Abstract: Social media data, as a type of spatiotemporal meta data, possesses real-time and location-based characteristics, and has been widely utilized in flood risk management abroad in recent years. However, the research on this data type in flood risk management in China is less, and related applications are limited. Through reviewing the research on flood disasters based on social media data both domestically and internationally, this paper explores methods for extracting and analyzing flood risk information. It also identifies directions for applying this information in flood risk management, including flood disaster monitoring and early warning, spatiotemporal distribution of disasters situation, sentiment analysis, response behavior analysis,disaster relief deployment, and damage assessment. Social media data has the advantages such as real-time updates, diverse data types, and visual representation, but it also faces challenges such as limited quality and accuracy, excessive false information,limitations in user groups, and cultural differences. Harnessing the strengths of social media data while addressing the shortcomings of traditional monitoring methods is of paramount importance for urban emergency management departments in China to efficiently respond to flood disaster risks.