This study investigates the influence of bed‐load transport on flow resistance and bed stability in steep step‐pool channels. A total of 86 flume experiments was performed. Stable step‐pool sequences were formed with increasing discharge under clear‐water conditions. The addition of fine bed‐load material to the flow led to a decrease in bed roughness and an increased mobility of step‐forming clasts. Velocity measurements showed that bed‐load transport significantly decreased flow resistance and increased near‐bed velocity for the conditions investigated. A set of existing resistance equations was tested against our data, of which the hydraulic geometry relation performed best. However, none of the existing equations was able to reproduce flow resistance during active bed‐load transport. Two new resistance equations based on a hydraulic geometry approach are proposed for conditions with and without bed‐load transport. The increase in near‐bed velocity associated with bed‐load transport was identified as the main cause for the observed increase in step mobility associated with bed‐load transport. For flows with bed‐load transport, step‐forming clasts were mobilized at discharges some 10% to 30% lower than under clear‐water conditions.

No abstract

Step-pool sequences typically form in steep mountain streams. Single step-pool units exist in steep channels with bed slopes exceeding 4% and a continuous step-pool morphology occurs in channels with bed slopes higher than 6%-7% (Church & Zimmermann, 2007;Recking et al., 2012). The flow regime in a step-pool channel is referred to as tumbling flow as the flow alternates between supercritical flow over the steps and subcritical flow in the pools (Whittaker, 1987). Thus, a large amount of flow resistance is attributed to the spill caused by hydraulic jumps (

Dual-tip phase-detection probes can be used to measure flow properties in gas-liquid flows. Traditionally, time-averaged interfacial velocities have been obtained through cross-correlation analysis of long time-series of phase fraction signals. Using small groups of detected particles, a recently developed adaptive window cross-correlation (AWCC) technique enables the computation of pseudo-instantaneous interfacial velocities and turbulence quantities in highly aerated flows, albeit subject to some smoothing which is due to the use of a finite window duration. This manuscript provides guidance on the selection of optimum processing parameters for the AWCC technique, additionally addressing shortcomings such as velocity bias correction in turbulent flows and extrapolation of turbulence levels to single particles. The presented technique was tested for three highly turbulent air-water flows: smooth and rough-wall boundary layers (tunnel chute and stepped spillway), as well as breaking shear layer flows of a hydraulic jump. Robust estimations of mean velocities and velocity fluctuations were obtained for all flow situations, either using dual-tip conductivity or fiber optical probe data. The computation of integral time scales and velocity spectra is currently limited by the data rate and must be treated with caution.

Gas–liquid flows occur in many natural environments such as breaking waves, river rapids and human-made systems, including nuclear reactors and water treatment or conveyance infrastructure. Such two-phase flows are commonly investigated using phase-detection intrusive probes, yielding velocities that are considered to be directly representative of bubble velocities. Using different state-of-the-art instruments and analysis algorithms, we show that bubble–probe interactions lead to an underestimation of the real bubble velocity due to surface tension. To overcome this velocity bias, a correction method is formulated based on a force balance on the bubble. The proposed methodology allows to assess the bubble–probe interaction bias for various types of gas-liquid flows and to recover the undisturbed real bubble velocity. We show that the velocity bias is strong in laboratory scale investigations and therefore may affect the extrapolation of results to full scale. The correction method increases the accuracy of bubble velocity estimations, thereby enabling a deeper understanding of fundamental gas-liquid flow processes.