Bayesian Hierarchical modelling of large scale geo-spatial processes
26th January 2018, 3:30 pm – 4:30 pm
Main Maths Building, SM3
In this talk I will introduce the framework of updating large scale geo-spatial processes using a model-data synthesis method based Bayesian hierarchical modelling. Gaussian Markov random fields (GMRF) are used to approximated the spatial process to reduce the computational cost and Bayesian inference is done by the INLA method. For non-stationary global processes two general models are proposed. The GMRF approximation and the INLA implementation are also adapted to these models. Finally, I will also show an application of updating the global glacial isostatic adjustment by using the GPS measurements.