The author has evaluated an existing method, Šoljan et al. (2024), for the retrival of cloud top height (CTH) using the 11 micro meter InfraRed channel, and used the retrieved CTH information for correcting the viewing parallax of Deep Convective Clouds (DCCs). The topic is very relevant due to the fact that parallax correction becomes more and more important with the ever-increasing resolution of satellite imagery. However, the study either lacks new information or it has not been clearly stated.

In particular:

(1) It is not clear if CLAAS-3 is a reliable reference against which the CTH retrievals with Solijan et al. 2024 can be compared. Especially because CLAAS-3 retrieves CTH using a Neural network based algorithm and does not represent ground truth. Why not use ground-based ceilometer measurements as reference ?

(2) Why is ICAO standard atmosphere, which is a static profile under an idealized condition, used for converting cloud top pressure to height ? Could it be more accurate or more closer to reality when profiles from NWP or Reanalysis are used ?

(3) There is a lack of reliable ground truth for assessing the accuracy of the parallax correction using the different CTH sources. Would it make sense to look at some surface parameters such as precipitation, direct solar irradiance, lightning, etc. that are directly impacted by the location of the DCC ?

(4) Section 5.1. Lines 350-355. The mean displacement vector length for parallax may not be very informative. As the author himself noted, parallax displacement is heavily dependent on the viewing satellite zenith angle (SZA). Therefore, it could be more intuitive to have the mean displacement vector and standard deviation for different SZA ranges.

The author evaluated a moist-adiabat cloud-top height retrieval for parallax correction of deep convective clouds. The topic is within the scope of Atmospheric Measurement Techniques. Overall, the clarity and rationality of the Methodology and Validation section need to be improved. My major concerns are as follows.

1、Lines 187-189: The description of the DCC definition is methodologically incomplete. The author reviews several BTD thresholds from the literature but does not clearly state which threshold(s) are actually applied in this study, nor how the regional differentiation (e.g., tropics vs. mid-latitudes) is implemented. Similarly, the CAPE threshold of 100 J kg⁻¹ is introduced as “arbitrarily selected”. Please specify: (1) the exact BTD threshold(s) and the spatial domain to which each applies; (2) whether the BTD and CAPE conditions are combined with “AND”; and (3) a justification or sensitivity check for the CAPE threshold.

2、Figure 1: Please verify the validity of the data presented in the figure. The statistical metrics shown (e.g., correlation coefficients) do not appear to be consistent with the visual results in the figure, nor do they align with the discussion in the main text.

3、The author stated that “after applying a simple regression-based bias adjustment, the absolute error is reduced to below 1.2 km.” However, it should be noted that this regression model was trained on a specific sample from 2007. From my perspective, the adjustment was not dynamic. If this regression correction is applied directly to historical Meteosat MVIRI data from four decades ago, how robust is it? Is there a risk that the regression formula artificially smooths out genuine physical signals?

4、For V24-based CTH estimates from SEVIRI measurements, why did the author choose to compare the results with CLAAS CTH, which itself also has inherent biases in CTH retrieval? The direct comparison with CALIPSO/CloudSat CTH products would be more reliable.

5、The evaluation logic in Section 5.1 is confusing. The author compared the VŠ24 cloud top height (CTH) retrievals against three different reference datasets (SEVIRI, MODIS, CloudSat/CALIPSO), and then evaluated the calibrated results against CLAAS dataset using parallax displacement vectors. The CLAAS-derived parallax displacements were regarded as ground truth without discussing the uncertainties inherent to CLAAS CTH products. I recommend that the author needs to better elaborate: (1) justify why CLAAS is a suitable evaluation benchmark, including its uncertainty, and why not CALIPSO; and (2) detail the bias adjustment method explicitly.

6、Author states that “Despite a ~10% underestimation of CTH, the impact on parallax correction was minimal: 84% of pixel geolocations coincided with those derived from CLAAS-based corrections, increasing to 97% when CLAAS retrieval uncertainty was accounted for.” Satellite viewing zenith angles (VZA) over Europe are generally large due to its higher latitude, which should amplify the effect of height errors. Why does such height error lead to a small pixel misalignment after parallax correction?