Coarsening a numerical mesh using methods of multiphase flow property upscaling is essential in almost any modeling application due to computational limits. In CO2 storage modeling, after injection wells have been shut off or far away from the wells, CO2 migrates upwards under the influence of gravity and capillary forces. Upscaling permeability (k), relative permeability (kr) and capillary pressure (Pc) is required for grid coarsening these models. However, it has been shown that this is a difficult problem and using conventional upscaling methods such as the common capillary limit upscaling approach leads to large modeling errors. This work presents a new upscaling method based on an effective property formula for k, power law averaging in the capillary limit for kr, and an optimization approach for Pc. The new method is tested on various example cases and coarse-grid simulations are shown to match fine-grid ones with sufficient accuracy. The challenge of upscaling the flows is found to be related to entry pressure trapping and the optimization upscaled Pc is shown to have a unique structure allowing to model the trapping. The method is global, requiring a fine-grid simulation for calibration of the optimized parameters. However, we show that the method reduces computational time dramatically if calibrated parameters are used in similar cases in which the fine-grid solution is unknown, such as for varying k realizations.

Using x-ray tomography in coreflooding experiments allows to characterize the sub-core, mm-scale, multiphase flow properties such as permeability, porosity, relative permeability and capillary pressure. This has been studied previously for CO2-brine drainage experiments and a procedure has been developed for sub-core property estimation, showing that their implementation in numerical models leads to accurate predictions of experimental measurements, such as core saturation distribution. Much less work has been conducted regarding CO2-brine imbibition modeling. In this work we characterize hysteretic sub-core properties using experimental data of CO2-brine imbibition coreflooding conducted on two core samples. We adopt the approach of [1] for sub-core capillary pressure modeling and that of [2] for relative permeability modeling, however, we find that these are not sufficient for accurate modeling of saturation distribution within the core. We improve the models by considering a unique turning point and Land trapping coefficient for each mm-scale grid block in our model and also by calculating new imbibition characteristic relative permeability curves based on a procedure developed in [3]. Results show improvements in matching experimental data. [1] R. Pini, and S.M. Benson. “Capillary pressure heterogeneity and hysteresis for the supercritical CO2/water system in a sandstone.” Advances in Water Resources 108 (2017): 277-292. [2] O. Dury, U. Fischer, and R. Schulin. “A comparison of relative nonwetting‐phase permeability models.” Water Resources Research 35.5 (1999): 1481-1493. [3] E. Anto-Darkwah, S.M. Benson, and A. Rabinovich. “An improved procedure for sub-core property characterization using data from multiple coreflooding experiments.” International Journal of Greenhouse Gas Control 105 (2021): 103226.