A Hierarchical Indicator System for Riverbank Collapse Risk Assessment: Methodology and Validation in the Jingjiang Reach, Middle Yangtze River

Abstract

Riverbank collapse is a recurrent hazard in alluvial rivers, driven by complex interactions between flow, sediment, bank materials, and human activities. Quantitative stability assessment remains challenging due to the multiplicity of influencing factors and their nonlinear couplings. This paper develops a hierarchical indicator system for bank stability evaluation. Following six selection criteria (scientific validity, discriminatory power, comprehensiveness, dynamic adaptability, mixed‑method quantification, and operational feasibility), we establish a three‑tier framework consisting of a target layer, an attribute layer, and an indicator layer. Six characteristic indicators are selected: current bank slope condition, bank geological composition, revetment integrity, near‑bank flow dynamics, near‑bank bed scour, and slope ratio change. These are grouped under three attributes – present stability state, controlling factors and evolutionary process. Scoring rules and a weight allocation scheme are derived from long-term observations in the Jingjiang Reach of the Middle Yangtze River. The overall stability score is then partitioned into three risk levels. Using topographic and hydrological data from 2020, the method is applied to 167.78 km of typical bank sections. Validation against actual collapse events recorded in 2021 demonstrates good predictive capability, with the majority of high risk sections either experiencing collapse or being identified as requiring urgent protection. The proposed system offers a practical and transparent tool for bank collapse early warning and river management.