This paper deals with a phenomenon often occurring in sewers during storm events - transition from free surface to pressurized flow and vice versa. This transition is also possible in sewers where the discharge is controlled by some control devices, like gates. Experiments were carried out at a test rig consisting of a circular pipe with gates at the upstream and the downstream ends. Because of the relatively steep slope of the pipe and upstream boundary condition (flow below the gate), free surface flow at the upstream end was always supercritical, so that a hydraulic jump was always present during transitions (transcritical flow). Experimental results were used for verification of a numerical model based on a shock capturing method, the McCormack explicit finite difference scheme.

During floods, the riparian vegetation in a watercourse significantly changes the velocity distribution and the turbulence structures of the flow. However, a certain influence on them is always exerted by the bed sediments. The aim of the present work is to study the bed roughness effects on the turbulence characteristics in an open-channel flow with rigid and emergent vegetation. Toward this end, an experimental campaign was conducted and consisted of three runs with different bed roughness conditions. The study is based on the analysis of the velocity, Reynolds shear stress, and viscous stress distributions. The results show that, in the region below the free surface region, the flow is strongly influenced by the vegetation. However, moving toward the bed, the flow is affected by a combined effect of vegetation, firstly, and bed roughness, secondly. This flow zone becomes more extended, as the size of the bed sediments increases. The shear stress distributions confirm the distinction between the two flow regions. In fact, the shear stresses are practically negligible in the upper zone of the water depth influenced by vegetation, whereas, owing to the bed roughness, they reach the maximum value near the bed surface. Finally, the analysis of the turbulent kinetic energy (TKE) revealed high values below the crest level and in the near-bed flow zone in the streamwise direction, whereas a strong lateral variation of TKE from the flume centerline to the cylinder occurred in the intermediate region.

No abstract

Vegetation on the banks and flooding areas of watercourses significantly affects energy losses. To take the latter into account, computational models make use of resistance coefficients based on the evaluation of bed and walls roughness besides the resistance to flow offered by vegetation. This paper, after summarizing the classical approaches based on descriptions and pictures, considers the recent advancements related to the analytical methods relative both to rigid and flexible vegetation. In particular, emergent rigid vegetation is first analyzed by focusing on the methods for determining the drag coefficient, then submerged rigid vegetation is analyzed, highlighting briefly the principles on which the different models are based and recalling the comparisons made in the literature. Then, the models used in the case of both emergent and submerged rigid vegetation are highlighted. As to flexible vegetation, the paper reminds first the flow conditions that cause the vegetation to lay on the channel bed, and then the classical resistance laws that were developed for the design of irrigation canals. The most recent developments in the case of submerged and emergent flexible vegetation are then presented. Since turbulence studies should be considered as the basis of flow resistance, even though the path toward practical use is still long, the new developments in the field of 3D numerical methods are briefly reviewed, presently used to assess the characteristics of turbulence and the transport of sediments and pollutants. The use of remote sensing to map riparian vegetation and estimating biomechanical parameters is briefly analyzed. Finally, some applications are presented, aimed at highlighting, in real cases, the influence exerted by vegetation on water depth and maintenance interventions.

A fixed-mass multifractal (FMA) analysis was used to investigate natural river networks and braided channels. In particular, while the study of natural river networks was performed with fixed-size algorithms (FSAs) in the past, the analysis of natural braided channels was not pursued before to our knowledge. Results showed the multifractal and non-plane-filling nature of all the digitalized data sets. Analysis of the digitalization step (constant or not) was performed and showed that it does not exert a strong influence on the assessed values of the Lipschitz-Hölder exponents and the support dimensions, even if a constant step permits better reconstruction of the right sides of the spectra, for negative moment orders of probabilities. The FMA approach presented two improvements with respect to the FSA one, in terms of oscillations of the scaling curves for negative moment orders of probabilities and of error bars. A more precise assessment of the multifractal spectra is of great importance in the development of multifractal models for the simulation of flood hydrographs.