Introduction: Submerged vanes are flow-pattern altering structures that are mounted vertically on channel-bed at a small angle of attack to the approach flow. A submerged vane generates a secondary circulation (a spiral flow), due to the vertical pressure gradients on the two sides of the vane, which originates below the top elevation of the vane and extends in the downstream of the vane. The vane-induced vortex redistributes sediment within the channel cross section and changes the alluvial bed profile. However local scour around the vanes is one of the problems in using of submerged vane technique. The extension of local scour hole is related to the shape of the vanes. Primary submerged vanes are generally flat rectangular plates. In the present research, cutting a part of the leading edge of the vanes out is studied as a countermeasure in reducing the local scour. Studied vanes include a rectangular vane (as the baseline vane), and five other modified vanes with tapered leading edges with angle of = 30° , 45° , 60° , 70° , and 73. 3° . The present study aims to evaluate the effect of this modification on the vertical velocity components at the leading edge and strength of the secondary circulation in the downstream of the vanes. flow-3D numerical model, version 10, is used to study the flow field around the vanes. Methodology: The commercial CFD model flow-3D was used in this research. Experimental velocity measurements were used for calibration of the model. For this purpose, a recirculating flume (7. 30 m long by 0. 56 m wide by 0. 6 m deep) was used. A centrifugal pump discharged the water into the stilling tank at the entrance of the flume. In order to create a uniform inflow of water, a screen was placed at a distance of 1 m from the flume entrance. A tail gate was used to adjust the depth (do) of water in the flume to a constant value of 0. 25 m. The dimensions of the vanes were determined using Odgaard’ s (2008) design criteria: a vane height-to-water depth ratio of Ho/do = 0. 3 and length of L = 3Ho. A mean flow depth of do = 0. 25 m yielded Ho = 0. 075 m and L = 0. 25 m. velocity measurements carried out using vanes V0 and V3 at a flow Froude number of Fr = 0. 16. In each test, the vanes were installed on the centerline of the flume at an angle of 20° to the flow. In order to study vane-induced velocity field, 4×4 cm 2 grids across the flume were taken at the center of the vanes. At each grid point, three-dimensional components of velocity vector (u, v, w) were measured by means of an electromagnetic velocimeter (EVM). Velocity very close to the walls of the flume was not measured. Results and discussion: On the high-pressure side of the vanes, vertical velocity components were upward (positive) and on the low-pressure side were downward (negative). Therefore, a clockwise secondary circulation was generated at downstream of the vanes. Downward velocity components at leading edge of primary rectangular vane (vane V0) were obvious. By cutting parts of leading edge out of vane V0 for tapered vanes V1 and V2, the magnitude of negative w-velocity components was respectively reduced by 40% and 69%. By increasing the taper angle for vanes V3, V4 and V5, downward velocity components were diminished, effectively. Moment of momentum (MOM) quantity was used in order to evaluate strength of vaneinduced circulation. MOM values were applied for comparison of performance of the vanes. For this purpose, velocity data at two sections at the distances of 2Ho and 4Ho, i. e., 15 cm and 30 cm downstream from center of the vanes was used. In the calculation of MOM, 100 velocity components (50 v-components and 50 w-components) were used. Therefore, this quantity is a useful criterion for evaluation of the performance and efficiency of the submerged vanes. Conclusion: Velocity distribution and moment of momentum (MOM) of the vanes indicated the reduction of erosive negative velocity components at the leading edge of the tapered vanes. Based on MOM values, cutting the leading edge out of the vanes causes lower performance. In other words, this modification restricts the vane-influenced field of the tapered vanes relative to the rectangular vane (vane V0). Results showed that the performance of tapered vanes (V1 to V5), relative to the rectangular vane, (at distance of 2Ho) is respectively reduced by 5. 8%, 7. 3%, 17. 8%, 33% and 42. 6%; at distance of 4Ho the amount of reduction respectively is 7. 4%, 11. 9%, 17%, 25. 5% and 34. 3%. On the contrary, the efficiency of the tapered vanes increased. The amount of increasing at distance of 2Ho from the center of vanes V1 to V5 respectively is 3. 2%, 9%, 11%, 14% and 14. 8% and at distance of 4Ho respectively is 1. 4%, 3. 6%, 12. 1%, 26. 7% and 31. 3%. Therefore, if tapered vanes are used to reduce the local scour, big values for the distance between the vanes arrays (δ s), according to the design criteria, are not recommended.