Re: PHREEQC

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

> Thank you for your response and offer to look at the file. This file
contains data from the most upstream site (solution 1) and the most
downstream site (solution 2). I thank you for any assistance you can give
me. I realize WATEQ4F and PHREEQC databases do not contain Cr. I haven't
decided what to do about Cr yet but I left it in the file anyway. The only
harm it seems to cause is that I get a warning message telling me it is not
found in the database.


> Jim asked why I chose inverse modeling instead of solution/solubility. As
I told him, results of the inverse modeling examples in the PHREEQC manual
were easier for me (a biology background not a chemistry one) to interpret
and understand in relation to the questions I am trying to answer. If
another model is better suited to identifying precipitation/dissolution of
metals, or will give me the same information in a different way, I am more
than willing to listen to your advice. In fact, I have been looking at the
solution species results, but I am not sure how to take the output
(molality, activity, saturation indices and solubility constants) and say,
for example, Cd precipitates at these sites in the river, Zn precipitates
at these sites, etc.

I took a quick look at the speciation model results. I normally just look
at calcite, CO2, and gypsum for starters. Iron and aluminum look a little
high, dissolved concentrations should be used in speciation modeling. Other
trace metals probably don't form pure solids but precipitate in solid
solutions or sorb on sediments, so the SI's don't indicate a lot.

As for inverse modeling, I would stick to the majors, Ca, Mg, K, Na, C,
SO4, and Cl. Your input file doesn't have any sources or sinks for elements
(-phases in INVERSE_MODELING), but then it's probably hard to guess what
would be reacting. In a stream, you may form some calcite and you will
exchange CO2 with the atmosphere. After that, my guess is that the
concentrations change mainly because of inflow of water of different
compositions. So unless you have those compositions, from which you could
estimate some mixing fractions, there's not a lot you can do by modeling.
For starters, I'd look at the changes in concentrations and try to figure
out whatever you can. Looks like Na and Cl go up and Ca, Mg, and SO4 go
down. You need to decide whether this is real or an artifact of sampling
and flow conditions. If it is real, then you need to devise a hypothesis
that explains the changes.

David


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