INTEGRATION OF SEISMIC, WIRELINE LOGS AND CORE DATA INTO A GENETIC CHARACTERISATION FRAMEWORK

Abstract

Whilst the adequate characterisation of reservoir heterogeneity is an acknowledged problem in the petroleum industry, surprisingly little use is made of seismic data to predict lithofacies and porosity in reservoir regions without adequate well control. Previous attempts at using seismic data to predict lithofacies and porosity variations are reviewed and a new technique based on simultaneous use of seismic and wireline logs is proposed. This new procedure is based on a two-step collocated co-kriging method. The first step simulates lithofacies architecture and is followed by the estimation of porosity values for each lithofacies group. Model studies show that this approach results in an improved description of reservoir lithofacies architecture and porosity variation. Moreover, the proposed technique is consistent with concepts of genetic reservoir characterisation, which aims to characterise the reservoir in terms of lithological flow units. The cokriging results are shown to be more reliable than those obtained using older and more established techniques.