A deterministic geomodeling case study for a thin clastic reservoir: The Second Wall Creek Sand, Teapot Dome field, Wyoming, USA

Abstract

In this case study, we geomodeled the Upper Cretaceous Second Wall Creek Sand (SWCS) of the southern Teapot Dome field, Wyoming, USA, using public domain seismic and well-log data. The SWCS is an at-or-below seismic-resolution (about 20 m), fluvial-deltaic reservoir and has been the most productive in the field. The goemodeling procedure consisted of structural modeling of the depth 3-D grid (30 m × 30 m × 1 m cells) and petrophysical modeling for porosity and net-to-gross (N/G). The seismic data were enhanced by spectral balancing to resolve the SWCS. Facies modeling was not necessary because a facies analysis of gamma-ray logs and lithofacies and stratigraphic-discontinuity seismic attributes suggests a uniform facies. The porosity and N/G were modeled by neural network multi-attribute transform (NN-MAT). NN training included the absolute and relative impedances and three representative isofrequencies in addition to the conventional amplitude- and frequency-based attributes. The modeled porosity varies from less than 8% to over 18% with an average of 16.7% which agrees with the reported averages. The modeled N/G values range from about 0.2 to over 0.8 with an average of 0.7 which is greater than the reported average (0.5). The porosity and N/G models may not be optimal because only a few wells and a small number of the time samples for the thin reservoir interval were used in NN training. The geomodeling workflow presented in this study is optimized for thin reservoirs and can be used when facies modeling is difficult because NN-MAT computes petrophysical properties without being guided by facies models.