Paper Summary
Whether we build a subsurface parameter model or deliver a subsurface image, our industry has been sadly lacking in attempting to assign ‘error bars’ to any of the products created. It transpires that this is an extremely difficult task to undertake in a quantitative manner. Assuming that we have an acceptable migration algorithm that honours the physics of the problem, then there are certain minimum acceptance criteria, which tell us that at least the derived model explains the observed data to within some acceptance threshold. Namely: image gathers that are ‘flat’ following migration with the obtained model, and which also match available well data — but these criteria do not tell us how accurate or precise the model or image is. Bayesian analysis of tomographic model error offers one approach to quantifying image positioning uncertainty, and here we give an overview of the elements involved in this procedure.