Dear Editor, dear Authors

This contribution of Brázdil et al. presents an analysis of weather-related fatalities in the Czech Republic during the period of 2000–2019. The study was very meticulously conducted and the manuscript is well organized. The manuscript is moderately well written and largely quite easy to understand. However, and as far as I can judge as a non-native English speaker, the sentence structure and punctuation is sometimes a bit special/awkward. This is a bit confusing here and there. Maybe a final proof-reading by a native English speaker could be helpful. The overall text seems a bit long to me. I make suggestions in various places on how it could be shortened slightly (if this is desired by the editors).

Presumably, data collection was very laborious, which makes this work all the more valuable. The authors have also given a lot of thought to the completeness of their data, which is also important, and discuss this point in great detail in the discussion. Furthermore, the authors show how difficult it is to compare, let alone combine, different data sources (e.g., national statistical offices or police). More attention should be paid to this point in the future. It is difficult enough for a country or region to compile databases on fatalities caused by natural hazards and weather extremes, useful synergies would be a great help.

In Europe, several similar overviews o those of the authors have emerged in recent years. The effort of the authors Brázdil et al. fits in perfectly here. This study is very important, not the least to help Czech authorities to better identify potential improvements in hazard prevention related to severe weather situations and natural hazard processes and to reduce the number of victims in the future.

I have compiled my criticisms and comments in quite a bit of detail in an attached document. Among other things, the title could be worded a bit more "crisply". Here I just want to briefly mention my clear main point of criticism, which is explained more in detail in the comments to the Methods and Discussion sections. For me, the division of the various fatalities considered here into nine classes/types is problematic. Meteorological causes and hydrological (as well as geomorphological and avalanche specific) consequences are mixed. In my opinion, this needs to be reconsidered or at least better and more extensively argued in the text.

Moreover, the importance of traffic accidents in this compilation is immense. Because (at least it seems so to the reader) as soon as the weather conditions were not optimal during an accident on the road, such a fatal accident was recorded. Here, in my opinion, the authors need to better explain how they proceeded and why.

In summary, this manuscript will definitely be of interest for the research community and therefore should in my opinion be published in NHESS. I am looking forward to it. Given the considerable number of various comments, I suggest that the paper be reconsidered after major revisions.

I provide (in an attached document) a list of comments and suggestions specific to the different sections of the article. I ask the authors to consider these.

Very important paper of sophisticated quality. Both the methodology and the literature background are valuable. Although, the paper is rather long, 47 pages, the description of the methods and results are consequent and systematic, so I could not recommend any substantial abbreviation. The Figures are also informative and well edited.

On p. 6, lines 164-167 you explain the difference between „windstorm” and „convective storm”. Were these details all sufficiently included in the documentary sources, or you applied additional meteorological information to select?

On p. 6, lines 175-181 you describe the two significance tests applied. It is not clear however, when you included the regression line into the corresponding diagrams. If both tests demonstrated significance? Please clarify it, or, even better if you decide and process according to the Mann-Kendall test, only!

Concerning the documentary sources, it would be useful to read the authors opinion and direct analyses on representativity of the long-term trend, annual cycle and distribution among the meteorological reasons established from the authors’ data base. The requested analyses could compare the above three aspects derived from the documentary and official data sources.