Review of “Brief communication: Measuring and modelling the ice thickness of the Grigoriev ice cap (Kyrgyzstan) and comparison with global datasets” by Van Tricht et al.

This brief communication presents new ground penetrating radar (GPR) ice thickness measurements for the Grigoriev Ice Cap in Kyrgyzstan. The manuscript describes the field campaign, the GPR measurements, and the interpolation method used to obtain a complete ice thickness map. Finally, the authors compare their results with global ice thickness datasets and highlight the discrepancies.

The figures are nice and the paper clear and well organized but the content of the paper is weak at this stage, even for a brief communication. Major revisions are required before it can be considered for publication.

General comments

Just saying that field measurements are needed because global thickness products are not accurate in this particular case is not very remarkable in itself. Global products do not attempt to be accurate everywhere, but rather give a volume estimate on a regional to global scale. The specific case of a polythermal, small ice cap is exactly where one would expect global scale estimates to be wrong.

In my opinion, what would make this communication publishable would be to highlight the reason why the different global estimates do not reproduce the observations. This would allow to identify which assumption done in those estimates can be improved and how. In the current version of the manuscript, this work is poorly done, as the authors have not really looked in detail at how these global estimates are made. This is shown by their assumption that these estimates are done for the year 2002 due to the SRTM DEM, which is wrong. This leads to a wrong correction of their thickness field and to irrelevant comparisons. For example, Milan et al. uses surface velocity from 2017/2018 combined to the shallow ice approximation to provide thickness estimate. The SRTM DEM from 2002 is only used to compute the surface slope. The method and assumptions of each estimate presented should be reviewed and analyzed in the light of what is known about the Grigoriev Ice Cap. This would allow to identify the origin of the errors in the reconstructed thickness (mass balance, ice viscosity, sliding, surface velocity .....).

Specific comments

You will find a list of correction and specific comments embedded in the annotated PDF in attachment.

I included the wrong supplement in my referee comments. I apologize. Attached here the correct one.

Lander Van Tricht and colleagues present the ice thickness estimations of the Grigoriev ice cap (Kyrgyzstan) collected in several field campaigns during August 2021 (summer) by using GPR technique. Then, the radar data was processed by applying the yield stress method and interpolated to produce an ice thickness layer. Finally, the authors evaluate if the global outputs resulting from 6 different experiments are able to capture the spatial patterns of the ice thickness at the local scale in the Grigoriev ice cap.

The manuscript is well structured, clear, and concise, making it easy to understand. I congratulate the authors because they have compiled a large amount of data with potential for scientific applications, however, they do not give enough detail on the statistical approach for demonstrating the unreliability of the global datasets, and it seems they remain in a visual description of the discrepancies.

The global ice thickness products were conceived as an approximation of the total volume of ice available on the Earth's surface, with its associated uncertainty. It is therefore logical to expect that their site-specific net representation will vary from site to site, depending on morpho-topographic conditions. In addition, several of the world's ice masses are inaccessible for logistical or risk reasons, in which case in-situ observations are simply not feasible. This does not seem to be the case with Grigoriev. Therefore, numerical modelling products can provide valuable complementary data to field measurements.

There is a methodological gap in this study and the authors need to work on major corrections before this manuscript can be published in TC.