Abstract.
Methodological approaches for modelling species distribution patterns have been lately proposed. Our approach makes use of hierarchical Bayesian models along with geographical and environmental characteristics at each sampling location. Maps of predicted probabilities of presence (or the abundance of the species) can then be generated using Bayesian kriging. Our interest here is to describe how to use the integrated nested Laplace approximation jointly with the Stochastic Partial Differential Equation approach to perform fast inference and prediction in such complex models. Moreover, our intention is to show that many extensions can also be easily implemented within this framework. In particular we will present practical examples involving preferential sampling, misalignment issues, spatio-temporal structures, etc.