[Objective] This study aims to investigate the effects of different vegetation arrangements (parallel and staggered) in the junction zone between channel and floodplain in urban rivers on the spatial distribution of bed shear stress, and to reveal how vegetation length and arrangement influence hydrodynamic characteristics, thereby providing theoretical support for ecological revetment design. [Methods] A two-dimensional hydrodynamic model was established using Delft3D-FM and validated with measured water levels and discharges. To account for the effects of vegetation, a vegetation module was incorporated into the hydrodynamic model. The model considered plant height, width, and density, with vegetation resistance simplified as bed roughness. Vegetation zones with lengths of 0.5L, 0.75L, L, 1.25L, and 1.5L(L represents 1 000 m vegetation zone length) were arranged in parallel and staggered patterns in the channel-floodplain junction zone. The two-dimensional hydrodynamic model that accounted for vegetation effects was used to simulate the flow fields under different conditions. Based on the simulation results, the distribution characteristics of bed shear stress in the vegetation zone and its downstream region were analyzed. The effect of vegetation on hydraulic resistance was evaluated using the blockage factor, dimensionless hydraulic radius, and surface area blockage factor (characterizing vegetation zone length). Finally, a dimensionless hydraulic radius function considering the effects of vegetation was proposed to predict the maximum bed shear stress. This function was introduced to quantitatively characterize the influence of vegetation. [Results] (1) In parallel arrangement, vegetation dominated the flow dynamics in the junction zone. The resulting shear stress zones extended from the main channel within the vegetation zone to its downstream end, forming elongated stress zones downstream of the vegetation zone. However, the stress field patterns varied with vegetation zone length, with significant differences observed in the maximum shear stress distribution. In longer vegetation zones, the location of maximum shear stress tended to shift farther downstream from the end of the vegetation zones. With a vegetation length of 0.5L, the maximum shear stress zone was located 0.25L-0.30L downstream from the end. When the vegetation length was L, the maximum stress zone nearly coincided with the downstream end. With a length of 1.5L, the maximum stress zone was 0.25L-0.5L downstream from the end. As vegetation length increased, the location of maximum shear stress zone in the main channel shifted upstream, showing a tendency to move away from the downstream end of the vegetation zone.(2) In staggered arrangement, the shear stress in the main channel reached its maximum within the staggered zone. Under the influence of the bend, the maximum cross-sectional shear stress shifted from the convex bank to the channel center. This indicated that vegetation reduced the effect of centrifugal forces on secondary flow and shear stress in the bend, with the maximum shear stress consistently occurring at the upstream face of vegetation zone. The shear stress in the main channel readjusted according to the vegetation zone distribution and then stabilized to meet the spatial variation of bed shear stress. The bed shear stress varied significantly with the length of the vegetation zone. Regardless of the vegetation zone length, the bed shear stress peaked at the cross-section adjacent to the vegetation units. Notably, the bed shear stress in longer vegetated zones was significantly lower than that in shorter ones at this cross-section. [Conclusion] Vegetation arranged in parallel pattern tends to form larger shear stress zones near the vegetation end and elongated stress zones downstream, with longer vegetation bringing stress concentration zones closer to vegetation zones. Vegetation arrangement and length significantly affect bed shear stress distribution by altering flow structures, with staggered arrangement forming large stress zones at staggered zones and the cross-sections of adjacent vegetation units, where longer vegetation zone results in smaller stresses. The proposed dimensionless hydraulic radius function proves effective for predicting the maximum bed shear stress, providing a basis for optimizing vegetation arrangements in channel-floodplain junction zones.