High resolution models of complex reservoirs can be obtained through stochastic inversion combined with lithology simulation. The models are constrained by the requirement to honour all seismic, log and geologic input data. Generation of multiple models fulfilling this requirement permits the uncertainties to be quantified. Model realisations at high resolution mean that the properties of individual reservoir units that are not resolved by conventional seismic reflection imaging techniques can be investigated.

Stochastic inversion with lithology simulation is a powerful tool for integrating a wide variety of data - geological, geophysical, petrophysical and statistical - to produce constrained, detailed models of reservoirs. These models can be used for static or dynamic modelling. The technique is well suited to mature fields such as those in the Central Sumatra Basin, where the objectives are to find bypassed reserves and optimise well planning for secondary or tertiary recovery.

A case study from the Libo SE field of the Central Sumatra Basin is presented. Deterministic inversion of the seismic data could not resolve all the sand units of interest in the complex reservoir of this field. Stochastic inversion with lithology simulation resulted in reservoir models from which the distribution and properties of the individual sand units could be predicted. The high quality of these models was confirmed by the results from recent drilling.

You can access this article if you purchase or spend a download.