Geophysical prospecting consists of making a quantitative inference about subsurface properties from geophysical measurements. Due to many ineluctable difficulties, observed data are almost always insufficient to uniquely specify the rock properties of interest. Hence, inevitable uncertainty remains after the estimation. The sources of the uncertainty arise from many factors: inconsistency in data acquisition conditions, insufficient available data as compared to the subsurface complexities, limited resolution, imperfect dependence between observed data and target rock properties, and our limited physical knowledge. While the uncertainty has been identified for a long time, quantitative framework to discuss the uncertainty has not been well established.; The objective of this dissertation is to quantify uncertainty of rock property estimation and to reduce it by using multiple seismic observables. Using existing laboratory data and rock physics model parameters, we establish the general relationships between rock properties and pairs of seismic attributes. We show how optimal selections of seismic attributes allow us to better distinguish different rock property effects. One of the novel innovations in this work is to combine statistical formulations—information theory and Bayes decision theory—with rock physics models to quantitatively describe the dependence of seismic attributes on several important rock properties. Various sources of uncertainty about rock property estimation are quantified using the developed formulations. Furthermore, We propose a method of combining stochastic simulations and Bayes inversion to quantify the uncertainty about the dependence between seismic observables and target rock properties, caused by ignorance of other rock properties. We apply this method to explore scale effects on sand/shale ratio estimation from seismic reflectivity. One of the new results of this investigation is to show from the full probability density function that the effective medium average tends to overestimate the sand/shale ratio when the reservoir is randomly layered. The proposed framework of quantifying information given by seismic data will serve as a decision making guideline in various exploration stages.
展开▼