Re: inverse modeling

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

 have tried some inverse modeling with PHREEQCI. As you can see from the
attached output there are errors flagged relating to a subroutine, although
the program gives models. Also, the second attempt yields no models
although I can construct one as shown in the exel spreadsheet. My questions
are:

1. do i need to worry about the Subroutine Cl1 errors.
2. why don't i get a model for the second attempt (INVERSE_MODELING 5)?

The Cl1 errors are coming from the small concentrations of ferrous and
ferric iron. Setting the iron concentrations to 0.0 will clear up most of
the problems. The second modeling fails because there is no way to account
for the small changes in ferrous iron between initial and final solution. A
few errors are related to models that generate significant water from
mineral reactions. These normally require moles of mass transfer of phases
and are not reasonable models. You can set a switch (-mineral_water false)
in INVERSE_MODELING to ignore the water generated by mineral reactions and
that gets rid of the remaining errors. I don't like to use this switch
universally, but this might be a good spot to get rid of some unreasonable
models and the error messages as well.

David


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