Anomalous solute transport in complex media Abstract
Download 276.08 Kb. Pdf ko'rish
|
7-203-216
3.2 Sediment transport
Sediment transport is one historical and challenging topic in hydrology. It is the core of rock cycles and one of the major mechanisms by which the Earth’s topography is built. Its investigation mainly focuses on suspended sediment transport and sedimen- tation. Suspended sediment moving in the vertical direction is mainly determined by the competition between the downward settling velocity (due to gravity) and the up- ward motion (caused by turbulent diffusion). The two factors result in the sedimenta- tion or erosion of the river bed/bank, and also affect the migration of pollutants as well as the ecological protection of the river. The process of bedload transport can be more complicated due to the complex velocity field near the river bed and the interaction between sediment, the open channel, and the (migrating) river bed. In bedload trans- port, sediment particles may be buried below the active layer of the river bed, or expe- rience an accelerating movement due to turbulence breaking the sediment cluster. The complexity of sediment dynamics comes from the complexity of the flow velocity and the interaction between the river bed and the channel flow. The widely existing multi- scale mass exchange makes the classical diffusion theory unsuitable to describe the turbulent diffusion behavior of bedload particles. The fractional derivative, stochas- tic differential equation model has been applied to analyze sediment transport, and some valuable results have been obtained. In the analysis of sediment’s vertical distribution, the diffusion of sediment is gen- erally treated as normal diffusion, but the existence of turbulence makes the turbulent diffusion of sediment substantially different from normal diffusion. Previous results showed that turbulence diffusion is a typical anomalous diffusion behavior, and the fractional derivative diffusion equation model can accurately describe this type of dif- fusion process [26]. Based on the turbulent diffusion analysis, Chen et al. [6] proposed a new vertical distribution formula for sediment concentration by establishing a frac- tional derivative model associated with the properties of the Mittag-Leffler function. Comparison between the experimental data and several existing models indicated that the new formula is attractive since it has fewer model parameters and leads to accurate descriptions. The parameter sensitivity analysis showed that the fractional derivative order is the key parameter of the model. The analysis result of experimental data of several groups showed that the fractional derivative increases with the particle size. Hence, anomalous diffusion occurs more with fine particles. Bedload transport is affected by many factors, including the quantity of sediment, the river bed structure, the flow velocity, and turbulence diffusion, which can cause 202 | H. G. Sun et al. strong path dependency and spatial non-locality for sediment movement. Meanwhile, the history-dependent and non-local characteristics of fractional derivatives make the fractional stochastic differential equation model an excellent tool to accurately char- acterize bedload dynamics. The sediment particle may be blocked or buried by river bed structures. Hence, using the random motion approach, the probability density function of the waiting time exhibits obvious power-law tailing characteristics, while the coarsening and cluster structure of the river bed force part of the sand particles to experience a dramatically fast movement, causing the non-Gaussian distribution of sediment jump lengths. Therefore, the fractional derivative is a natural candidate to quantify the random motion of bedload sediment. Recently, fractional derivative models are gradually being used to characterize bedload transport. For example, Hill et al. [10] found experimentally and theoreti- cally that the random jump distribution function of a single particle is an exponential function, after assuming that the river bed is composed of uniform particles. How- ever, the random jump distribution function is a power-law distribution when the river bed is composed of mixed-size sediment, and the fractional derivative is an excel- lent tool for this situation. Ganti et al. [9] conducted a numerical simulation research to investigate the physical model for bedload transport under different conditions. Their results showed that the random jump distribution is close to the exponential distribution under the stable flow and near-uniform sand conditions, and the tradi- tional convection-diffusion equation can well describe the sediment dynamics. How- ever, since the unstable flow field and mixed particles are usually observed in natu- ral rivers, the statistical distribution of stochastic motion can be better described by the power-law distribution. Hence, the fractional derivative model might be a better model. Bradley et al. [5] applied the spatial fractional convection-diffusion model to describe the bedload transport in a natural river under instantaneous source condi- tion. Their simulation result showed that the spatial distribution of sediment is close to the α stable distribution, and the spatial fractional model agrees well with the ex- perimental data. Zhang et al. [32] considered the influence of the velocity field on ran- dom motion of bedload transport along sand beds, and then established a fractional advection-dispersion equation with fewer parameters to accurately describe the com- plex dynamics for bedload transport in real-world rivers. Download 276.08 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling