Petrophysical Inversion on Carbonates from Pre-stack Seismic Data

Coléou, Thierry ¹; Allo, Fabien ¹; Colnard, Olivier ¹; Machecler, Ingrid ¹; Dillon, Lucia ²; Schwedersky Neto, Guenther ²; Colpaert, Arnout ³; van Wijngaarden, Aart-Jan ³
(1)CGGVeritas, Massy, France. (2) Petrobras, London, United Kingdom. (3) Statoil, Oslo, Norway.

This paper presents the results of the R&D collaboration project between Petrobras, Statoil and CGGVeritas for the 3D petrophysical inversion study on a carbonate field using pre-stack seismic data.

The carbonate sequences analysed are oolitic grainstone intervals (with various diagenetic alterations) alternated with tight carbonate mudstones. Two pore-types, inter-particle and micro-porous intra-particle, are dominant in these reservoir intervals and often co-exist in the same sample. The observed dual porosity system is modelled using a T-matrix petro-elastic model.

The petro-elastic modelling is of particular importance in the process as it translates the petrophysical properties into their elastic counterparts (Vp, Vs and density) necessary for the seismic modelling. The use of a petro-elastic model within the seismic inversion introduces constraints in the ill-posed inverse problem. It increases robustness and resolution and provides better understanding of the results as the petrophysical parameters are more easily understood than the elastic variables. Constraints are also more conveniently controlled in terms of porosity and saturations than expressed in velocities and density.

Petrophysical seismic inversion provides a direct estimation of reservoir geometry and petrophysical parameters from pre-stack seismic data, in our case: thicknesses, porosities and saturations. Starting with an initial model and constraints expressed in terms of petrophysics like fluid type and possible porosity range, the inversion scheme updates the model to honour the pre-stack seismic data. This approach enables to test and validate various reservoir modelling hypotheses. Quantitative statistical evaluation of the match between the seismic response of the model and the observed seismic data enables to evaluate if a model is compatible with the seismic measurements and also to analyse the influence of each of the parameters in the seismic response. The same quantitative statistical analysis is also used to select and validate seismic pre-conditioning parameters, often critical for the quality of the inversion results.

The application integrates the knowledge of the rock physics in the analysis of the pre-stack seismic data. The petrophysical seismic inversion provides quantitative estimate and insight in the lateral and vertical distribution of porosity and fluid content within the various carbonate formations.

AAPG Search and Discovery Article #90135©2011 AAPG International Conference and Exhibition, Milan, Italy, 23-26 October 2011.