C. Ekberg et al., MINVAR and UNCCON, computer programs for uncertainty analysis of solubility calculations in geological systems, COMPUT GEOS, 26(2), 2000, pp. 219-226
Computer simulations of different phenonmena have become increasingly impor
tant as computer time has become cheaper and computers have become faster.
In some cases, it may be that simulation results are used instead of experi
ments, since experiments are expensive and require a great deal of time. On
e such example is the determination of solubilities of different elements u
nder different conditions, e.g. different temperatures and water compositio
ns. It is important when working with computer simulations, however, to rem
ember that the result is a direct function of the input data, i.e. if input
data are slightly uncertain, so will be the results. The main input data t
o solubility calculations are thermodynamic data, such as stability constan
ts and enthalpies of reactions, and water composition data. The thermodynam
ic data have been discussed elsewhere, and the focus of this paper is thus
on the water parameters. The uncertainties in water composition may origina
te from at least two different sources. Either the water composition at a c
ertain location has been measured, thus making the main measurements uncert
ain, or the uncertainties in water composition arise from computed estimate
s of water composition on the basis of rock-water interactions. To investig
ate these different approaches, two programs, MINVAR and UNCCON, have been
developed. The MINVAR program gives uncertainties in water composition caus
ed by uncertainties in rock composition using the CRACKER code. These uncer
tainties or measurement uncertainties may then be transferred to the UNCCON
program, which estimates the effect on the calculated solubilities. Both p
rograms use the Latin Hypercube sampling technique. (C) 2000 Elsevier Scien
ce Ltd. All rights reserved.