Re: (long) question on phreeqc

["David L Parkhurst" <dlpark@xxxxxxxx>]

Dear Chris,

I don't quite follow the mass balance on water. Seems like 1.0 pore volumes
have been added to the column at step 2, instead of just 0.5?


cell_no:      1   2   3   4   5
step 0:      .5  .5  .5  .5  .5
step 1:      1 X .5  .5  .5  .5
step 2:      1   1 X 1   .5  .5
step 3:      1   1   1 X 1   1

Still, however the flow should be, it would take some programming to do it,
but actually might not be too difficult. Advection is accomplished in one
short loop in either transport or advection, where solutions are copied
from one position to the next. It would be possible to modify that part of
the program to advect in a different way. There is a variable in a solution
structure for the mass of water. It is also possible to define solution 0
with a mass of water other than 1 kg. In concept, you can look at these
masses of water (assuming fully saturated is 1 kg water) and determine how
much to advect between cells. There is a routine add_solution that can be
called multiple times to accumulate a solution composition; saving the
mixture, memory. and storage would take a bit of work.

There is a problem with the dispersion calculation. The mixing fractions
are based on equal volume mixing. It would still work, but I don't think
the dispersion would be correct when you have different volumes and that
would be messier to try to reprogram. The dispersion calculation would
change the mass of water in cells if the volumes were not equal.  It would
be simpler to work in the advection module and ignore dispersion, at least
initially.

As for K(h), one way to do it is to have an unsaturated zone transport
model that sequentially transports each element in the system to make new
solution compositions in each cell, then run PHREEQC on each cell. We've
used this splitting approach in PHAST for 3D ground-water systems, but I
don't see any reason it shouldn't work if an unsaturated transport code
were used. I suppose if you knew the fluxes between cells from a 1D
unsaturated flow code, it would be possible to use that information in
PHREEQC at the point where advection is calculated.

David




David Parkhurst (dlpark@xxxxxxxx)
U.S. Geological Survey
Box 25046, MS 413
Denver Federal Center
Denver, CO 80225