--> Bualuang Field Geological Cellular Model- A Multipoint Geostatistics Approach For Modelling Complex Fluvial Hydrocarbon Reservoirs

AAPG Asia Pacific Region GTW, Back to the Future – The Past and Future of Oil and Gas Production in the Asia Pacific Region

Datapages, Inc.Print this page

Bualuang Field Geological Cellular Model- A Multipoint Geostatistics Approach For Modelling Complex Fluvial Hydrocarbon Reservoirs


The middle to late Miocene hydrocarbon sandstone reservoirs of the Bualuang Field (informally termed as T2) were deposited by meandering fluvial channels developed within a variety of coastal plain settings with variable marine to estuarine influence. Geological cellular modelling of these poorly seismically imaged, low NTG reservoirs has been proven technically challenging. Previous facies cellular models were built using traditional Sequential Indicators Simulation (SIS) conditioned by well data and seismic-based lithology probability volumes generated from stochastic seismic inversion. However, it has become apparent that there is still scope to improve the static model to be able to represent the inherent complexity of fluvial depositional architectures and to obtain a more satisfactory history matching and subsequent reservoir performance prediction. The multipoint geostatistics approach is a pixel-based method whereby the traditional single-point variogram is replaced by a training volume that captures the vertical and horizontal patterns of facies and depositional elements of the meandering channels. This approach has resulted in significant improvement of the latest generation of geological cellular models by capturing and integrating multiple scales of geological heterogeneity observed in core, well logs and seismic datasets. The stacking patters, sedimentology, biostratigraphy and chemostratigraphy are consistent with deposition by meandering channels developed within a lower coastal plain setting. The facies associations are linked to two main depositional elements, namely fining-up pointbars and coarsening-up crevasse splays. The pointbars dominate circa 90% of the sandstones packages of the T2, comprising a vertical succession of facies associations which, from base to top, consist of polymictic conglomerates (CG); cross-bedded coarse- to medium-grained sandstones (CS); fine grained planar laminated to massive sandstones (MS); and capping hydromorphic paleosols (HPS). The less common crevasse splays consists of MS and/or CS capped by HPS. The pointbar sandstone packages form up to 11 vertical sequences separated by meter-scale two paleosols are repeated throughout the T2 stratigraphy. These cycles, in turn, have a typical muddying-upward signature identifiable in GR, density-neutron and porosity logs. These logs and core responses were therefore used to guide the calculation of facies associations and depositional elements in the non-cored intervals. Together with the training volume, stochastic seismic inversion shale probability volumes were also used as a soft conditioning for the HPS (background facies) in the facies modelling. The Gaussian Random Function Simulation algorithm was used to populate reservoir properties which were, in turn, conditioned to the facies model. The porosity model was conditioned to mean P-Impedance trend scaling transformation. The horizontal permeability model was conditioned to porosity using a collocated co-kriging method. Saturation height function conditioned to each facies is used for water saturation modelling. This study has shown a successful application of multipoint geostatitics in complex pointbar-dominated fluvial reservoirs. The modelling results give an improved and more realistic fluvial-pointbars architecture and provides a reasonable dynamic history match and reservoir performance prediction.