Search

Meandering flows in rectangular shallow reservoirs were experimentally investigated. The characteristic frequency, the longitudinal wave length and the mean lateral extension of the meandering jet were extracted from the first paired modes, obtained by a proper orthogonal decomposition of the surface velocity field measured by large scale PIV. The depth-normalised characteristic lengths and the Strouhal number were then compared to the main dimensionless numbers characterizing the experiments: Froude number, friction number and reservoir shape factor. The normalised wave length and mean lateral extension of the meandering jet are neither correlated with the Froude number nor with the reservoir shape factor; but a clear relationship is found with the friction number. Similarly, the Strouhal number is found proportional to a negative power of the friction number. In contrast, the Froude number and the reservoir shape factor enable to predict the occurrence of a meandering flow pattern: meandering jets occur for Froude number greater than 0.21 and for a shape factor smaller than 6.2. [ABSTRACT FROM AUTHOR]

Bottom friction modelling is an important step in river flow computation with 1D or 2D solvers. It is usually performed using energy slope based formulations established for uniform flow conditions, or using a turbulent regime based approach relying on turbulence analysis. However, these formulations are often applied under conditions of relative roughness which lie far outside of their validity fields. Furthermore, the theoretical definition of the roughness coefficients, defined by the different authors of both approaches, is not valid for usual numerical flow modelling, considering numerical approximations. The value of this coefficient becomes generally dependent on the flow conditions. Following the definition of the flow validity field of the main friction formulations proposed in literature, an original formulation has been developed. It combines 2 explicit turbulent regime based formulations smoothly linked by a polynomial expression, providing a continuous formulation covering the wide range of roughness usually encountered in river flows. The formulation is suitable to model, with a unique value of the friction coefficient, river flows with a wide range of hydrodynamic properties (water depth, discharge). The efficiency of this new formulation, fitted to explicit friction formulations and numerically adjusted, is demonstrated through various 1D and 2D practical applications. [ABSTRACT FROM AUTHOR]

Shallow-water models with porosity are used to compute floods at a relatively coarse resolution while accounting indirectly for detailed topographic data through porosity parameters. In many practical applications, these models enable a significant reduction of the computational time while maintaining an acceptable level of accuracy. In this paper, we improve the use of porosity models on Cartesian grids by three original contributions. First, a merging technique is used to handle cells with low porosity values which tend otherwise to seriously hamper computational efficiency. Next, we show that the optimal method for the determination of the porosity parameters depends on the modelling scale, i.e. the grid resolution compared to the characteristic size of obstacles and flow ways. Finally, we investigate the potential benefit of using a different porosity parameter in each term of the shallow-water equations. Five test cases, two of them being original, are used to validate the model and assess each contribution. In particular, we obtained speedup values between 10 and 100 while the errors on water depths remain around few percent. [ABSTRACT FROM AUTHOR]