Preprints
https://doi.org/10.5194/egusphere-2024-2556
https://doi.org/10.5194/egusphere-2024-2556
30 Aug 2024
 | 30 Aug 2024

Content Analysis of Multi-Annual Time Series of Flood-Related Twitter (X) Data

Nadja Veigel, Heidi Kreibich, Jens A. de Bruijn, Jeroen C. J. H. Aerts, and Andrea Cominola

Abstract. Social media can provide insights into natural hazard events and people's emergency responses. In this study, we present a natural language processing analytic framework to extract and categorize information from of 43,287 Twitter (X) posts in German since 2014. We implement Bidirectional Encoder Representations from Transformers in combination with unsupervised clustering techniques (BERTopic) to automatically extract social media content, addressing transferability issues that arise from commonly used bag-of-word representations. We analyze the temporal evolution of topic patterns, reflecting behaviors and perceptions of citizens before, during, and after flood events. Topics related to low-impact riverine flooding contain descriptive hazard-related content, while the focus shifts to catastrophic impacts and responsibilities during high-impact events. Our analytical framework enables analyzing temporal dynamics of citizens’ behaviors and perceptions which can facilitate lessons learned analyses and improve risk communication and management.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Nadja Veigel, Heidi Kreibich, Jens A. de Bruijn, Jeroen C. J. H. Aerts, and Andrea Cominola

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CC1: 'Comment on egusphere-2024-2556', Knut Seip, 02 Sep 2024
    • AC1: 'Reply on CC1', Nadja Veigel, 03 Sep 2024
      • CC2: 'REGUSPHERE-2024-2556', Knut Seip, 03 Sep 2024
  • RC1: 'Comment on egusphere-2024-2556', Samar Momin, 04 Oct 2024
  • RC2: 'Comment on egusphere-2024-2556', Anonymous Referee #2, 05 Oct 2024
Nadja Veigel, Heidi Kreibich, Jens A. de Bruijn, Jeroen C. J. H. Aerts, and Andrea Cominola
Nadja Veigel, Heidi Kreibich, Jens A. de Bruijn, Jeroen C. J. H. Aerts, and Andrea Cominola

Viewed

Total article views: 449 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
236 89 124 449 27 3 3
  • HTML: 236
  • PDF: 89
  • XML: 124
  • Total: 449
  • Supplement: 27
  • BibTeX: 3
  • EndNote: 3
Views and downloads (calculated since 30 Aug 2024)
Cumulative views and downloads (calculated since 30 Aug 2024)

Viewed (geographical distribution)

Total article views: 504 (including HTML, PDF, and XML) Thereof 504 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 08 Dec 2024
Download
Short summary
This study explores how social media, specifically Twitter (X), can help understand public reactions to floods in Germany from 2014 to 2021. Using large language models, we extract topics and patterns of behavior from flood-related tweets. The findings offer insights to improve communication and disaster management. Topics related to low-impact flooding contain descriptive hazard-related content, while the focus shifts to catastrophic impacts and responsibilities during high-impact events.