Beyond kriging: dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling
Published:January 01, 2004
John Stephenson, K. Gallagher, C. C. Holmes, 2004. "Beyond kriging: dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling", Geological Prior Information: Informing Science and Engineering, Andrew Curtis, Rachel Wood
Download citation file:
The technique of kriging is widely known to be limited by its assumption of stationarity, and performs poorly when the data involve localized effects such as discontinuities or nonlinear trends. A Bayesian partition model (BPM) is compared with results from ordinary kriging for various synthetic discontinuous 1-D functions, as well as for 1986 precipitation data from Switzerland. This latter dataset has been analysed during a comparison of spatial interpolation techniques, and has been interpreted as a stationary distribution and one thus suited to kriging. The results demonstrate that the BPM outperformed kriging in all of the datasets compared (when tested for prediction accuracy at a number of validation points), with improvements by a factor of up to 6 for the synthetic functions.
Figures & Tables
Geological Prior Information: Informing Science and Engineering
Geological prior information represents a new and emerging field within the geosciences. Prior information is the term used to describe previously existing knowledge that can be brought to bear on a new problem. This volume describes a range of methods that can be used to find solutions to practical and theoretical problems using geological prior information, and the nature of geological information that can be so employed. As such, this volume defines how geology can be influential far beyond the confines of its own definition.