Data Conditioning for a Combined Inversion and AVO Reservoir Characterisation Study

Abstract

Summary The Cocuite gas field, located onshore Mexico, is producing since the 1960’s from 4 main reservoir levels (E, G, channels, M) with a steadily declining production rate. The structural interpretation indicates important extensions of the main gas accumulation, undrained by existing wells. Detailed amplitude studies are conducted, using both pre- and post-stack seismic data, to validate this hypothesis and minimise the risk in out-step drilling. A sub-volume of 100 km 2 from the Cocuite 3D seismic survey was reprocessed. The data conditioning includes a full Kirchhoff pre-stack time migration, detailed velocity analysis and special noise attenuation. The zero-phased time-migrated CDP gathers were stacked and inverted to acoustic impedance (AI) by applying a model-driven inversion algorithm. The same CDP gathers formed also input for an AVO analysis, whereby the “intercepttimes-gradient” and “fluid-factor” attributes are extracted. The complexity of the inversion and AVO results called for the generation of synthetic 1D AVO models in order to better understand the seismic response. A large number of existing wells have been examined on their AVO behaviour. The prestack seismic imaging showed a good differentiation between gas and brine filled reservoirs. A combined approach - of structural interpretation, amplitude mapping, together with the presence of anomalous AVO behaviour and a drop in acoustic impedance – is essential for reducing the drilling risk attached to new prospects. Proper data conditioning is crucial for reliable seismic imaging and gives access to quantitative interpretation of reservoir characteristics.