Diverging hydrometeorological drivers of flashier floods in a warming climate

Journal article

The monitoring and forecasting of flash floods present significant challenges due to their rapid onset. While extreme rainfall events, a primary driver of flooding, are becoming flashier, there is no consensus on whether floods will exhibit similar trends in a warming climate. Here, we assess flash flood risk development over four decades across 741 basins globally using the flashiness index. Our findings reveal a significant shift towards flashier floods in nearly one-third of the basins, predominantly concentrated in cold and arid regions of the northern mid-latitudes. In these regions, the increase in flashiness is largely associated with larger flood peaks rather than changes in timing. We identify basins characterised by rising flood peaks, which are likely to require flood defence infrastructure capable of withstanding more severe flood events, as well as those marked by shorter onset times, which may face challenges in implementing short-term early warning systems. Specifically, the increase in flashiness in most North American basins is mainly associated with rising flood peaks, whereas in most European basins, it is dominated by shorter onset times. These findings underscore the urgency of implementing region-specific strategies to adapt to flashier floods in a warmer future.

Authors: Zhou, F; Wang, S; Slater, L; Lin, P; Yang, H

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Elsevier
• Journal: Journal of Hydrology More from this journal
• Volume: 679
• Issue: Part A
• Article number: 136163
• Publication date: 2026-08-03
• Acceptance date: 2026-08-01
• DOI: 10.1016/j.jhydrol.2026.136163
• EISSN: 1879-2707
• ISSN: 0022-1694
• Language: English
• Keywords: flash flood; explainable machine learning; climate change