On tomography velocity uncertainty in relation with structural imaging
Evaluating structural uncertainties associated with seismic imaging and target horizonscan be of critical importance for decision-making related to oil and gas exploration andproduction. An important breakthrough for industrial applications has been madewith the development of industrial approaches to velocity model building. We proposean extension of these approaches, sampling an equi-probable contour of the tomographyposterior probability density function (pdf) rather than the full pdf, and usingnon-linear slope tomography. Our approach allows to assess the quality of uncertainty relatedassumptions (linearity and Gaussian hypothesis within the Bayesian theory)and estimate volumetric migration positioning uncertainties (a generalization of horizonuncertainties), in addition to the advantages in terms of computational efficiency.We derive the theoretical concepts underlying this approach and unify our derivationswith those of previous publications. As the method works in the full model space ratherthan in a preconditioned one, we split the analysis into the resolved and unresolvedtomography spaces. We argue that the resolved space uncertainties are to be used infurther steps leading to decision-making and can be related to the output of methodsthat work in a preconditioned model space. The unresolved space uncertainties representa qualitative byproduct specific to our method, strongly highlighting the mostuncertain gross areas, thus useful for QCs. These concepts are demonstrated on asynthetic dataset. In addition, the industrial viability of the method is illustrated ontwo different 3D field datasets. The first one consists of a merge of different seismic surveys in the North Sea and shows corresponding structural uncertainties. The second one consists of a marine dataset and shows the impact of structural uncertainties on gross-rock volume computation.