Inversión geoestadística conjunta de datos sísmicos y electromagnéticos
L. Azevedo and A. Soares
Journal name: Geociencias Aplicadas Latinoamericanas
Issue: Vol 1, No 2, October 2014 pp. 45 - 52
Info: Article, PDF ( 2.34Mb )
Price: € 30
Three-dimensional numerical models are key tools for reservoir characterization and management. In order to decrease the uncertainty associated with these models, it is essential the integration of all the available data within the geo-modeling workflow. Geostatistical seismic inversion has proven its ability to integrate well-log data and seismic reflection data to infer reliable subsurface elastic models. Recently, the use of electromagnetic data is being used to infer the spatial distribution of the subsurface pore fluid distribution. We present a new iterative geostatistical methodology that is able to jointly invert seismic reflection and electromagnetic data for petro-elastic models, i.e. water saturation, porosity and acoustic impedance. The proposed approach is based on a genetic algorithm that works as a global optimizer to converge the iterative methodology and uses stochastic sequential simulation as the perturbation technique of the model parameter space. A calibrated rock physics model acts as the link between the elastic properties derived from the seismic reflection data and the petrophysical properties derived from the inverted resistivity models. The geostatistical joint inversion of seismic and electromagnetic data was successfully tested and implemented in a synthetic dataset, Stanford VI-E. The inverted petro-elastic models of acoustic impedance, porosity and water saturation agree with the real three-dimensional petro-elastic models in both large- and small- scale details.