New and simplified numerical method for reactive transport simulations

Numerically simulated reacton of mineralization for CO2 at the Ketzin site (graphics: GFZ)

18.03.2015: Scientists from GFZ German Research Centre for Geosciences have developed a new method which allows to greatly accelerate computationally intensive modelling of geological storage of the greenhouse gas CO2.

The sustainable use of the subsurface requires long-term predictions of changes in the rocks. Chemical processes due to water-rock interactions play a significant role in this matter. The issue at hand here is represented by the large time scale and the number of active chemical processes and interactions, as well as their interplay with mass transport. Without computer models it is not possible to produce quantitative estimations thereof. However, due to the needed calculation time, even modern supercomputers are not able to tackle high-resolution, fully coupled simulations for time scales on the order of 10,000 years.

The geoscientist Marco De Lucia and his colleagues of GFZ section "Hydrogeology" now have developed a simplified method for calculating the outcome of the processes active in a CO2 storage system, which under certain conditions can dramatically increase the efficiency of the simulations. The reaction paths predicted by fully coupled simulations, in fact, show a large degree of self-similarity, which is an important prerequisite for the minimization of the needed model parameters.Main assumption for the simplified model is that the presence of injected CO2 is the only driving force triggering the chemical reactions. Furthermore, it is assumed that the extent of migration of CO2 in the reservoir is not significantly affected by its mineralization. The simplified model is a one-way coupling, which only takes into account the feedback of hydrodynamics processes on chemistry, and not the contrary.

The loss of precision due to this approximation, however, is limited. Simulations for the Ketzin pilot site and their comparison with fully-coupled models demonstrate that the uncertainty is more than acceptable. The alternative coupling allows reactive transport simulations highly resolved in time and space with a fraction of the calculation time needed by the classical fully coupled models.

M. De Lucia, T. Kempka, and M. Kühn: “A coupling alternative to reactive transport simulations for long-term prediction of chemical reactions in heterogeneous CO2 storage systems”, Geosci. Model Dev., 8, 279-294, 2015, doi:10.5194/gmd-8-279-2015