UNCERTAINTY QUANTIFICATION OF CALCITE REACTIVE TRANSPORT IN FRACTURED POROUS RESERVOIR

The dissolution of carbonate minerals, calcite in particular, has received considerable study in the geochemical kinetics literature. Despite the accumulation of a large dataset over several decades, there is significant uncertainty in the value of the dissolution rate under given conditions. Calcite dissolution is controlled by surface-reaction (as opposed to transport) kinetics with increasing pH under alkaline conditions far from equilibrium (high under-saturation). Although the data exhibit internal consistency within results from a given laboratory, absolute rates at high pH vary by well over an order of magnitude which is surprising given the large quantity of experimental data available. Some of the differences in rate reflect differences in experimental conditions (e.g., ionic strength, P_CO2, alkalinity). However, it is difficult to evaluate what additional sensitivity may be present as a function of experimental and analytical methodology, starting materials, or other factors. Here, we are not concerned with the true value but given an ensemble of possible rate law one can determine the impact of those models on the ensemble average of distribution of the species. If we cannot understand the origin of differences in rates derived from changes in solution chemistry, then neither can we fully understand the relationship of those rates. We thus opt for probabilistic uncertainty quantification through direct Monte Carlo simulations. The rate of dissolutions is assumed stochastic and model derived from data will serve as a driver for the conditional probability of rates given a pH. Here the conditional probability is uniform between the ranges of the possible rates. There are several other uncertain factors which could be geological or physical which increase the dimensionality of the problem. We focus here, without loss of generality of scheme, on the uncertainty within the rate of dissolution. It is our ultimate aim to understand how this uncertainty, regardless of its origin, can propagate through a thermo-hydro-chemical process. Using StoTran, a numerical framework for simulating coupled thermo-hydro-mechanical-chemical process, we will present two applications motived by enhanced geothermal environments.