Bayesian Seismic Inversion

Determining the uncertainty inherent in reservoir property estimation is a key requirement for the risk assessment of an existing asset. PGS uses a state-of-the-art Bayesian inversion which directly inverts reservoir properties in the depth-domain. This, coupled with depth dependent rock physics, provides a unique workflow for asset de-risking.

 

Bayesian Seismic Inversion ExampleDelta oil saturation at the reservoir level between the base and the monitor survey.

Model-based Bayesian seismic inversion combines prior knowledge of the reservoir with seismic data constraints and produces a suite of stochastic inversions:

  • The prior layer-based model is built using rock physics information from well log analysis and initial layer times from interpreted picks
  • The inversion allows for uncertainty in inverted reservoir properties such as net-to-gross, fluid type, porosity and saturation
  • Simultaneous AVO inversion is supported using specified wavelets and seismic noise levels
  • The inverted stochastic results can be used to generate predictive distributions of useful reservoir quantities such as net-sand, hydrocarbon probability and thickness
  • Useful statistical quantities can also be calculated such as P10, P50 and P90 values for all stochastic properties
  • The workflow includes well tie and wavelet extraction, end-member trend interpretation, prior model building, calibration and inversion testing at well locations, inversion production and tweaking of inversion results.