Please see my comments in the attached file

The manuscript presents a critical review of the use of social media data in the context of disaster management, reviewing a total of 250 articles from a range of disciplines. The paper is well-written and offers a valuable contribution to the growing literature on social media in disaster contexts. 

The paper would benefit from a clearer focus on its core research questions and a more concise presentation of the findings. The current version includes a significant amount of descriptive content, which could be streamlined to better highlight the most important insights.

Major Comments:

1. Length and focus of the study: The manuscript is quite long (44 pages, including 19 figures and 4 tables). While comprehensive, much of the content is descriptive and does not directly address the core research questions. I recommend moving some of the less critical descriptive sections (particularly in Sections 3.3, 4.2, and 4.4) to the appendix. For me, the most insightful sections were 4.3 and the discussion in Section 5. A focus on these could improve the impact of the paper.

2. Clarification of research questions and key insights: The paper could more clearly articulate its central research questions and ensure that the findings directly address them. In particular, I found the second research question on the “actionable information” derived from social media more relevant and of interest to a broader audience. In contrast, the first question on “exclusion criteria in relevance filtering” is more technical and may be of limited interest to non-specialist readers.

3. Practical examples in the introduction: The introduction would benefit from one or two concrete examples illustrating the kind of information that can be derived from social media in disaster contexts and how these tools have been used by researchers. This would help orient readers who are less familiar with the data sources or their potential applications.

4. Keyword selection: The search strategy seems to have focused on Twitter, with no mention of other major platforms such as Facebook, Instagram, or Weibo. This is particularly surprising given that some of these platforms have significantly higher user bases and broader geographic reach. They have also been extensively used in disaster contexts (e.g., Facebook). The rationale for this focus should be clearly explained, and the implications of this potential bias should be acknowledged more explicitly in both the methodology and the interpretation of findings (e.g., when the authors find that the great majority of studies in their database uses Twitter as noted in line 338 and elsewhere).

5. Inclusion of studies not using social media: Section 3.3.4. describes the data used in studies distinguishing whether article have utilized social media data or not. It is unclear why studies not using social media were included in the database, given the paper’s stated focus on social media. Of course, the authors could use these other articles as reference points to show the advantages or disadvantages of the use of social media, but this does not really happen in the analysis. Instead, when describing their database, the authors also refer to those articles not using social media data, which may confound the analysis. 

6. Stronger emphasis on key messages: While the paper shows great technical detail, the manuscript would benefit from a more focused discussion of key takeaways. Specifically, what are the major advantages, disadvantages, and challenges of using social media data in disaster contexts? These insights could be more prominently featured in the abstract, conclusion, and discussion sections (see also my comment 2 on the research question above).

Minor Comments:

7. Line 35: The sentence “Existing literature reviews on social media data (SMD) platform evaluations, data collection tools, and analysis methods over time” is missing a verb.

8. Line 176: Typo in “The may however be other keywords...” 

9. Line 180ff: The authors note that relevant studies may have been missed. Would it not have been advisable to revise the keyword strategy to more comprehensively capture the literature?

10. Line 288: Missing word: “Such data is considered credible when compared to the data extracted from social media platforms as it may include false information.”

11. Line 557: The phrase “affected more than 20 lakh people” may be unclear to international readers. Maybe better to replace with “2 million people” 

12. Section titles: Some subsection titles could be made more meaningful and specific. For example, “4.2. Early Works” could clarify the time period; “4.2.2. Previous Works on Social Media Analytics” could clarify what is meant with “previous”, etc.