An integrated modeling approach was used in a thin, highly heterogeneous complex clastic reservoir to build a more robust three-dimensional geo-cellular model by incorporating all available data from geoscience disciplines such as geophysics, geology, petrophysics and reservoir engineering. The model was used to replicate actual reservoir flow dynamic and generate optimum development strategy.
The early versions of the model relied heavily on seismic attributes to populate the reservoir facies. The poor seismic resolution of this thin reservoir has resulted in a patchy facies distribution which could not represent the actual fluid behavior during dynamic modeling. An alternative modeling approach was followed which populates the sand based on net sand mapping which honors the conceptual depositional model of this reservoir. As a result, this approach was used to improve the static model based on actual observed data like pressure and injection response in nearby producers and performance of well behavior.
A more reliable reservoir model was built after a lot of iterations between static and dynamic models without changing basic conceptual geology. The characterization of the reservoir as fluvial/deltaic mouth bar lobes has helped to capture a realistic heterogeneity level of the reservoir which made the history matching process much easier. Critically well-wise production performance analysis was conducted before proceeding to dynamic modeling. Production allocation anomaly was observed in few wells and necessary weight has been given to these wells during history matching. One of the key parameters that needs high attention particularly for reservoirs under waterflooding is relative permeability. Because of lack of availability and high uncertainty of relative permeability data in this reservoir, a systematic sensitivity analysis was performed on end point relative permeability and Corey parameters. As a result, the impact of each examined parameter on reservoir performance was scrutinized. The concept of producer injector pairing was used during history matching process and satisfactory history match achieved with minimum property modifiers.
A real case example was used to introduce an alternative modeling approach for thin, below seismic resolution clastic reservoirs to achieve a satisfactory history matched model. The presented approach would provide rigorous and sound reservoir models. The sensitivity analysis on relative permeability parameters expedites to the better understanding of flow behavior of the reservoir.