Benchmark initiative on coupled multiphase flow and geomechanical processes during CO2 injection

CO2 injection into deep saline aquifers involves multiple strongly interacting processes such as multiphase flow and geomechanical deformation, which threat to the seal integrity of CO2 repositories. Coupled simulation codes are required to establish realistic prognoses of the coupled process during CO2 injection operations. International benchmark initiatives help to evaluate, to compare and to validate coupled simulation results. However, there is no published code comparison study so far focusing on the impact of coupled multiphase flow and geomechanics on the long-term integrity of repositories, which is required to obtain confidence in the predictive capabilities of reservoir simulators. We address this gap by proposing a benchmark study. A wide participation from academic and industrial institutions is sought, as the aim of building confidence in coupled simulators become more plausible with many participants. Most published benchmark studies on coupled multiphase flow and geomechanical processes have been performed within the field of nuclear waste disposal (e.g. the DECOVALEX project), using single-phase formulation only. As regards CO2 injection scenarios, international benchmark studies have been published comparing isothermal and non-isothermal multiphase flow processes such as the code intercomparison by LBNL, the Stuttgart Benchmark study, the CLEAN benchmark approach and other initiatives. Recently, several codes have been developed or extended to simulate the coupling of hydraulic and geomechanical processes (OpenGeoSys, ELIPSE-Visage, GEM, DuMuX and others), which now enables a comprehensive code comparison. We propose four benchmark tests of increasing complexity, addressing the coupling between multiphase flow and geomechanical processes during CO2 injection. In the first case, a horizontal non-faulted 2D model consisting of one reservoir and one cap rock is considered, focusing on stress and strain regime changes in the storage formation and the cap rock. For the second case, a fault is introduced to investigate the risk of fault reactivation and fracturing due to CO2 injection for a single and a multiple cap rock system, respectively. A multiple injector setting exposed to different tectonic stress regimes is proposed for the third case. Hereby, a 3D model is used compartmentalized by low permeability faults, which become permeable due to injection. Injection scenarios will be evaluated for extensional and compressive stress regimes. All model set-ups are based on already published simulation results of coupled multiphase flow and geomechanical processes during CO2 injection. To end with, a real site geometry including parameterization and realistic reservoir conditions is provided. The benchmark design and cases will be presented as well as some preliminary simulation results for the first cases. Interested institutions and researchers are invited to discuss and to participate in the study.