Abstract Geothermal fluids are highly variable in chemical composition and ion concentration. Parametrising density, as one of the most important fluid properties for geothermal reservoir development, is therefore challenging. The purpose of this paper is to test the hypothesis that saline geothermal fluids can well be characterised for density when only the dominating dissolved salts (i.e., NaCl, KCl, and CaCl 2 ) are taken into account, thus neglecting any minor fluid constituents. For the example of four geothermal sites with known chemical fluid composition and significant differences in total salt content (Groß Schönebeck and Insheim, Germany, Balmatt, Belgium, and Heemskerk, The Netherlands) synthetic aqueous solutions of the main salts without and with three different other salts (i.e., LiCl, SrCl 2 , and MgCl 2 ) representing the minor fluid constituents were parametrised for density at atmospheric pressure and temperatures between 293 K and 353 K. Moreover, density was derived numerically using the PHREESCALE chemical code and evaluated against the analytical data. The results demonstrate that: (1) an effect of the ion type is evident with density increasing in the order of added LiCl, MgCl 2 , and SrCl 2 at a given concentration. (2) The uncertainty in density when neglecting any minor fluid components is at most 2 % which indicates that fluid density can be well characterised when applying the database of the main salts only. (3) The match between analytical and numerical data is excellent with differences generally less than 1 % evidencing that PHREESCALE permits to reliably predict the density of complex and multicomponent geothermal fluids.
Hoffert et al. (Tue,) studied this question.