M17: Bayesian inverse problems and seismic inversion
| Researcher: | Sean Lim |
| Team Leader(s): | Dr Chris Farmer & Dr Irene Moroz |
| Collaborators: | N/A |
Background
Seismic inversion, the inverse problem on the wave equation, is key to understanding the geological structure of hydrocarbon reservoirs. The project will explore the statistical formulation and solution of inverse problems for wave equations. It will focus on full waveform inversion for the characterisation of rock seismic properties and time-dependent wave fields in space and time.
Techniques and Challenges
We will adopt a Bayesian approach to the solution of the inverse problems, making use of existing information from disparate sources and taking care about how uncertainty is modelled. We will begin by studying the one-dimensional wave equation, with priors based on Gaussian random fields and having reviewed the literature, we will study the seismic inversion problem using both an acoustic wave equation and a convolution forward model. Various filtering methods will also be explored.
The Future
We hope to improve the efficiency and accuracy of the algorithms and techniques currently used to solve the inverse problem.