Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling
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 dis...
Main Authors: | , , |
---|---|
Format: | Journal article |
Language: | English |
Published: |
2004
|
_version_ | 1797058363579695104 |
---|---|
author | Stephenson, J Gallagher, K Holmes, C |
author_facet | Stephenson, J Gallagher, K Holmes, C |
author_sort | Stephenson, J |
collection | OXFORD |
description | 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. © The Geological Society of London. |
first_indexed | 2024-03-06T19:49:27Z |
format | Journal article |
id | oxford-uuid:2375c84c-579c-4441-b3e5-32e9e50d291e |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-06T19:49:27Z |
publishDate | 2004 |
record_format | dspace |
spelling | oxford-uuid:2375c84c-579c-4441-b3e5-32e9e50d291e2022-03-26T11:44:27ZBeyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modellingJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:2375c84c-579c-4441-b3e5-32e9e50d291eEnglishSymplectic Elements at Oxford2004Stephenson, JGallagher, KHolmes, CThe 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. © The Geological Society of London. |
spellingShingle | Stephenson, J Gallagher, K Holmes, C Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title | Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title_full | Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title_fullStr | Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title_full_unstemmed | Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title_short | Beyond kriging: Dealing with discontinuous spatial data fields using adaptive prior information and Bayesian partition modelling |
title_sort | beyond kriging dealing with discontinuous spatial data fields using adaptive prior information and bayesian partition modelling |
work_keys_str_mv | AT stephensonj beyondkrigingdealingwithdiscontinuousspatialdatafieldsusingadaptivepriorinformationandbayesianpartitionmodelling AT gallagherk beyondkrigingdealingwithdiscontinuousspatialdatafieldsusingadaptivepriorinformationandbayesianpartitionmodelling AT holmesc beyondkrigingdealingwithdiscontinuousspatialdatafieldsusingadaptivepriorinformationandbayesianpartitionmodelling |