Unsaturated efficient diffusion coefficient in OGS-5

Usability
1

Hi all,

does anyone have some further insights into how OGS5 computats the effective Diffusion coefficient? In the keywords it says

 Fick’s law in the porous media (effective diffusion coefficient):

\[ D^* = \omega \cdot D_d \]

where omega is a coefficient that is related to the tortuosity (Empirical coefficient, lab: 0.5~0.01)

Is omega simply the tortuosity given in $TORTUOSITY ?

And is there a possibility to account for unsaturated cases, i.e. to make tortuosity a function of water content as e.g. in the Millington-Quirk relation:

tortuosity = (water content)^n / (porosity)^m

or is saturation accounted for once the mass dispersion tensor is computed?
(It seems so from the code in rf_mmp_new.cpp under MassDispersionTensorNew, but saturation is only a linear factor there… )

Cheers, Erik

2

Hi Erik,
yes the effective diffusion coefficient (D_e = omega \cdot D_d) is computed using the value for tortuosity (omega) given by $TORTUOSITY.

Unfortunately, I do not know about a possibility to account for water saturation to compute the diffusion coefficient.

Best,

3

Thanks for the clarification, Johannes!