Answers are grounded in this paper and cite their sources.
Dr. Bear's approach nicely shows for the isotropic two-dimensional case that the coefficient of dispersion is a fourth-rank tensor. The question arose whether it is possible to construct a differential equation which describes this phenomenon. In fact, this may be done for a uniform velocity field by the following analysis. Let a rectangular cartesian coordinate system y1, y2, y3 be placed so that the y1 coordinate has the direction of the uniform flow υ. The lateral dispersion, i.e. in the direction of flow (y1), with coefficient DI is larger than the transverse dispersion DII in the directions y2, y3 perpendicular to the flow. The rate of change of the concentration due to dispersion is therefore described by three terms: for the lateral dispersion, |υ|DI∂2C/∂y12; for the transverse dispersion, |υ|DII [∂2C/∂y22]+∂2C/∂y32; and for the convection, −υ ∂C/∂y1. Combination of these terms gives (∂C/∂t)=|v|{DI∂2C/∂y12+DII∂2C/∂y22+DII∂2C/∂y32}−v(∂C/∂y1)((1)) This relation can be written (∂C/∂t)=|v|{(DI−DII) ∂2C/∂y12+DII∇2C}−v(∂C/∂y1)((2)) Let the rectangular cartesian coordinates x1, x2, x3 be rotated with respect to y1, y2, y3, so that α1, α2, α3, are the angles between y1 and x1, x2, x3, respectively.
No takes yet. Share an insight, caveat, or question.
Jong et al. (1961) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: