[General comments]

The paper overviews the history of “burning embers” comprehensively in the Introduction chapter, explains the structure of database for archiving knowledge on climate risks and communicating them with the burning-embers format in Chapter 2, exemplifies analyses in Chapter 3, and finally discuss possible contributions of the database to future risk assessments and update of the burning-embers.

I highly evaluate this paper with the following reasons in summary and hope that it is published as a reviewed article on ESSD to be read by wide range of readers.

• Timely article for the initial period of the IPCC-AR7 cycle:  Just at the initial period of the IPCC-AR7 cycle, this article will be beneficial both for researchers contributing to the assessment report as a lead author of WG2 and researchers who are willing to conduct research to be assessed properly in the report. Traceability and objectivity of the burning embers assessment have been strengthened gradually for the previous 20 years. This paper will significantly contribute to the further improvement of the RFCs and burning embers approach both from theoretical and practical aspects. Things discussed in Section 4 are describing current research gaps concisely and will send useful signals to impact projection researchers who are willing to contribute to the IPCC-AR7’s risk assessment. Researchers may also use this paper for explaining the potential value of their new research proposal to funders in the coming years.
• Potential flexibility of the proposed database structure: We are not sure how long the proposed database continues to work effectively. Key aspects or uncertainties of risk analyses may radically change in future and database for storing analyses outputs will need to be flexibly revised or extended to be continuously functional. The authors of the paper seem conscious about it and they are not selling the current design of the database as the ultimate and perfect one. I suppose the attitude will allow effective extension and improvement of the database structure in future.
• Well balanced technical documentation: This paper not only explains the technical detail of the database structure but also exemplifies how the database can be really used for storing and communicating climate risk assessments outputs in Chapter 3, that would help readers contribute to the community effort for fulfilling risk analyses.

[Specific technical suggestions]

• Table 5 (P28): From the viewpoint of decimal position, “2” in some cells should be written as “2.0”.

• 4.2.1 (P39): There is no 4.2.2 to be put in parallel here. Considering the logical flow and structure of the story, it may not be needed to be separately put as 4.2.1 but connected to the previous paragraphs (as a part of 4.2).

The development of a database on knowledge and assumptions underlying the IPCC burning embers is much needed and is likely to prove useful in understanding the evolution of risk judgments over time and in providing a critical foundation for future judgments and their comparability with previous embers. The illustrative applications of the associated climate risks ember explorer are effective. They show how analysis can use the by-now large number of embers that have been produced to draw broader conclusions about the balance of risks at different warming levels, the risk reduction possible from adaptation, and the types of risks that are relatively more or less serious.

I have no major reservations or suggestions for major revisions for this manuscript. I also find it clear and well organized.

I have two broader suggestions, and then a number of more minor comments that I have listed below, in the order they appear. The first broader comment is that the aggregation of risks across embers is useful but of course is subject to the distribution of the types of risks considered in the embers (some types of risks may be over- or under-represented). It would be useful to show early in the paper the distribution of risks considered. For example they could each be assigned to one of the Representative Key Risk (RKR) categories defined in the AR6 WG2 Ch 19, and the number of risks by category displayed in a figure. A similar figure could be made for risks by world region. Some of this categorization occurs in figures 7 and 8, but for a different purpose and it comes late in the paper.

The second comment is that I find the discussion section to be somewhat long and delving a bit further than necessary into topics that are related but not central to this paper. That includes some of the discussion of adaptation framing, limits to adaptation, and the final section on the future of burning embers. All of these sections are relevant to some extent, however, so there are arguments to keep them. The authors might consider ways to make them somewhat more concise.

Minor comments:

line 15: "due to the colours used" -> "with risk judgments reflected by the colours used"

line 18-19: it should be specified what time period the database covers, that is, all embers created since the TAR

line 47: "temperature" -> "global average temperature"

line 49-55: It is probably worth mentioning here that the burning embers diagram did not appear in the IPCC Fourth Assessment Report, although the Reasons for Concern were re-assessed in the text. Instead, the BE diagram appeared in a paper (Smith et al.) that appeared in PNAS after the report was published. This is mentioned later in the paper, but it seems also appropriate here.

line 51: "with four discrete risk levels" -> "with four (rather than three) discrete risk levels"

line 61-62: The shift from "key vulnerabilities" in AR4 to "key risks" in AR5 was not only a matter of a change in terminology. AR4 was somewhat murky in the distinction between vulnerability and risk, sometimes explicitly acknowledging their differences while at other times seeming to substitute one for the other. AR5 clarified the conceptual framework for risk, focusing the RFC assessment appropriately on risk as the ultimate outcome of interest. I suggest changing the wording from "which were later referred to as “key risks”" to "; AR5 refocused this approach on 'key risks'".

line 85: "United Framework" -> "United Nations Framework"

line 88: by saying this paper "proposes to structure a database" it is unclear whether such a database has been created yet; this shouuld be stated more definitively.

line 98: the wording "quantitative estimates of how risk increases with the level of climate change" can be interpreted as estimates from a variety of studies of impacts relevant to a particular ember, which underlie the expert judgments made (eg, various estimates of the damages from flooding that might underlie the RFC related to extreme events). Since this is not what you intend, I suggest changing to "quantitative estimates of the global average temperature at which the risk for a given ember changes from one level to another".

line 172-175: This is a useful finding: "While getting the numerical data to reproduce the embers has become easier in the recent IPCC reports, it remains difficult to get a synthetic description of the risks illustrated in each ember and an explanation for each risk transition. This information is rarely associated with the quantitative data and was not always collected in a systematic way."

line 256: Section 2.3 on Adaptation levels and scenarios: This section describes well the difficulties of conflating adaptation assumptions with SSP scenarios (which do not include adaptation by definition), which the database appears to do by lumping them into the same field of information. However the section does not present a solution to the challenge described. I don't think there is a good reason to record adaptation levels and scenario in the same field; it excludes for example the possibility of having an SSP3 with high adaptation vs an SSP3 with low adaptation. A better solution here is needed; I suggest the scenario and adaptation assumptions should be separated.

p. 15: Table 3: the caption should include a clear and comprehensive description of the color scheme used for shading rows.

line 353: Section 2.5.3: an excellent plan for collaboration on completing the information in the database.

line 471: in this figure caption, I found the explanation of panel (c) hard to understand. I suggest changing "indicates the fraction of assessed embers, at each GMT, for which the risk is above the midpoint within each transition" to: "indicates the fraction of assessed embers for which a given GMT exceeds the midpoint of each of three risk transitions"

p. 26: Table 4 is quite useful in illustrating how the assessment of the database of embers can yield useful information about risks at the low or high end of the distribution. 

p. 33-34: I am not sure figures 7 and 8 work very well. They are very hard to understand, particularly the lines that connect different results across adaptation levels. I believe that these are supposed to be interpreted that each line represents a separate result: what is the change in risk at the same warming level but with different adaptation assumptions. However since many of the risk judgments overlap, there are multiple lines that all look like they are connected to each other and one doesn't know what to do with that. Maybe each line could be made into an arrow, so that they appeared more separate than connected?