This is a Preprint and has not been peer reviewed. This is version 1 of this Preprint.
Floods as a window of opportunity: When and why extreme events trigger adaptation
Downloads
Authors
Abstract
Disasters can open windows of opportunity for climate adaptation—moments when heightened public attention makes it possible to advance policies and measures that would otherwise struggle to gain traction. Yet whether and how floods shape public adaptation discourse remains unclear. Here, we quantify how flood severity influences media coverage on 12 adaptation measures across 274 German districts from 2000–2020, using 625,813 news articles and daily streamflow observations from 1,238 gauging stations. Through two-way fixed-effects models, we find that floods significantly increase adaptation coverage across all hazard severity levels, with extreme floods producing 26 times the effect of minor ones. However, this coverage is selective: structural measures (e.g., dikes, dams) and warning systems dominate, while financial and nature-based solutions receive proportionally less coverage, particularly during severe events. In general, post-flood discourse follows a three-phase cycle: an immediate increase in coverage to structural measures, a gradual shift toward financial recovery, and a return to baseline within 8 to 12 weeks. However, repeated floods alter this pattern. With successive events occurring in the same district or neighboring districts, discourse becomes more diverse and shifts beyond structural responses, with greater attention to financial preparedness measures. This suggests a learning process in which societies move from reliance on single solutions toward a broader portfolio of adaptation options. Together, our results demonstrate that floods create windows of attention in news media. Yet, these are systematically biased, potentially narrowing the range of adaptation options considered in the public debate.
DOI
https://doi.org/10.31223/X51R3F
Subjects
Physical Sciences and Mathematics, Social and Behavioral Sciences
Keywords
newspaper analysis, focusing events, disaster discourse, natural language processing, flood risk communication.
Dates
Published: 2026-06-08 09:08
Last Updated: 2026-06-08 09:08
License
CC BY Attribution 4.0 International
Additional Metadata
Conflict of interest statement:
None
Data Availability:
The CAMELS-DE hydrological dataset is publicly available via Loritz et al. (2024) [https://doi.org/10.5281/zenodo.13837553]. The severity-calibrated flood event dataset, the analysis panel (comprising NUTS-3 weekly article counts and flood variables), and all underlying analysis code will be made available on Zenodo upon publication.
Metrics
Views: 13
Downloads: 1
There are no comments or no comments have been made public for this article.