THE ACOUSTIC BACKSCATTER AT TWO WELL-SPACED GRAZING ANGLES TO CHARACTERIZE MOVING SEDIMENTS AT RIVERBED

The use of acoustic-based methods for hydrographic purposes was first introduced for the seafloor investigation: oceans depth, seafloor composition and roughness have been extensively studied since the mid of the last century (Medwin & Clay, 1998). Indeed, the direct measuring and observation at the seafloor are extremely costly and, in some cases, even unfeasible, explaining the thriving of sonar-based methods. The advantages of indirect investigations by means of ultrasound emissions have been also exploited in coastal areas and riverine environments. In these cases, the efforts primarily addressed the measuring of sea waves spectrum and river flow discharge, achieved by means of the Acoustic Doppler Current Profiler (ADCP). This is a non-static device that applies the Doppler shift of coherent pulses to eventually profile the water velocity along the ensonified range (i.e., the water depth). More recently, the ADCP was also applied to profile suspended sediment concentration in river channels by means of echo intensities from the water column (Guerrero et al., 2017) and to investigate sediment fluxes occurring close to the riverbed, i.e., the bedload (Rennie et al., 2002, Latosinski et al., 2017) by using the bottom track feature. In addition, few attempts were produced to map the mean size of sediments at riverbed by interpreting the intensity of the reflected sound from the bottom (Douglas Shields, 2010). Particularly relevant for this study, the attempt of using the ADCP for bedload assessment is based on the “apparent velocity”, representing a velocity calculated as the difference between the Bottom Tracking (BT) mode and the GPS velocity, where the former is computed using the frequency change of the scattered acoustic signal from the mobile riverbed. Nevertheless, the scattered signal from the bottom depends on a variety of environmental and instrumental parameters that may largely deteriorate the desired correlation among bedload and apparent velocity. A relevant issue is the acoustic beam penetration into the active layer that defines the thickness of the measurement volume. This work aims to investigate the possibility of using the acoustic backscatter from riverbed at two well-spaced approaching angles (i.e., 70° and 90°) of the acoustic beam to river bottom (i.e., grazing angle) to characterize the sediment layer moving close to the riverbed. This is performed using the RiverPro ADCP, a relatively novel model from Teledyne-RDI that projects one vertical beam (90° grazing angle) and four 20°-degree slopped beams (70° grazing angle). This device was extensively used on the River Po in Italy, investigating the variation of layer thickness across the channel that may be combined with the apparent velocity to eventually provide a more reliable estimation of the bedload. At the same time, this study offers a corrected method for the assessment of skin roughness, related to the sediment grainsize at the riverbed, to be eventually accounted for in numerical modelling (Nones et al., 2017).