Enhanced Urban Flood Hazard Assessment by Stochastic Event Catalog

Abstract Assessing flood severity in urban areas is a pivotal task for urban resilience and climate adaptation. However, the lack of in situ measurements hinders direct spatial estimation of flood return periods, while conventional assumptions about rainstorm‐flood consistency introduce significant...

Full description

Saved in:
Bibliographic Details
Main Authors: Haochen Yan, Kaihua Guo, Mingfu Guan
Format: Article
Language:English
Published: Wiley 2025-08-01
Series:Water Resources Research
Subjects:
Online Access:https://doi.org/10.1029/2025WR040459
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Assessing flood severity in urban areas is a pivotal task for urban resilience and climate adaptation. However, the lack of in situ measurements hinders direct spatial estimation of flood return periods, while conventional assumptions about rainstorm‐flood consistency introduce significant uncertainties due to rainstorm spatiotemporal variability (STV). This study proposes a novel framework that utilizes multivariate frequency analysis of flood variables at the street level (50 m) through a stochastic rainstorm‐flood event catalog. The rainstorm events in the catalog are generated by a random field generator and resampled to match the joint distribution of STV variables consistent with radar observations. Urban flood processes are then simulated by a hydrodynamic model for flood hazard assessment (FHA). We applied the framework to a rural‐urban watershed using 3,000 cases randomly resampled from the catalog. Results reveal that inundation characteristics respond more rapidly to increasing rainfall intensities than downstream flood peaks, particularly during the early stages of rainstorms. The complex joint probability structures of rainstorm severity and STV variables obscure the mechanistic control of individual factors on flood response. A significant underestimation of street‐level flood hazards occurs when assuming the same return periods (RPs) as those for watershed‐level hazards. The inconsistency between rainstorm and flood severities results in widespread underestimation of street‐level flood hazards in upstream regions, while traditional storm designs that neglect STV lead to overestimations in mid‐ and downstream areas. This study highlights the complex probabilistic behavior of spatially distributed flood hazards across multiple scales, enhancing the insights and methodologies for street‐level FHA.
ISSN:0043-1397
1944-7973