Evaluating the French Flash Flood Warning System Using Hydrological and Impact Data in Southeastern France

ABSTRACT Due to climate change and rapid urbanisation, flash floods are becoming increasingly threatening, making real‐time flash flood monitoring and early warning systems a crucial adaptation measure. In France, the Vigicrues Flash system, operational since 2017, uses a simplified rainfall‐runoff...

Full description

Saved in:
Bibliographic Details
Main Authors: Juliette Godet, Olivier Payrastre, Qifan Ding, Julie Demargne, Eric Gaume, Pierre Nicolle, Anne Belleudy, Pierre Javelle
Format: Article
Language:English
Published: Wiley 2025-06-01
Series:Journal of Flood Risk Management
Subjects:
Online Access:https://doi.org/10.1111/jfr3.70053
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:ABSTRACT Due to climate change and rapid urbanisation, flash floods are becoming increasingly threatening, making real‐time flash flood monitoring and early warning systems a crucial adaptation measure. In France, the Vigicrues Flash system, operational since 2017, uses a simplified rainfall‐runoff model to issue short‐range flood warnings for small‐to‐medium ungauged basins, covering about 10,000 municipalities. Seven years on, we assess its performance on a large spatio‐temporal scale, using a multi‐indicator approach. First, hydrological validation is conducted using discharge data from gauging stations and post‐event campaigns, applying a contingency‐based approach with flood thresholds from 2‐ to 1000‐year return periods. Results show promising flood detection rates (mean CSI: 40%–70%) and indicate that expanding the system's coverage is feasible without reducing performance. Next, we evaluate its capacity to detect damaging events using the French CatNat database of natural disaster decrees. Finally, we document a survey of system users to capture their perception of its effectiveness and operational value. However, the high number of false alarms in the impact‐based assessment highlights the urgent need for impact‐oriented forecasts and warnings.
ISSN:1753-318X