Modelling Flood‐Induced Riparian Vegetation Disturbance and Hydro‐Morphodynamic Feedbacks

Abstract

Floods reorganize gravel‐bed rivers through coupled hydrodynamics, sediment transport and vegetation mortality. However, event‐scale morphodynamic predictions remain limited when vegetation loss is represented by a single, simplified mechanism. Here, we address how explicitly representing multiple, concurrent flood‐induced vegetation disturbance processes and their feedbacks can improve the physical realism of vegetation–hydro‐morphodynamic interactions during floods. We propose an integrated framework for modelling flood‐induced vegetation disturbance processes, including mechanical failure through overturning or breakage, followed by downstream washout, burial beneath deposited sediment or persistence in a failed but nonremoved state. To evaluate the model, we compared four configurations: Case 1 (proposed framework), Cases 2 and 3 (with different postfailure mechanisms than Case 1) and Case 4 (a static vegetation case in which flood‐induced disturbance was not considered). These four configurations were incorporated into a two‐dimensional hydro‐morphodynamic model. The framework was applied to the 2019 flood of the Chikuma River, Japan and evaluated against preflood and postflood topography and vegetation distribution data. The proposed framework successfully captured the spatial heterogeneity of vegetation removal by explicitly simulating the co‐evolution of riverbed change and vegetation disturbance processes. Furthermore, the spatial distribution of washout‐dominated vegetation removal and burial‐dominated disturbance were closely associated with bar morphology. These event‐scale feedbacks, whereby vegetation disturbance directly alters flow resistance and morphodynamic evolution, are critical for reliably predicting embankment erosion and associated infrastructure risk. Future work should extend this framework to multi‐event sequences and diverse river planform types to further validate its generalizability.