2017
The inversion of seismic data often adds valuable information in quantitative interpretations but the value of this information is susceptible to bias depending on the selected low frequency model (LFM) provided to generate absolute instead of relative elastic properties. The seismic data is band limited and hence the inversion output from seismic alone is a relative impedance volume. To obtain the broad spectrum inversion output or the true earth model in terms of its elastic properties, an a-priori low frequency model from recorded well logs has to be provided. This low frequency model also act as a guide for population of impedance properties away from the well. The problem lies in the fact that the low frequency model itself is a statistically populated property. Hence great care must be taken to create an reliable low frequency model (LFM). Infact seismic inversion can be thought of as the combination of low frequency model and band limited relative inversion derived from the seismic data. In this study four approaches for LFM building was looked into for minimizing spurious artefacts in inversion results.
Low frequency model (LFM), Geo-cellular modeling (GCM), Geobody encoding