Modelling the velocity coefficient alfa for the optimization of discharge estimation by non-contact techniques

Abstract

Recent technological development provided a huge impulse in river monitoring field. Non-intrusive innovative discharge estimation techniques have been proposed and applied to address typical issues of traditional methods. Radar- and image-based techniques do not require direct-contact between personnel, instruments, and water, and can safely provide reliable discharge estimates by observing remotely the motion of the liquid surface. Radar instruments exploit Doppler effect to evaluate the velocity of floating matter over the liquid surface, by calculating the delay between an emitted wave and the reflected signal. Handheld devices make radar technology rapidly deployable and easy-to-use; fixed radar gauges allow for continuous river surface velocity monitoring. Image-based techniques, such as Large-Scale Particle Image Velocimetry (LSPIV) and Large-Scale Particle Tracking Velocimetry (LSPTV), are based on commercially available, inexpensive, and easy-to-use recording devices. Environmentally sustainable tracer, naturally present or manually introduced, can be also used for tracking river surface. Image processing is performed with free and open-source software, applying cross-correlation analysis to identify the frame-by-frame tracer displacement. Non-intrusive monitoring techniques typically use the velocity-area method for river discharge evaluation. From the estimated surface velocity, a simplifying hypothesis is introduced to calculate the depth-averaged velocity. The ratio between depth-averaged and surface velocity is called velocity coefficient, alfa, and, using simplified methods based on surface velocity to derive discharge, it is often approximated equal to 0.85. Velocity coefficient, as well as the depth velocity profile, is indeed influenced by a series of geometric and hydraulic factors (e.g., depth, distance from the banks, roughness, turbulence). The present work aims to analyze the variability of alfa with respect to some of the aforementioned factors, exploiting a large dataset collected during an extensive field measurement campaign carried out by means of an Acoustic Doppler Current Profiler (ADCP) in Sicily, Italy, between 2020 and 2023. The availability of ADCP velocity data allowed for the application of inference techniques to characterize alfa coefficient, considering rivers in different hydraulic, environmental, and geometric conditions and to propose a more accurate method for estimating the velocity coefficient alfa, instead of a constant value of 0.85, to improve discharge estimation via non-contact techniques.