Morphodynamic responses to the 2000 Yigong dam-break flood: Insights from back-analysis and cross-scale modelling challenges

Landslide Dam-Break Outburst Floods (LDBOF) are devastating natural hazards that drastically reshape downstream river morphologies. However, their inaccessibility, high risk of equipment loss, and sparse field data collection severely hinder hazard understanding and timely warning capabilities. The 2000 Yigong LDBOF event in China is one of the most significant modern recorded cases, yet it suffers from limited observational data. To address this gap, we integrated multi-source data—including open-source elevation datasets, literature-derived records, satellite-based flood inundation extents, and direct field observations—to develop a comprehensive input dataset for hydro-morphodynamic modeling of the event. Model validation against field observations and comparable studies confirmed the reasonableness of simulated lake emptying, dam breaching, flood inundation, bank erosion, and channel infilling processes. Our results reveal key morphodynamic characteristics of the Yigong LDBOF: dam material transport was dominated by translational motion during the flood rising stage and dispersive transport during the falling stage. The outburst flood peak discharge reached ~60 times that of typical meteorological floods, significantly amplifying the effects of river width on dam material transport. We further proposed a sediment transport equation that incorporates the regulatory effect of large boulders. Post-event channel recovery simulations, validated with remote sensing data, indicated minimal planform changes, with bed incision driven by headward erosion as the dominant morphological adjustment. Large boulders acted as a stabilizing factor, limiting upstream erosion and forming sediment supply-limited reaches. This study provides a robust multi-source data integration and modeling framework for LDBOF events with sparse observations, offers new insights into cross-scale hydro-morphodynamic processes of extreme floods, and the proposed sediment transport equation improves the accuracy of simulating boulder-influenced sediment dynamics—supporting hazard risk assessment and downstream river management for future LDBOF events.