Premium
Incorporating the Memory Effect of Turbulence Structures Into Suspended Sediment Transport Modeling
Author(s) -
Tsai Christina W.,
Huang ShihHsun,
Hung Serena Y.
Publication year - 2021
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2020wr028475
Subject(s) - turbulence , random walk , mechanics , statistical physics , diffusion , turbulent diffusion , brownian motion , particle (ecology) , flow (mathematics) , open channel flow , continuous time random walk , anomalous diffusion , stochastic process , physics , geology , mathematics , computer science , statistics , thermodynamics , knowledge management , oceanography , innovation diffusion
Modeling of the random movement of fine sediment particles in open‐channel turbulent flow is mostly built upon the memoryless Brownian motion process. Such a process describes the chaotic behavior of small particles without considering temporal correlations in terms of the particle moving velocity and direction (i.e., memory). However, when particles are transported in time‐persistent turbulent flow, the movements of the suspended particles may exhibit persistency that depends on the various temporal durations of turbulent flow structures—such persistence results in direction and magnitude variations in the velocity of the fine moving particles. The diffusion property of the particles may then deviate from normal diffusion. The conventional memoryless random walk models may not provide a comprehensive description of the particle diffusion process for the duration of turbulence structures when the particles are subject to memory effects. In this study, a novel random walk model is proposed to present the temporal correlation of the suspended sediment particle velocity caused by turbulence structures in open channel flow. The probabilistic properties of the proposed model are discussed. In particular, enhanced physical insights are obtained regarding the particle diffusion behavior in turbulent flows. Numerical simulations are conducted to demonstrate that, similar to the conventional memoryless random walk models, the proposed model shows normal diffusion for long‐term observations, despite its local superdiffusion behavior. The effective diffusion coefficient of the proposed stochastic process on a long‐term time scale is formulated.