Exploring Spatio-Temporal Variability By Eigen-Decomposition Techniques

The analysis of the relationships within and among environmental data observed at different spatial locations and temporal units can be done in many different ways. In this paper we propose a unifying overview of a set of multivariate data analysis techniques which are very powerful and useful for signal detection. In the context of spatially continuous processes, all the techniques are presented in the framework of Generalized Eigenvalue Decomposition (GED). The methodology is useful for exploratory spatial analysis but we also show that it can be used for prediction purposes.