4.1.1. Standard Velocimetry

The use of micro high-speed propellers is appropriate when the quantification of local time average velocities is the measurement objective, such as in discharge measurements. However, the use of such propellers can be constrained by the presence of sediment due to the potential damage caused in the mechanical system. A modern approach for pointwise velocimetry is the use of Acoustic Doppler Velocimeters. Even though this instrument has been employed to measure the time averages and turbulent quantities in a broad range of flows, and even though the instruments have reduced dimensions, in cases of supercritical conditions, their intrusiveness may create a flow pattern similar to those caused by obstacles (Figure 2). Another consequence is the generation of oblique surface waves affecting the whole cross-section of the channel downstream.

## 4.1.2. Advanced Measurement of Hydrodynamic Processes

The use of advanced optical or acoustic measurement techniques for laboratory research in the supercritical regime presents many technical restrictions compared to subcritical cases. Supercritical flows, and particularly the flows developed around obstacles, show important unsteady free surface deformations, which make any attempt to access them from the surface impractical.

In the case of a flat smooth bed, measurements along the vertical planes can be performed through the bottom wall. One possibility is to use an acoustic profiler [52], located within the bed, which emits a vertical acoustic signal towards the free surface (Figure 6d). This instrument provides access to a time resolved vertical profile of three velocity components measured with a vertical resolution precision of approximately 1 mm. A second alternative is an optical technique using sourced illumination and non-orthogonal cameras located at the bottom of the channel. The most common optical technique is two-dimensional Particle Image Velocimetry (2D PIV) or 2D particle Tracking Velocimetry [53]. In this case, the reduction of optical distortions is achieved through the use of Scheimpflug adapters and optical prisms attached to the channel bottom-wall (as in standard stereoscopic or tomographic setups). A high image resolution and sampling frequency for the system becomes crucial for measuring the dynamics of small-scale structures around obstacles. A double pulse laser with a short-time interval between images is recommended to produce images with a high signal to noise ratio. An additional issue is air entrainment through the hydraulic jump or reverse spillage. In these cases, it would be necessary to consider the use of tracer particles, e.g., fluorescent particles, able to reflect the light source in a di fferent wavelength than the light scattered by the air bubbles. This and additional band-pass filters in the camera give access to an independent quantification of particle and bubble dynamics. Finally, other alternatives available, using similar optical arrangements from the bottom, are the use of Stereo- and Tomographic-PIV ([54], Figure 5b), as well as the use of pointwise measurements performed with Laser Doppler Velocimetry [55].

The use of a movable bed constrains the use of optical measurements from the bottom. An alternative is to use refractive index-matching between the sediment and the fluid, as presented in [56], which would facilitate a similar technique to the one described above for the plane bed condition (Figure 6a).

**Figure 6.** Innovative experimental techniques to be applied to supercritical hydrodynamics and scour processes in the laboratory: (**a**) an example of refractive-index and density matching from [57], (**b**) a view of a tomographic system, (**c**) CCP (Conductivity Concentration Profiler), (**d**) and ACVP (Acoustic Concentration and Velocity Profiler) from [58], (**e**) the application of Fourier transform profilometry to free-surface deformation, as per [59], (**f**) and an endoscopic camera placed inside a transparent pier for measuring scour evolution (from [47]).

#### 4.1.3. Image-Based Reconstruction of a Free Surface

The time-resolved measurement of free-surface deformation is commonly performed using point-wise sensors, such as ultrasonic sensors (from above) or electrical resistive probes (within the water column). The study in [20] showed that these techniques apply to both supercritical flow patterns: the detached hydraulic jump and the wall-jet like bow wave. On the other hand, access to instantaneous 2D free-surface deformation around the pier/abutment requires projection techniques, such as the RGB-D sensor method [60], the projected grid method [27] (with a limited spatial resolution), or the Fourier transform profilometry method [61], with a higher spatial resolution (Figure 6e). These techniques require high quantum efficiency for the CCD and a high frequency sampling camera to ge<sup>t</sup> access to the dynamics of the surface deformation.

