Abstract: Prediction of Carbonate Reservoir Heterogeneities from the Integration of Stratigraphic, Seismic, and Log Data

Gregor P. Eberli

Predicting reservoir heterogeneities from log and seismic data is difficult because the resolution of these tools is often insufficient to image reservoir-scale variability. This difficulty can be partially overcome by first integrating outcrop analogs to construct robust stratigraphic models that reflect, in detail, the distribution of facies and petrophysical parameters in the reservoir and, second, by combining porosity and velocity logs to estimate permeability distribution. Such models require the assessment of the distribution of facies and diagenesis in a sequence stratigraphic framework and a correlation of rock fabrics to their petrophysical parameters. Two examples illustrate this approach of assessing heterogeneities in carbonate reservoirs for enhanced product on.

Reservoir heterogeneities are the combined result of variations in deposition and diagenesis. In carbonates, both of these parameters are largely controlled by high-frequency sea level changes. High-resolution sequence stratigraphy is a powerful technique to assess these sedimentological variations in genetic units through time. In addition, because early diagenesis in carbonates follows a characteristic pattern during each sea level fall and rise, these genetic units also contain information about how diagenesis may be distributed. Facies and diagenetic alterations control the petrophysical parameters within the genetic units. Therefore, the sedimentary and diagenetic rhythm of the depositional environment can be recognized on the logs, and the distribution of facies and diagenesis p edicted based on the established model. A difficult next step is to assess the permeability distribution within the stratigraphic units. No direct method exists, but the combination of sonic and porosity logs can be used to estimate pore types and partially assess permeability. A study in Bahamian carbonates will illustrate this approach.

The integration of an outcrop stratigraphic model to subsurface log and seismic data has proven to be a successful approach in predicting reservoir heterogeneities in the Carboniferous algal mound reservoirs in the Paradox basin in the southwestern United States. The key is the detailed correlation of the outcrop analog to the wireline logs and 3-D seismic data from subsurface reservoirs. Thus, the most successful approach in reservoir prediction is the integration of a variety of data sets into reservoir-specific, but robust models that incorporate sedimentology and diagenesis into a high-resolution sequence stratigraphic framework. The calibration of these models with petrophysical parameters allows then the extraction of reservoir heterogeneities from log and seismic data.

AAPG Search and Discovery Article #90948©1996-1997 AAPG Distinguished Lecturers