Overall Quality

The research objective of the submitted manuscript is an important and understudied topic area, particularly in light of increasing exposure to climate-change associated natural hazards and disasters. However, the authors have not included the necessary level of detail in their methods description to assess the overall quality of the submitted research. The text of the manuscript frequently contains ambiguous and nonspecific language and is difficult to interpret. This paper would likely greatly benefit from a grammatical editor as well as enhanced details on the selected methodology. Additionally, this manuscript would benefit from a limitations section which includes (at a minimum) an assessment of the representativeness, accuracy, and precision of survey data collected for this research, critical assessment of the validity and reproducibility of the weighting criteria, an evaluation of how generalizable the data and conclusions are to the selected communities and target population, and a discussion of the benefits and limitations of creating an individual-level index compared to a community-level index of social vulnerability.

I have focused my review primarily on the area of my expertise (i.e., epidemiology, research methods, and area-based social vulnerability indices).  

Specific Comments

If possible, indicate the four community types (Type 1 - Type IV) in the Wuhan and Hongshan District map (Figure 2)

Including a population estimate in Table 1 would provide important context for interpreting how the sampling fraction contributes to the generalizability and precision of survey data.

In the methods section, the authors should clarify how the data were "weighting" in this context. For example, are you weighting the survey data to be representative of the target community and for non-response bias or are you applying a weighting algorithm to give more importance to certain measures of vulnerability?

Page 9, lines 27-31: This main idea of paragraph is ambiguous. Are the authors arguing that migrants are de-stabilizing the adaptive capacity of a community or are the authors highlighting the reduced adaptive capacity of migrant individuals?

Section 3.2 (Determination of weight): This section needs much more detail. Describe how you selected experts, the response rate, demographics of experts, methodology for eliciting ratings including the modality (in person, by phone/email, etc.), what prompts did you use, what scale did you provide, how did you define each of the 15 indicators for the experts, etc. If the AHP method you used followed a standard procedure, please at least include a citation. Define what first-level and second-level index means. Describe how you "check the consistency of the judgment matrix". Include an interpretation of CR, CI, and RI for the reader.

Section 3.2 (lines 24-26, Calculation of Final Weight): This section requires much more detail. I don't understand what the authors are referring to in Step 2 (final weight) if Step 3 describes how they calculated the final weight.

Section 3.3. (Data collection and analysis): This section is lacking in important details needed to evaluate the quality of the data generated from the survey. The authors should describe how they determined a minimum sample size, how they constructed a sampling frame, and if they stratified communities based on demographic characteristics (e.g., migrant status). The authors should also include a description of the survey questions used to assess social vulnerability factors, how the survey was administered to respondents, and the refusal/non-response rate?  The authors do not indicate that the survey data was weighting to account for demographic differences and/or non-response. The lack of weighting seriously undermines the generalizability and validity of the survey data.  Without any indication that the data collected are representative of the underlying community, it is inadvisable to extrapolate the results beyond the sample of individuals included.

Figure 5: This figure is difficult to visually interpret. I suggest using a more simplified chart to display the distribution of each of these dimensions separately.

Table 5: It is more informative to show the percent of individuals in each community type that were high/mid/low vulnerability than the percent of individuals in each vulnerability category that lived in each community. For example, you show that 61% of high-vulnerability individuals lived in Type IV communities. However, only 27.6% of individuals who lived in Type IV communities are classified as high-vulnerability.

Figure 6: Trend lines for each plot in scatter plot matrix would help with interpretation

Page 17, lines 5-6: The authors state that, “The disparity in social vulnerability among inhabitants in various neighborhoods implies “residential segregation” in the metropolitan environments. However, their previous statement appears to contradict this conclusion: “A previous study (Turner et al. 2003) found that not only do social vulnerabilities vary between societies, communities, and groups, but also among residents in the same area/community. We have verified that using quantitative analysis receives similar findings (see Figure 5).”

Page 19, lines 20-21: I believe that the authors are implying that occupation, household registration, gender and debt cannot reflect the variations in individual social vulnerability because there are no natural quantitative hierarchies to these factors. If so, that argument should be made more explicit. However, I would argue that it would be informative to show the proportion of individuals within each vulnerability group that belong to a specific vulnerable group (e.g., percent of workers employed in low-skill occupations, percent of individuals without household registration, etc.).

Page 20, lines 22-23: It is unclear the conclusion the authors present (“Although there are also some low-vulnerability individuals with the rural household registration, it can be argued that they are mainly engaged in state-owned enterprises, including public service units”). is supported by the survey data or if this is a hypothesis extrapolated by the authors.

Technical Corrections

Page 16, line 20: The author indicate that communities of Type III have fewer scores than those of Type IV in terms of exposure and adaptive capacity, higher in sensitivity. I believe that they might have meant lower rather than fewer scores.

Overall quality

This is a well-written, easy-to-understand paper. This study presents new data on the evaluation of social vulnerability. This study establishes indicators from a micro-level perspective to assess social vulnerability and conduct an assessment of social vulnerability among the population of Hongshan District, Wuhan City, China, via individual questionnaire surveys. Additionally, the study utilized K-means cluster analysis to compare diverse community types and identify susceptible groups. The results offer valuable insights into the connections between distinct community types, physical and built environments, and levels of social vulnerability to disasters. This study contributes to the realm of disaster risk reduction by providing a quantitative comprehension of the variation in the extent of social vulnerability among distinct communities and populations. These findings provide insights into the associations between various community types, physical and built environments, and levels of social vulnerability to disasters, which could be valuable for policymakers and practitioners in the field. However, it has limited novelty and contribution, at least in its presentation.

Specific comments

1. Line 15 : Vulnerability is a key concept for both disaster risk and climate change adaptation. By analyzing the potential factors causing losses, it is possible to predict the extent to which a disaster will impact society in the future (Vincent 2004). The author mentions "factors contributing to losses"; are they referring to the concept of "root causes of a disaster"? Further clarification of this matter is required.
2. Line 34 : Social vulnerability is influenced by various factors beyond social and economic status. There are as well as political conditions that affect an individual's or group's position and power in society and additionally, people's level of vulnerability may differ based on their life circumstances, age, and the time of year. Why the study did not consider the potential interactions between different social vulnerability indicators, which may affect the overall level of vulnerability?
3. Line 13 “ At the same time, the results also show that about 50% of urban registration holders are also at high and medium levels of social “ Despite the quantitative results, did the author examine/documented urban social vulnerability from a more optimistic viewpoint, such as the innovative use of existing neighborhood groups for preparedness or the utilization of hazard and vulnerability mapping? Additionally, did the author investigate cases of excellent coordination between municipalities and NGOs/CBOs regarding improvements in risk communication or increased sensitivity to the needs of population, both legal and illegal?
4. Why did the study not consider the potential role of cultural and social factors in shaping social vulnerability and disaster risk?
5. Did the author recognize any limitations of this study? If so, it may be advantageous to incorporate these limitations in the manuscript.

Technical corrections

Given the dynamic nature of vulnerability, it would be advantageous to delineate a timeline that specifically identifies periods of heightened vulnerability over the course of the year, particularly in relation to the influence of hazards. Such an approach would enable a more comprehensive appreciation of the "mosaic" of vulnerability within the research site.