Zhang et al. (2026) Spatiotemporal evolution and driving mechanisms of superposition effects of rainfall-induced urban waterlogging and levee breach flooding

Identification

• Journal: Journal of Hydrology
• Year: 2026
• Date: 2026-10-01
• Authors: Ye Zhang, Xin Su, Qin Guanghua, Leizhi Wang, Jinhui Li, Lingjie Li, Yintang WANG, Xiting Li
• DOI: 10.1016/j.jhydrol.2026.136522

Research Groups

• State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, Chengdu 610065, China
• Nanjing Hydraulic Research Institute, National Key Laboratory of Water Disaster Prevention, Nanjing 210029, China
• Harbin Institute of Technology (Shenzhen), Shenzhen, Guangdong 518055, China
• Hohai University, Institute of Water Science and Technology, Nanjing 211164, China

Short Summary

This study developed an integrated explainable machine learning framework to quantify and attribute the nonlinear superposition effects of coupled riverine–pluvial flooding hazards, revealing their spatiotemporal evolution and driving mechanisms in urban areas. The findings demonstrate pronounced nonlinear amplification effects under high recurrence intervals and highlight the shifting dominance of drivers from breach dynamics to topography in extreme scenarios.

Objective

• To quantify and attribute the nonlinear interaction effects, hazard sources, and underlying driving mechanisms of "Pluvial flooding – Levee breach – Urban waterlogging" coupled riverine–pluvial flooding hazards using an integrated explainable machine learning framework.

Study Configuration

• Spatial Scale: Urban polder area (Yundong polder area, Wuxi City, China).
• Temporal Scale: Recurrence intervals of 20 to 200 years.

Methodology and Data

• Models used: Hydrodynamic modeling, Flood Superposition Index, Flood Superposition Increment, coupled riverine–pluvial flooding hazard interaction transfer matrix, XGBoost-SHAP approach.
• Data sources: Scenario-based simulations (single-rainfall, single-levee-breach, and coupled riverine–pluvial flooding hazard scenarios).

Main Results

• Coupled riverine–pluvial flooding interaction is not a simple linear combination but exhibits pronounced nonlinear amplification effects under high recurrence intervals.
• Superposition effects show substantial spatiotemporal heterogeneity, characterized by hazard lag effects, severe inundation near the breach, strong amplification at the impact margin, and shifts in hazard attributes with increasing recurrence intervals.
• Compound superposition effects are jointly governed by topography, distance to the breach, pluvial inundation depth, and breach inundation depth, with dominant drivers gradually shifting from breach dynamic factors to topographical factors under extreme scenarios.

Contributions

• Proposes an integrated framework using explainable machine learning to quantify and attribute the hazard superposition effects of coupled riverine–pluvial flooding.
• Reveals the spatiotemporal evolution and driving mechanisms of "Pluvial flooding – Levee breach – Urban waterlogging" coupled riverine–pluvial flooding hazard interactions.
• Provides insights supporting urban flood risk management and mitigation by characterizing nonlinear interactions and dominant drivers.

Funding

• Not specified in the provided text.

Citation

@article{Zhang2026Spatiotemporal,
  author = {Zhang, Ye and Su, Xin and Guanghua, Qin and Wang, Leizhi and Li, Jinhui and Li, Lingjie and WANG, Yintang and Li, Xiting},
  title = {Spatiotemporal evolution and driving mechanisms of superposition effects of rainfall-induced urban waterlogging and levee breach flooding},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.136522},
  url = {https://doi.org/10.1016/j.jhydrol.2026.136522}
}