Author(s): William Crawford-Jones; Gustavo A. M. de Almeida; Sergio Maldonado
Linked Author(s):
Keywords: Sediment transport; Impulse; Incipient motion
Abstract: Understanding the intricate interaction between a turbulent boundary layer and bed sediment particles remains a longstanding challenge for Earth scientists and engineers. A key and fundamental aspect of fluid-particle interaction involves the conditions under which a stationary bed particle begins to move due to hydrodynamic forces, which can lead to the particle being fully dislodged from its original position. This condition is known as the initiation of particle motion, and accurately predicting it has proven difficult, especially in turbulent flows. Many sediment transport models, which are largely based on empirical data, make simplifying assumptions that may not accurately reflect the realistic conditions found in fluvial environments. As a result, these models often lead to poor predictions of sediment transport rates and associated processes like bed erosion and morphological changes. A promising approach to understanding initiation of motion, stemming from the original work by Diplas et al. (2008), emphasises the influence played by turbulent fluctuations. In particular, this approach focuses on the empirical observation that some fluctuations, while containing and transferring enough energy to momentarily destabilise a bed particle, may nonetheless be insufficiently long lasting to lead to full dislodgement of the particle from its original resting position. A proposed metric to account for this effect is the impulse (time integral) of the hydrodynamic force induced on the bed particle. However, experiments so far have focused on rather idealised settings, such as analysis of forces on a static particle (Celik et al. 2014) or use of a ferromagnetic sphere subject to well-controlled electromagnetic pulses (Diplas et al. 2008). We extend this line of research by studying the time-history of forces on a bed sphere during the entire dislodgement process, using high-speed imaging. Our experimental and theoretical results demonstrate that these events are nearly binary: once particle motion begins, either full dislodgement takes place or the destabilised particle displays a very small amplitude oscillation within its resting pocket, with virtually no cases in between. These remarks are important given the implied importance of the time history of hydrodynamic forces in the impulse-based criterion for initiation of motion, which, in light of these results, may have a limited impact on the potential enhancement of predictive models for initiation of sediment bed erosion.
Year: 2025