Re: Rio Grande mxing model

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

 > I'm trying to model the mixing of two streams (Rio Grande
             and the Pecos) in Lake Amistad.  I've obtained the
             values below for Aug/1984 (I'd like to compare to 8/1998
later).
             I'm using the measured inflow rates for the mixing ratios.
             1- Rio Grande, 2-Pecos & 3- is a Bouy in the lake.

>            Is there something I can do to decrease uncertainty
             and reduce the # of models (I get > 100).

The simplest way to reduce models is to use the -minimal option. That
should make many fewer models by eliminating phases that are not needed
within the given uncertainty.

I would also remove some phases. I don't know the system that well, but I'm
guessing clay and biotite reactions probably are not the dominant
reactions. So I'd limit reactions to CO2, calcite, and silica; reactions
that could be biologically driven. Calcite is near saturation and could
probably dissolve or precipitate. Gypsum is well undersaturated, as is
halite. Maybe these are reasonable aeolean phases, but I don't know if they
could really affect the water chemistry.

Your mixing calculation is not saved, so it is not used in the inverse
modeling, but the inverse model calculates the mixing fraction anyway. In
the file below, it is about 60/25 percent for the two waters. It also
requires a dilution with pure water (~10%). From here you need to add your
own knowledge of the variability in the input waters, alternative phase
selections, the importance of direct precip and runoff. The dilution seems
a bit too much, but I would probably attribute the problem to uncertainties
in the two source waters.

Good luck,

David



TITLE Rio Grande & Pecos Mixing at Buoy1
SOLUTION_SPREAD
    -units    mmol/l
 Number       pH     Ca      Mg         Na         Cl  S(6)     Si
K      temp  Alkalinity
      1      7.6   1.85  0.3745        3.6       0.96   2.5  0.267
0.138  27.5         1.0
      2      7.6   2.25   2.016      13.04       13.8  2.91  0.217
0.15   28.5        1.10
      3        8   1.80   0.864       5.65       3.95  2.18   0.30
0.13   21.0        1.20
MIX 1
             1 0.85
             2 0.15
             3 1.0
INVERSE_MODELING 1
    -solutions      1        2        3
    -uncertainty    0.1
    -phases
        # Ca-montmorillonite
        CO2(g)
        H2O(g)
        Calcite
        Chalcedony
        # Gypsum
    -balances
        Cl          0.1      0.1      0.1
        Ca          0.1      0.1      0.1
        Mg          0.1      0.1      0.1
        K           0.1      0.1      0.1
        S(6)        0.1      0.1      0.1
        Si          0.1      0.1      0.1
        Na          0.1      0.1      0.1
    -range             1000
    -tolerance         1e-010
    -mineral_water     true
PHASES
Biotite
    KMg3AlSi3O10(OH)2 + 6H+ + 4H2O = Al(OH)4- + 3H4SiO4 + K+ + 3Mg+2
    log_k     0
CH2O
    CH2O = 0.5 CH4 + 0.5 CO2
    log_k     0
END

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

Project web page: https://wwwbrr.cr.usgs.gov/projects/GWC_coupled