Authors present an interesting application of SIP on a permafrost site. The study is exhaustive and well presented, results are supported by the data presented and the work is of interest for TC readers. I have no doubt the paper deserves publication after some minor corrections on text and graphics.

- Line 73 and 76 seems describe the same concept, maybe the lab tests can be presented in one sentence.

- line 140 ρ is the here the real component (Ω m) ?

- Ln 176 Maybe the relaxation frequency of ice should be here better introduced

- chap. 3.2 The SIP setup description should include essential information as the electrode spacing, the arrays lengths, to  etc. to be specified line by line

-chap.3.2.2 .  August or September 2019 ? (see line 218). I think part of this paragraph is setup (3.2) rather than mapping. You assert resistance contact was in the range 5-60 kΩ , a very low values in such a debris condition (if compared to the literature ones, in the common order of hundreds). Did you help the contact some way ? and if yes, how do you increase the contact locally? This is of extreme importance in SIP results obtained below, and of huge interest for the  TC community.

-Ln 258 is it always 5 m the spacing ? did you collect dip dip data also with 10 m spacing?

-ln 302-304. This last sentence about robust inversion is not clear, is it real necessary?

-Fig.2, I suggest to label the figures (a,b,c,d) to help reader's comprehension. Modify the text and caption accordingly.

Ln 402. Sentence is not clear, I think the concept is ρa has no frequency dependence but it is still able to discern frozen parts.

- Ln 546-549, Sentence about Duvillard work is not clear.

Fig.8 the  c) panels must be differentiated, cause they are not clear in the caption. Please label the different depths in the figure.

-Ln 584-87 As I understood you performed lab test on Lapires sample rocks. This is part of the research (and then Method) and should no be presented here in the discussion section.

-Ln 601 Since the spacing was 5 m, why you average pixel of 1 m width. Maybe keeping the same spacing address more your survey lateral resolution.

Ln 614 How Coperey work differs from Limbrock one? This is of interest in the discussion of your results.

Fig. 9 I suggest label the panel. Looking the figure, the frequency dependence of AL and permafrost seems to have the same behaviour. This should be more emphasised in yours discussion. 

Thank you for the very interesting reading

The paper `Spectral induced polarization imaging to investigate an ice-rich mountain permafrost site in Switzerland` (authors: Teresa Maierhofer, Christian Hauck, Christin Hilbich, Andreas Kemna and Adrián Flores-Orozco) is one of the only very few studies investigating the applicability of the IP method in mountain permafrost. This paper brings valuable technical knowledge in discriminating between permafrost and unfrozen substrate within a talus slope in the Swiss Alps. Furthermore, the field protocol presented in this paper is essential to the permafrost community since its application is expected to provide supplementary information on the permafrost characteristics elsewhere. All the methodological parameters were carefully analyzed, and the results were compared with ERT measurements, borehole data and other studies. Overall, SIP discriminated very well against permafrost and unfrozen bedrock.

The paper is clearly structured and well-illustrated, contains adequate critical reflections and reaches appropriate conclusions. The methodology is rigorous and consistent, integrating several complementing techniques, whereas data back the interpretations. The results support the findings. There are no factual errors. Therefore, I suggest acceptance of the manuscript after some minor revisions. These are outlined below:

• In the Introduction, you mention that there are still limitations of 4PM because of the difficult interpretation of the geophysical signature of air, water, ice etc., based solely on 4PM, and thus, an improvement is desirable. Since SIP appears to be a promising complementary technique, it would be interesting to write a phrase within Section 5 (Discussion) to highlight how/if SIP can overcome these limitations and if this technique might be integrated into a more complex model in the future. However, I feel the potential of this new technique for overcoming the existing limitations was not clear enough discussed.
• It seems that sensor spacing is essential for delineating the active layer. Can you add an idea within the Discussion to refer to the importance of SIP sensor spacing for discriminating between ice rich-permafrost and unfrozen substrate and precisely delineating the active layer based on your experience/other studies findings?
• Figure 1 needs some adding. First, there is no scale on the inset map with Switzerland. Then, you should add to the legend the ERT profiles' location. PERMOS monitoring profile also refers to ERT? Please clarify! Moreover, there are no values for the contour lines. Finally, you should mention the source of the background image.
• Please add the coordinates of the Lapires talus slope in Section 3.1.
• In section 3.1, you mention that lithology consists of gneiss and schists, but in Section 5.2 (lines 580-585), you say that granite was collected and examined in the Lapires site. It is not consistent; please clarify.
• Some studies were conducted in the Lapires site, which were not cited and might be needed to consider (e.g., Delaloye et al., 2001; Scapozza et al., 2010).
•  Figure 4 should also contain information on the RMS error and electrode configuration. You added the intersection between crossing profiles but didn`t specify the name of the intersected profiles.
• Geophysical measurements were conducted in two different years. Have you noticed significant changes in the permafrost spatial extent between 2018 and 2019?
• Did the ERT monitoring multicore cable installed within PERMOS ERT network influence your SIP measurements?
• There is a significant reduction in permafrost extent compared with Delaloye (2004). Therefore, in Section 5.2 (lines 530-535), I suggest providing the difference in spatial extent between your findings and Delaloye (2004). On the other hand `the slight decrease of the spatial extent of permafrost with depth` is also supported by findings of Mollaret et al. 2019, which showed that the most significant resistivity decrease at LAH (profile H1) is `below the ice-rich body` due to air circulation. Please refer to this within Section 5.2.
• Technical corrections: The reference list lacks information about several papers cited in the text: Limbrock et al., 2021 (line 584), Duvillard et al. 2020 (line 546), Haeberli et al., 2018 (line 33), Krainer et al., 2015 (line 33); Revil et al., 2012 (line 63) (at the reference list is Revil et al., 2012a and 2012b), Waxman et al., 1968, Biskaborn 2016 (line 28) (at the reference list is 2019)? However, I think Biskaborn et al., 2019 is correct. Coperey et al. 2019a (line 43) should be replaced with Coperey et al. 2019 and Scapozza et al., 2015a should be replaced with Scapozza et al., 2015. Flores Orozco et al., 2018 and 2019 are in the Reference list but not cited in the text. The same goes for Haeberli et al., 1988, Krainer et al., 2014, Shengting et al., 2018!, Waxman and Smits, 1968!

References:

Delaloye, R., Reynard, E. & Lambiel, C. (2001). Pergélisol et construction de remontées mécaniques : l’exemple des Lapires (Mont-Gelé, Valais). Le gel en géotechnique, Publications de la Société Suisse de Mécanique des Sols et des Roches, 141, 103- 113.

Scapozza, C., Lambiel, C., Abbet, D., Delaloye, R., Hilbich, C., 2010. Internal structure and

permafrost characteristics of the Lapires talus slope (Nendaz, Valais). 8th Swiss

Geoscience Meeting 2010, Fribourg, Switzerland, 19–20 November 2010. Extended

Abstract 7.16, 166–167