Abstract

Interpretation of deterministic inversion does not prognosticate the inherent uncertainties in the data, so essentially any interpretation of inversion data should be probabilistic. Deterministic inversion followed by probabilistic interpretation approach is more desirable. Probabilistic interpretation incorporate in a systematic way all major factors contributing to the inherent uncertainty associated to reservoir models. In this study, facies and fluid probabilities of reservoir sand was determined from pre-stack seismic inversion data to estimate the probabilities of lithology using joint probability density functions (pdf). A joint/multivariate pdf was generated to model the statistical relationship between multiple properties from deterministic inversion using Bayesian probability theory in clastics reservoir sand in a field of Krishna-Godavari offshore basin. Rock physics analysis of logs was done and multivariate probability density functions (pdf) were computed from well logs of P impedance and Vp/Vs for reservoir properties defined by lithofacies. This classification gives four facies volume of oil, gas & brine sand and shale. The aim of this study is to quantify the uncertainty of different lithofacies estimated from inversion results.