Abertay Research Collections >
Research Centres >
SIMBIOS Collection >
Please use this identifier to cite or link to this item:
|Title: ||Determination of soil hydraulic conductivity with the lattice Boltzmann method and soil thin-section technique|
|Authors: ||Zhang, Xiaoxian|
Deeks, Lynda K.
Bengough, A. Glyn
Crawford, John W.
Young, Iain M.
|Affiliation: ||University of Abertay Dundee. Scottish Informatics, Mathematics, Biology and Statistics Centre|
|Keywords: ||Lattice Boltzmann|
|Issue Date: ||May-2005|
|Publisher: ||Elsevier B.V.|
|Type: ||Journal Article|
|Rights: ||Published version (c)Elsevier Science B.V., available from http://www.sciencedirect.com/science/journal/00221694|
|Citation: ||Zhang, X., et al. 2005. Determination of soil hydraulic conductivity with the lattice Boltzmann method and soil thin-section technique. Journal of Hydrology. 306(1-4). pp.59-70. [Online] Available from: DOI: 10.1016/j.jhydrol.2004.08.039|
|Abstract: ||A pore-scale modelling of soil hydraulic conductivity using the lattice Boltzmann equation method and thin-section technique is presented in this paper. Two-dimensional thin sections taken from soil samples were used to measure soil pore geometry, and three-dimensional soil structures were then reconstructed based on the thin-section measurements and an assumption that the soil structure can be fully characterised by its porosity and variogram. The hydraulic conductivity of the reconstructed three-dimensional soil structures was calculated from a lattice Boltzmann simulation of water flow in the structures, in which the water-solid interface was treated as a non-slip boundary (zero-velocity boundary) and solved by the bounce-back method. To improve the accuracy of the bounce-back method, the particle distribution functions were located at the centre of the cube that was fully occupied either by water or by solid. The simulated hydraulic conductivity was compared with measured hydraulic conductivity and the results showed good agreement. We also analysed the probability distribution of the simulated water velocity and the results indicated that the transverse velocity components had a non-Gaussian symmetric distribution, while the longitudinal velocity component had a skewed-forward distribution. Both distributions had a marked peak for velocity close to zero, indicating a significant portion of stagnant water.|
|Appears in Collections:||SIMBIOS Collection|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.