Turbulent flow over an erodible bed is influenced by complex interaction of fluid, grain of sediment and bed forms. Dunes can form the majority of the deposits of alluvial rivers. They are developed in a wide range of textural materials from silt to gravel.These bed forms occur at discrete sites of the bed, although they commonly recur in periodic fashion and therefore can be considered as extension of channels. Their formation, destruction and stability are highly dependent on localized flow conditions and their presence, in turn, is known to play a crucial role in governing flow. They can affect the spatial and temporal distributions of local velocity and shear stress and as a result, local sediment traort rate. Understanding the feedback in this complex interaction among fluid flow, sediment flux, and bed morphology are at the head of studies of river engineering. No theory yet exist which can adequately predict the formation of dunes and their subsequence effects on the turbulent flow field. Since most previous investigations have been limited to sand dunes; this research investigates the effect of gravel dunes and the change of wavelength on flow structure. The experiments were conducted in a straight flume, 30m long, 1.5m wide and 0.5m deep with zero bed slope and maximum discharge of 200 lit/s in the Civil Engineering Hydraulics Laboratory at the University of Ottawa . Its side and bottom constructed of cement and the base constructed of aluminum. The flume recirculated water from a large underground reservoir through a series of pipes to the channel entrance where turbulence is dampened by a large setting basin. Approximately 1.6 ton gravel with the median diameter of 10mm was used to build the dunes. the primary advantage of utilizing fixed bed forms are that they allow detailed measurements of velocity without complication of both a migrating and changing bed form and the difficulty of flow measurement in the presence of sediment traort. However, measurements must be averaged over fairly long period of time to obtain accurate turbulence statistics and because bed forms migrate, the flow is typically not stationary over requisite averaging periods. Taking them into account, all trials made on laboratory setup with fixed 2D bed form. As it was mentioned above, the goal of this research is to investigate the wavelength effect on flow structure. To achieve this objective, 3 different sizes of wavelength accompanied a constant height; slip face angle and depth were studied in the test section. In total, 3 runs were selected. The bed dimensions were chosen on the basis of previous studies and existing empirical relationship for dune size. In all cases, 8 two-Dimensional gravel dunes, 0.08m high with a slip face angle of 28