Author(s): Benoit Camenen; Loic Deville-Cavellin; Fabien Thollet; Adrien Bonnefoy; Thierry Fretaud; Gilles Pierrefeu

Linked Author(s): Benoît Camenen, Fabien Thollet

Keywords: Suspended sand sampling; Peristaltic pump; Isokinetic

Abstract: The evaluation of fine sediment fluxes in rivers remains a difficult task, especially for the coarsest fraction in suspension, i.e. sand. For the clay and silt fractions, the assumption of a homogeneous distribution of the concentrations throughout the river section is generally valid. A single sample from the side of the river is then sufficient to evaluate the section-averaged concentration and grain size distribution. In addition, surrogate monitoring using turbidity sensors is nowadays commonly used to monitor SSC (Suspended fine Sediment Concentration) continuously. However, when dealing with sand, one observed large vertical and lateral gradients in concentration. There exists some technics to evaluate sand concentration continuously such as hydro-acoustics. However, these methodologies need to be calibrated with in-situ measurements of the total suspended sand flux. Such sampling campaigns, usually applied using suspension samplers, necessitate numerous water samples to describe the spatial distribution of sand concentrations. This is very time-consuming since the sampler needs to be brought back to collect each sample. Therefore, the description of the river cross-section is often limited to a relatively small number of samples. One solution could be the use of a pumping system. Indeed, by just modifying the position of the intake nozzle within the river section, one could get more efficiently a large number of samples. One issue is to sample using an intake velocity close to the streamflow velocity. Indeed, non-isokinetic sampling can lead to significant bias in concentrations and grain size distributions especially for sand. The objective of this work is to evaluate the capabilities of a peristaltic pump (Albin Pump ALP17) to sample sand suspension. Such peristaltic pump is robust and easy to deploy although their pumping power remains limited compared to immerged pumps. Tests have been performed in a test bench at the CNR, Lyon, France, where particles with a controlled concentration and velocity circulate through a pipe in a closed circuit. Calibrated turbidity sensors allow continuous measurements of the concentration and different intake nozzles allow sampling using a pump. A sensitivity analysis is made to evaluate the error in the pumped sand concentration depending on the intake velocity, the concentration, the orientation of the nozzle, etc. Eventually, the most important issue for the peristaltic pump appears to be its jerky intake velocity leading to instantaneous intake velocities larger than its average. Such behaviour induces larger sampling errors due to the difficulty to respect an isokinetic sampling.

DOI: https://doi.org/10.3850/IAHR-39WC252171192022191

Year: 2022