This is a review of the manuscript titled “Thermokarst lakes disturb the permafrost structure and stimulate through-talik formation in the Qinghai–Tibet Plateau, China: A hydrogeophysical investigation” by X. Ke et al. This manuscript details an investigation primarily focused on geophysical measurements related to permafrost properties around lakes in the Qinghai Tibet Plateau. The main objective is to characterize permafrost structure and the morphology of sublake taliks using direct current electrical measurements and time-domain electromagnetic measurements of permafrost electrical properties. Overall, the text is written clearly and English usage is good. The manuscript lacks a clearly articulated science question, and therefore it is challenging to determine if the main objective of the research was achieved. Additionally, concerns are raised below about geophysical data acquisition, processing, and presentation – while it is not clear that there are detrimental issues with the methods at this point, insufficient information was provided to fully evaluate these issues. 

Specific Comments:

The research question remains unclear. In the manuscript, I was unable to locate the word “question” nor any use of a question mark “?” – a question has not been posed here, and therefore it is difficult to determine if they authors have answered the research question. Furthermore, there is no evidence of a hypothesis that is stated or tested.

Line 151: The measurement parameters for TEM are confusing. The authors state that a 40,000 m^2 loop was used for transmitting, however this is an unusually large loop size for such shallow measurements. Typical loop areas for TEM within the top 400 m would be in the range of 1,600 m^2 to 10,000 m^2 (the vast majority being towards the lower end of this range). I was unable to retrieve any information from the manufacturer about this instrument to confirm if 40,000 m^2 is indeed correct, and if so, why such a large loop size would be used in shallow investigations (in the context of TEM, I consider anything <200 m to be a shallow target).

 Line 158: The use of the approach in Constable et al 1987 is acceptable, however the authors do not reference (either in the manuscript or supplement) which codebase or commercial software was used for the inversions. If the authors created their own implementation of the inversion detailed in Constable et al 1987, I would encourage them to share the codebase in accordance with open data policies and benchmarks should be provided to demonstrate that their code produces consistent results with existing free and paid software that is available. Furthermore, key data pre-processing details are omitted.

I appreciate that the measured and modeled TEM data are provided in Figure 6 (the same should be done for pseudosections of the ERT data), however these figures seem to reveal that much of the apparent electrical structure has not been fit during the modeling. This is evidenced by the spread of the data and non-linearity of the relationships shown in Figure 5 particularly (but also in Figure 6). In conjunction with the sparse information on TEM pre-processing and inversion make me concerned about the reliability of the TEM results. 

Line 190: The method to calculate the maximum detection depth is not stated. 

Line 193: “lack of borehole temperature measurements” I don’t understand this, ground temperature to >50 m is presented in figure 3, and can easily be modeled to greater depths. Temperature correction should be considered for all geoelectrical images given that large temperature gradients may be present in the subsurface (e.g., Figure 3), particularly within the top 5-10 meters.

Line 324: How was ALT interpreted from this image? It is unwise, if not impossible, to reliably interpret ALT from ERT data because 1) at any reasonable electrode spacing, the ALT will be too close to the surface to image because the layer thickness is ~1x – 4x the electrode spacing and may only occupy 1 or 2 vertical elements in the mesh, and 2) ERT images are inherently smooth and do a poor job of resolving sharp interfaces such as encountered at the ALT.

Section 3.3: What is the purpose of this section? It does not appear to play a role in the Discussion section (i.e., Figure 9 is not referenced in the discussion, nor are observations from GTM explicitly discussed in the context of the geophysical measurements and previous research), and if not, why is it included? Presumably the authors would want to consider all of their presented work in the context of other results.

Section 3.4 may be better suited for the Discussion section, and if moved, should be augmented with appropriate references. 

Figure 1: It is unacceptable to have the tomograms on different color scales because it makes unbiased interpretation impossible. Each tomogram in this figure must be presented on the same colorscale. Also, I suggest considering the current consensus on colormaps for the presentation of scientific results, and pick one that is more accessible: Crameri, F., Shephard, G. E., & Heron, P. J. (2020). The misuse of colour in science communication. Nature communications, 11(1), 5444.

I appreciate the opportunity to review your manuscript. I think it will make a good contribution to the discipline.

Overall, I think the authors present a well-structured study on the impact of thermokarst lakes on permafrost structure in the Qinghai–Tibet Plateau. The combination of geophysical prospecting methods (ERT, TEM) with temperature measurements is a robust approach to investigate sublake talik formation. The manuscript is generally well-written and organized, making it easy to follow the authors' line of reasoning. The topic is relevant, particularly given the context of climate warming and permafrost degradation, and the study appears to be the first of its kind in this specific region. I recommend publication after minor revisions.

