After a reservoir's structural framework has been established using seismic and well tops data, a conceptual heterogeneity pattern is used to control the spatial distribution of core and log data. Traditionally, seismic acoustic impedance or depositional facies map are the conceptual heterogeneity patterns used. Sometimes, facies map may be derived from limited wells, and seismic data is expensive to acquire. In such cases, a purely statistical approach like sequential gaussian simulation is used and the resulting heterogeneity pattern may have large uncertainty, being driven by probability rather than actual data. The goal of this study is to derive a new conceptual heterogeneity pattern using inexpensive and readily available historical pressure data.

It is known that pressure cluster differentiation can be caused by faults or fractures, we investigated in this study whether a non-faulted and non-fractured reservoir can exhibit pressure cluster differentiation. Numerical experiments are performed using hypothetical models to study the impact that stepwise and continuous permeability variations, the permeability ratio, duration of shut-in, and flowrate all have on the historical pressure trends of wells located within regions of similar and different permeability magnitudes. The historical pressures for all wells are plotted on the same axis to assess cluster differentiation. Closed polygons are drawn around the wells within each differentiated cluster.

The simulation results indicate that pressure cluster differentiation can occur in non-faulted and non-fractured reservoirs and typically implies heterogeneity. Therefore, in a reservoir where pressure cluster differentiation is observed, mapping the spatial locations of wells within each pressure cluster would result in a reservoir heterogeneity map which is shown to be comparable to seismic acoustic impedance map in an example dataset. A key contribution of this study is the development of an alternative conceptual heterogeneity pattern, using inexpensive and readily available data, that can be used as a trend map for guiding spatial property distribution during geo-modeling. An application example is used to further strengthen the conclusions derived from numerical experiments.

You can access this article if you purchase or spend a download.