a graph (Figures 2 and 7), they should be horizontal bars, rather than point-to-point lines. That was the old way it was done, and was more representative. It is easily done with most graphing programs.

- When citing references in the text, place them in date order (ascending or descending), and be consistent, please.

Thanks for the detailed responses to the last set of reviewer comments Florent. I'm satisified the paper is now ready for publication. Martin Schneebeli is ok to have his name included in the Acknowledgements as we discussed.

Best regards, Ross B