A hundred ways to get flooded: systemic variation in disaster preparedness and response

Da Costa, J., Cloke, H. L. ORCID: https://orcid.org/0000-0002-1472-868X, Neumann, J. ORCID: https://orcid.org/0000-0003-3244-2578 and Salvidge, N. ORCID: https://orcid.org/0000-0002-9716-5156 (2026) A hundred ways to get flooded: systemic variation in disaster preparedness and response. International Journal of Disaster Risk Reduction, 143. 106321. ISSN 2212-4209 doi: 10.1016/j.ijdrr.2026.106321

Abstract/Summary

Flood early warning systems have improved considerably in hazard detection, forecasting, and warning dissemination, yet a persistent gap between warning availability and protective action remains. This paper examines how flood warning information leads to action in municipal preparedness and early response, asking why municipalities act differently when operating within the same system. Drawing on a nationwide cross-sectional survey of 46 municipal administrations (of the 100 municipalities nationally) in Luxembourg, a setting in which nationally standardised warnings are received directly by municipalities without an intermediate administrative tier, we analyse how organisational arrangements condition whether the interpretation of multi-hazard warning information results in action. Findings show that municipalities do not enter warning situations from equivalent starting points. Although the survey was designed as multi-hazard, flooding was the primary concern for 93.5% of municipalities. Most municipalities lack formal intervention plans (69.6%) and predefined warning thresholds linking alert levels to specific measures (67.4%), creating uneven pre-warning institutional conditions across the municipal landscape. Right across this landscape, warning information is not used as a direct trigger for action but assessed in relation to local conditions, in which local relevance, expected impacts and observed conditions are assessed before measures are defined. This interpretive process is consistent across municipalities, with most reporting use of multiple information sources (97.8%) and local observation (76%) to assess warnings. However, interpretation does not lead to action with the same frequency. Frequent action is observed exclusively in the few municipalities that have both a documented plan and a predefined threshold, while those without either element report the lowest action frequency. These findings indicate that institutional arrangements are associated with how often similar interpretive practices lead to action. The current governance structure, in which municipalities retain responsibility for preparedness and early local measures while operational emergency response is centralised under CGDIS (the national fire and rescue service), favours confirmation-based responses over anticipatory activation. Standardising warning signals does not standardise municipal response: one warning system produces multiple pathways through which municipalities fulfil their statutory duties for preparedness and response because warning information becomes operational within uneven institutional contexts.

Item Type	Article
URI	https://centaur.reading.ac.uk/id/eprint/131018
Identification Number/DOI	10.1016/j.ijdrr.2026.106321
Refereed	Yes
Divisions	Science > School of Archaeology, Geography and Environmental Science > Department of Geography and Environmental Science Science > School of Mathematical, Physical and Computational Sciences > Department of Meteorology
Publisher	Elsevier
Download/View statistics	View download statistics for this item