ABSTRACT: Integration of NMR, borehole image, and conventional log data for improved petrophysical characterization of thinly bedded reservoir sequences

Shorey, David S., Gary Ostroff, Rocco Difoggio, and Daniel Georgi, Baker Atlas, Houston, TX

Across thinly bedded sand-shale sequences, such as deep-water turbidities, where the bed thickness is less than that of the logging tool vertical resolution, reservoir properties of the individual laminations cannot be resolved. Consequently, the measured bulk response across these sequences reflects the partial volume-weighted average of the sand and shale properties over the logging tool's vertical resolution. This results in apparent low bulk porosity and resistivity measurements; low computed permeability; and high computed water saturation. In the absence of shale distribution information, the bulk response is often assumed to reflect the actual reservoir properties. Although this may be true in the case of dispersed shale, in laminated sequences the actual sand lamination reservoir properties are progressively masked by reductions in sand net to gross. The implications to reservoir description in the later case are to dramatically under estimate the reservoir quality, productivity, hydrocarbon storage capacity, and to frequently bypass productive intervals. Through the integration of NMR porosity data, borehole images and conventional logs, shale distribution can be characterized and laminar shale volume quantified over the logging tool's vertical resolution. If the shale properties can be resolved across a bed whose thickness is at or above the tool resolution, the laminar shale contribution can be removed from the bulk tool measurement providing a solution for the average reservoir sand properties.

AAPG Search and Discovery Article #90913©2000 AAPG International Conference and Exhibition, Bali, Indonesia