Specific Comments:

Line 37: The authors state that permafrost degradation is the main driver. However, it should be acknowledged that other factors, such as precipitation and temperature, also play significant roles. Please revise the sentence to reflect this.

Line 80: The use of "spatial distribution" is too broad. The study focuses on the vertical distribution of permafrost. For accuracy, I recommend changing it to "vertical distribution" or "depth distribution."

Line 95: The abbreviation "TDS" is used without prior definition. Please introduce "TDS" (Total Dissolved Solids) at its first occurrence.

Line 140: The figure reference should be corrected to read "Figs. 4c and 1d. "

Line 189: The streams mentioned to the ERT transect ER1 are not marked in Figure 1. Please add the streams to Figure 1 for clarity. Additionally, consider removing or justifying the inclusion of lakes that are not directly relevant to the study to avoid clutter in Figure 1.

Line 295: The anomaly in ground temperature at a depth of 10-36 m in Figure 9a requires further explanation. Please provide a possible reason for this anomaly. It is critical to address this to give readers confidence in the data's veracity.

Line 297: The text mentions the application value of ground temperature monitoring in cold regions and suggests using the data in Figure 9 to estimate the lower limit of permafrost. Elaborate on how the temperature gradient can be used for this estimation. This would add value to the discussion.

Figure 9h: The vertical axis of Figure 9h should be reversed to maintain consistency and ease of interpretation.

Abbreviations: A list of acronyms used throughout the manuscript should be included at the end of the paper to improve readability.

Line 23 in SI: The unit for hydraulic conductivity should be corrected from "(y/m/d)" to "(m/d)".

Line 29 in SI: The temporal series data should include the appropriate unit (e.g., temperature in °C).

The study presents a hydrogeophysical investigation of permafrost and talik distribution around thermokarst lakes in the Qinghai Tibet Plateau. The topic is of interest to me due to the still limited understanding of how permafrost and thermokarst lakes interact, as well as the limited number of studies conducted in the region where this study was performed. The manuscript reads easily and is relatively well organized. I also appreciate seeing the use of TEM for deeper investigation than the ERT, as well as seeing the differences between both methods. However, I have several major concerns, outlined below. I recommend major revisions before publication.

Major Comments:

The aim of the paper is not clearly stated in the introduction. I suggest that the authors clarify this by introducing a specific research question they are trying to answer or a hypothesis they are aiming to test. Based on the title, I assume the study is focused on the impact of thermokarst lakes on permafrost distribution. Although the discussion returns to this topic, it reads more like a general summary of current literature than a clear comparison of this study’s results with previous findings, or a demonstration of how the study confirms or reveals new insights.   Additionally, based on the manuscript content, it appears the main focus might be the estimation of permafrost structure around thermokarst lakes, and the comparison between ERT and TEM methods. I suggest the authors refine the key research questions or focus of the study in the introduction, and revise the discussion section accordingly.

It is very difficult to evaluate the ERT and TEM results as presented. The various transects are shown with different resistivity color scales. In Figure 4, ER1 and ER3 intersect, and ER4 and ER5 are subsets of ER1, yet each has a different color/value range. I suggest using a consistent color scale throughout the manuscript, or at least adding all the figure with a consistent color scale in the supplementary material, to allow for meaningful comparison.

The use of color scale is also misleading in Figure 8, I think. It gives the impression of a sharp subsurface boundary without showing the actual “smoothed” inversion result. If the authors want to show results in this manner, I believe the inversion should first be presented with a more typical color scale at least in the supplementary material for transparency. In addition, it is very likely that there are discrepancies btw the resistivities values and boundaries identified with the TEM and ERT. I do believe that improving the discussion of differences and reasons (in section 3.4 or section 5) would benefit the paper.

Specific Comments:

Line 127: Apparent resistivity (ρₐ) is not the same as resistivity. In this paragraph, the term “apparent resistivity” seems to be used incorrectly or inconsistently. Consider clarifying by discussing the measured resistance and the inferred electrical resistivity separately.

Line 173: Consider moving Section 2.5 earlier in the manuscript to explain the survey strategy before providing details about each method. This will help the reader understand the overall approach more clearly.

Line 227: The statement “The ERs for each profile are relative values rather than true values...” is confusing. Electrical resistivity is not a relative value. Please clarify what is meant here.

Figure 5: Consider explaining why the fit appears low. If the misfit were expressed as a percentage, I believe it would appear significant. Please consider providing the misfit in % and discussing the possible reasons more clearly.

Line 364: It is great to see a section dedicated to “Limitations.” However, I am surprised there is no discussion of challenges related to data inversion in TEM, the comparison of ER values between ERT and TEM, or the difficulty in defining a boundary value to delineate permafrost.