--> Abstract: Using Seismic Facies to Constrain Electrofacies Distribution as an Approach to Reduce Spatial Uncertainties and Improve Reservoir Volume Estimation, by Bruno de Ribet, Pedro Goncalves, Luis H. Zapparolli, and Cesar A. Ushirobira; #90124 (2011)

Datapages, Inc.Print this page

Making the Next Giant Leap in Geosciences
April 10-13, 2011, Houston, Texas, USA

Using Seismic Facies to Constrain Electrofacies Distribution as an Approach to Reduce Spatial Uncertainties and Improve Reservoir Volume Estimation

Bruno de Ribet1; Pedro Goncalves1; Luis H. Zapparolli2; Cesar A. Ushirobira2

(1) Paradigm, Houston, TX.

(2) Petrobras, Rio de Janeiro, Brazil.

In the 3D geological modeling workflow, the knowledge of the distribution of facies is a critical step due to the customary non-stationary behavior of the depositional environments, for enhancing a good match between simulation and history matching. Drilling campaigns are looking for areas of interest with better reservoir rock quality, economical hydrocarbon accumulations and help to get in-situ valuable information. This sparse type of recovered information frequently introduces a strong bias in subsequent estimative of facies proportions within the geological grid. Additionally, stationary techniques do not guarantee the appropriate distribution of geobodies, regularly mapped by seismic facies studies.

To overcome the associated uncertainties of the 3D geological model, we propose a workflow based on the combination of seismic facies and electrofacies. The objective is to constrain the spatial distribution of electrofacies and reduce geological and volumetric uncertainties.

For illustrating such workflow, we use a dataset from a giant field offshore Brazil (Campos Basin). The main challenge is the integration of data from different sources and scales like cores, logs and seismic. Electrofacies are generated from logs (conventional or from images), using statistical methods like multivariate regression or neural network and then extended to the grid scale at well location. A seismic facies classification, based on Neural Network technology, is used for generating a seismic facies volume. The input to the classification process is the result of a simultaneous inversion (Ip and Is) from a set of angle stacks. Neural Network Technology is used rather than any other classification algorithm to better discriminate the lithology and its spatial distribution, keeping the continuity of the facies.

To bring the seismic facies information to the scale of the reservoir grid, a statistical pairing is applied in order to determine the correlation between seismic facies and the electrofacies at well location. The resulting probabilities are propagated throughout the entire reservoir grid, conditioned to the spatial distribution of seismic facies.

Assigning probability of electrofacies occurrence for every seismic facies has proved to be a reliable approach to constrain the distribution of electrofacies on geostatistical algorithms, reducing spatial uncertainties and better estimating volumes of connected geobodies.