Abstract:
Facies correlated seismic attributes (RMS amplitude, instantaneous amplitude) were combined with rock physics parameters, for probability property modeling of Turbidites and channel sands. The aim is to reduce geologic uncertainties inherent in well log correlation, and facies modeling of characteristic deepwater environments. This is critical for inter-well petrophysical property distribution, reservoir characterization,
field development strategy, and predictability of future field performance. An integrated approach using litho-sensitive seismic attributes, rock physics, facies log, artificial neural network and variogram analyses; was used within a geostatistical framework to model rock facies and related petrophysical properties. This study has revealed, discrete facies classes that can be linked to the architectural facies pattern of turbidites and channel sands. These facies are characterized by high output probability properties. The modeled reservoir bulk volume from the integrated approach is observed to be relatively lower, when compared with the result of the traditional log-based correlation and structural modeling. Consequently, this over estimation of bulk volume, characteristic of the sequential indicator simulation modeling of upscaled log facies alone, have resulted to high static and volumetric uncertanties. The application of seismic attributes to probability property modeling of turbidites and
channel sands, has improved inter-well facies distribution in the study area. This implies, better property distribution, better reservoir characterization, reduced uncertainties in static reservoir properties and volumetric estimation of hydrocarbon in-place. Also, the combination of seismic attributes and well logs in the sequential indicator simulation has led to drilling recommendation for optimal well placements, as driver for increased future production and reduced dynamic uncertainties.