#### 4.1.4. Distance Sensors for Scour Measurement

Laser distance sensors (LDS) are used to measure the scour-hole radius at different depths, thus quantifying the scour hole's geometry. In [62,63], researchers developed a measuring system composed of an LDS to measure the scour radius with an accuracy of ±0.4 mm. This LDS is placed inside a Plexiglas pier and aligned in a horizontal and radial direction, such that no refraction on the cylinder wall is observed during the measurements. The sensor is driven in the vertical direction by a step-motor with a precision of ±1/50 mm, and in the azimuthal direction, a vertical positioning system is driven by a second step-motor with an accuracy of <sup>±</sup>1/100◦, allowing the distance sensor to turn around in the scour-hole, taking various vertical profiles in different azimuthal half-planes. In this way, the geometry of the scour-hole below the original flat bed is automatically measured. The sensor performed well in tests with live bed conditions, having a flow intensity of *I* = 2. The same technique was applied to measurements in gravel [37] and sand-clay mixtures [30]. The application of LDS is, however, restricted to conditions close to clear-water and with low turbidity. It is, therefore, expected that LDS will not work properly in a number of supercritical cases where fine sediment is suspended.

Conductivity concentration profilers (see Figure 6c), originally developed in [64], were used to measure concentration profiles under sheet flow conditions in [58] using lightweight PMMA particles (1 and 3 mm). This technique is based on the inversion of the linear relationship between the sediment concentration and the conductivity of the medium. Using a grid of conductivity probes stuck on the pile, it would be possible to detect, at about 10 Hz, the position of the fixed bed interface at the pile position. Compared with LDS, the obtained information would be restricted to the bed elevation at the pile location, and no information on the 2D geometry of the scour hole would be provided. However, the advantage of such a technique is that it would be applicable even under intense live-bed conditions encountered in supercritical flow conditions for which no optical access is possible.

#### 4.1.5. Image-Based Reconstruction of Scour during Running Experiments

The study in [65] presented a stereovision-based technique for continuous measurement of the bed morphology. This technique is capable of reconstructing instantaneous surface representations of the evolving bed with high spatial resolution during scour experiments. Two calibrated cameras must be partially submerged in the flow and record videos of the evolving bed geometry. This technique considers the texture of sediment beds and does not require the use of targets or structured light. A set of computer-vision and image-processing algorithms were developed for accurately reconstructing the surface of the bed. This technique was further applied to the spatio-temporal characterization of scour at the base of a cylinder in [66]. Optical access to the scoured region might be an issue in supercritical flows, due to noise from the unsteady wavy water surface. The authors in [67] developed a bed level tracking system with micro cameras placed inside a pier to record the maximum scour depth under clear-water conditions. The study in [47] used an endoscopic camera placed inside the pier to record images of a graduated pier, registering the maximum scour depth in scour experiments conducted under live-bed conditions (Figure 6f). Similarly, the authors in [68] used a snake camera of 0.5 cm in diameter that slightly penetrated the flow surface (less than 1 cm) to look at a graduated strip on the upstream face of the pier. In these studies, algorithms for the automatic recognition of the scour hole bottom through digital image processing were developed. The ability of cameras placed inside the pier to record the maximum scour depth has emerged as a promising alternative for measuring the temporal evolution of the maximum scour depth at piers in supercritical flows. Indeed, surface waves and oscillations, as well as air entrainment occurring in supercritical flows, constrain the applicability of stereoscopic systems using cameras placed above the water surface. Recognition of the scour-hole bottom, and thus the detection of the maximum scour depth in the images, might be possible even in presence of suspended sediment particles because the color of the turbid water is different to the color of the bottom embedded foundation.

#### *4.2. Field Techniques for Flow and Scour Dynamics*
