Print this pageIntegrated of Multiattribute Analysis, Inversion Modeling and Reservoir Modeling as a New Approach for Porosity Prediction and Identifying Sand Distribution in Tanjung Raya Field, Barito Basin, Southeast Kalimantan
The integration of well-log and seismic data become increasingly important due to bridging the exploration and development needs. The new method developed for predicting log properties from 3D seismic with statistical approach is Multiattribute Analysis. Incorporated by reservoir modeling with stochastic method, the multiple realizations yield non-unique solution to enhance the resolution. This study has concentrated for predicting porosity to identify the distribution of sand reservoir through several production layers within Lower Tanjung Formation.
Tanjung Field is Pertamina own operation area located to the north east of Barito basin, which is known as the largest basin in South Kalimantan. It was discovered in 1937 by Shell with successfully discovery well T-001. Recently, production performance is successfully achieved from 88 production wells.
The common problems arise in reservoir characterization in Tanjung Field are to determine the lateral connectivity of sand distribution along the Tanjung structure and to define the facies changing. The objective reservoir is silisiclastic from Lower Tanjung Formation which is consists of six production layers deposited above volcanic basement. Fluvial-deltaic setting controls the depositional environment of this area.
Two mathematical formulations have been calculated: multivariate linear regression and probabilistic neural network. The use of those algorithms using seismic attributes, acoustic impedance data as the external attribute and detailed petrophysical analysis is found to be useful to identify the reservoir character between layer and facies changing.
In summary, when seismic inversion modeling was not sufficient enough to identify facies changing, multiattribute analysis and probabilistic neural network can be thought as the extension of inversion of post-stack data to overcome this problem. Furthermore, stochastic reservoir modeling is give more oportunities to identify reservoir connectivity, better spatial distribution and reducing the uncertainty. The integrated approach enhanced the confidence level in developing undrilled prospect within this field.
AAPG Search and Discover Article #90100©2009 AAPG International Conference and Exhibition 15-18 November 2009, Rio de Janeiro, Brazil