ABSTRACT

As most seismic based estimation problems in exploration geophysics are ill posed, there is a great need to use additional information to improve prediction capabilities and reduce the degrees of freedom in the seismic based inference problem. To that end, the use of geomechanical models and data is attractive due to the intimate link between seismic wave velocities and geomechanical parameters such as effective stress and strength. Specifically, the ability to use rock physics and geomechanical models can be shown to improve predictability and reduce uncertainty. The challenge from a practical point of view is to gain confidence in the geomechanical model and to properly merge observed data with model prediction such that uncertain model estimates will not conflict with observed seismic data. In this presentation we will discuss two examples where geomechanical information is applied to (1) reservoir characterization problem and (2) velocity estimation for anisotropic model building. In the first example we will show how effective stress estimates from pore pressure model calibrated to drilling data can improve seismic reservoir characterization. The basic workflow is presented in Figure 1 (Bachrach et al, 2007). In the second example we show how geomechanical stress modeling with rock physics stress to velocity transformation can be merged with CIP tomography to better constrain subsalt velocity estimation and anisotropic velocity model building. This is done by analyzing the null space of the CIP tomography problem using the method described by Osypov et al, 2008 and then merging the geomechanical prediction such as the one presented in Bachrach et al (2008) in such a way that only the poorly resolved component of the model space are changed.

This content is only available via PDF.
You can access this article if you purchase or spend a download.