B0181
Title: Bayesian spatial modeling for data fusion adjusting for preferential sampling
Authors: Paula Moraga - King Abdullah University of Science and Technology (KAUST) (Saudi Arabia) [presenting]
Abstract: Spatially misaligned data are becoming increasingly common due to advances in data collection and management. A Bayesian geostatistical model is presented for the combination of data obtained at different spatial resolutions. The model is flexible and can be applied in preferential sampling and spatiotemporal settings. The model assumes that underlying all observations, there is a spatially continuous variable that can be modelled using a Gaussian random field process. Fast inference is performed via the integrated nested Laplace approximation (INLA) and the stochastic partial differential equation (SPDE) approaches. In order to allow spatial data fusion, a new SPDE projection matrix for mapping the Gaussian Markov random field from the observations to the triangulation nodes is proposed. The performance of the new approach is shown by means of simulation and air pollution applications. The approach presented provides a useful tool in a wide range of situations where information at different spatial scales needs to be combined.
