Disentangling the effects of Three Gorges Dam operation and post-dam channel evolution on flood stages along the middle and lower Yangtze River

Study region The middle and lower Yangtze River (MLYR), spanning ~1800 km from the Three Gorges Dam (TGD) outlet at Yichang to the East China Sea, China. Study focus Flood responses to large dams are spatially heterogeneous, while the contributions of short-term regulation and long-term channel evolution remain poorly quantified. Using the 2020 flood as a common reference, we attribute specific-gauge flood-stage change to TGD operation and channel-form evolution between the 2010 and 2020 surveys. A 2D Delft3D model with baseline, operation-only, channel-evolution-only, and combined scenarios separates the two effects and quantifies their non-additivity. New hydrological insights for the region For the 2020 flood and under a fixed-lumped-resistance assumption, the 2010–2020 channel-form change is the leading control on specific-gauge flood stages, lowering them by 0.03–1.85 m (an upper-bound, geometry-only estimate because bed coarsening is not modelled). The residual effect of TGD operation on the stage–discharge relationship is small (within 0.34 m), whereas the operation strongly reshapes the flood hydrograph: the peak discharge is reduced by up to 2.85 × 10⁴ m³ s⁻¹ and the peak stage by up to 5.29 m. The interaction term is generally small but is threshold-sensitive and is interpreted primarily as a non-additivity diagnostic. The framework, rather than the Yangtze-specific magnitudes, is transferable to other regulated rivers under comparable data and boundary-condition settings.