Publications

Integration of a low frequency model during seismic inversion enhances the qualitative and quantitative interpretation level of inversion results. Information required to build a low frequency model primarily comes from well data. In some geological environments it is possible to make simple extrapolations of the well data within a basic geological framework to provide a reasonable low frequency model. Often greater geological constraint is required, which can be incorporated in various ways. In this case study we present a novel approach based on the use of pseudo-wells.The Gendalo field in the Kutei basin comprises a series of laminated gas sands deposited in a deepwater setting. Quantitative models of the distribution of gas sands are difficult to produce because of ambiguities in the seismic response to variations in fluid content, sand thickness, and reservoir quality. We designed a workflow that brings together geological understanding of the depositional environment with the geophysical process of prestack seismic amplitude inversion. An initial geological model was built in terms of reservoir netto- gross (NTG) and fluid flag with information from four wells, seismic attributes, and a conceptual geological model. A transform between NTG and elastic rock properties enabled conversion of the model into synthetic seismograms. These synthetics were used to identify areas where the geological model had significant mismatch with seismic data. At these locations pseudo-wells were inserted into the geological model and the reservoir NTG and fluid flag were modified using geological constraints such that corresponding synthetics matched the seismic data. The process was iteratively repeated until a satisfactory result was obtained. After the final inversion, inverse transforms were used to convert the elastic properties to NTG and porosity.This integrated approach significantly improved understanding of the inversion process and provided the project geoscientists with greater confidence in the results. As a consequence, it is expected that the inversion results will be used more effectively in supporting development of the field.