Abstract: Application of Probe Permeametry, CT-scanning and Petrographic Image Analysis to the Understanding of Petrophysical and Pore Geometry Properties: Carmopolis Conglomerate Reservoirs, Sergipe-Alagoas Basin, Northeastern Brazil

Rodrigues, Carlos R. O.*; Paulo L. B. Paraizo; Gustavo C. Pinho and Marcelo S. Almeida - Petrobras/Cenpes/E&P

The Carmopolis reservoirs in the Carmopolis Field, Sergipe-Alagoas Basin, is composed of a set of very complex siliciclastic reservoir rocks, ranging from clast-supported conglomerates to fine-grained sandstones, organized in four broad lithofacies. Reservoir quality is controlled by several sedimentological and diagenetic factors, which are responsible for the low core to log correlation factor obtained when comparing petrophysical parameters measured in core plugs with those calculated from well logs.

Probe permeametry, CT-scanning and petrographic image analysis were used to improve the understanding of porosity and permeability distribution with respect to the different pore geometries identified in these rocks.

Probe permeametry was considered a strategic tool in bridging the scale gap between well logs and petrographic thin sections. A well core was analyzed using a Core Lab Pressure-Decay Profile Permeameter (PDPK-300^TM) to generate a permeability profile through all lithofacies. This profile shows that permeability is controlled by grain size distribution, mainly in the sandstone facies. Conglomeratic facies show a wide variability in horizontal and vertical permeabilities as well as thin highly permeable zones which can act as preferential pathways to fluid flow.

The CT-scanning was used to generate a porosity profile through all lithofacies using the CT-scanning density curve processed through the integration of perpendicular sections along the core, evenly spaced 10 cm apart.

The petrographic image analysis was performed in impregnated thin sections. First, the thin sections were classified according to their porous systems and then correlated to the four lithofacies. Then, the images were digitized using a petrographic and scanning electron microscopes and processed in the Graftek's Ultimage software, which allowed the generation of binary images. After that, a set of 10 binary images per thin section were processed in the "Virtual Core Laboratory" software. This product from Waterloo University, Canada, performs the 3-D stochastic reconstruction of porous media from 2-D image statistical data. The extracted geometrical and topological informations are used as input to network simulators capable of predicting a number of petrophysical properties including permeability, formation factor, resistivity index, capillary pressure curves, pore size and throat size distributions. These parameters were compared with mercury injection curves, distinguishing the different pore systems characteristics.

This approach allowed the better understanding of petrophysical and pore geometry properties contributing to improve the quality of permeability prediction and the strategy of water injection and oil sweep in the Carmopolis Field.

AAPG Search and Discovery Article #90933©1998 ABGP/AAPG International Conference and Exhibition, Rio de Janeiro, Brazil